Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216 DOI 10.1007/s12529-014-9418-2

ABSTRACTS

Abstracts from the ICBM 2014 Meeting

Workshop #1

Workshop #2

DESIGN AND CONDUCT OF RANDOMIZED BEHAVIORAL CLINICAL TRIALS – SESSION 1 Peter G. Kaufmann, Ph.D.1, Lynda H. Powell, Ph.D.2, Kenneth E. Freedland, Ph.D.3 1 National Heart, Lung and Blood Institute, Bethesda, Maryland 2 Rush University, Chicago, Illinois 3 Washington University, St. Louis, Missouri

USING N-OF-1 DESIGNS IN BEHAVIOURAL MEDICINE Prof Derek Johnston and Prof Marie Johnston Aberdeen University, Scotland

Introduction: Randomized controlled trials are the standard objective method for evaluating efficacy and effectiveness of interventions. Their implementation in behavioral medicine poses specific challenges that warrant careful consideration. Speakers will draw on their substantial experience in conducting, reviewing, and monitoring clinical trials involving behavioral interventions to provide solutions for commonly encountered issues. Objectives: this workshop is designed to provide a basic overview of the most salient principles that underlie the conduct of valid randomized behavioral clinical trials. Material will include a discussion of equipoise and blinding, efficacy and effectiveness, pragmatic trials, large simple trials, multisite trials, relationship of pilot studies to outcomes in the main trial, a consideration of Phases of trials, applicability of “Good Clinical Practices” as promulgated by regulatory agencies, issues in analysis including intent-to-treat, responder, and per-protocol analyses, primary and secondary endpoints, ancillary studies, and the structure and function of data and safety monitoring committees. Recognizing the diversity of issues in clinical trials, faculty will endeavor to integrate answers to specific questions or requests submitted to [email protected] not later than two weeks before the workshop. Altogether, this session will present the principal challenges associated with the design of clinical trials involving behavioral interventions, encourage questions from participants and encourage questions regarding any of the presented material. Workshop activities: The workshop will consist primarily of lectures, drawing on the contemporary clinical trials literature in behavioral medicine and related fields. Power point slides and selected readings will be distributed. We encourage questions submitted in advance of the workshop. The faculty will be available for consultation after the workshop. Description of the intended participants: The session assumes no particular level of familiarity with clinical trials involving behavioral interventions. It should be of interest to everyone with an interest in improving the quality of clinical research. Presenters: All three presenters have been involved in numerous single- and multi-site behavioral clinical trials involving individual or group-randomized designs, and have served on research grant review committees, data and safety monitoring boards, and taught clinical trials methods for more than a decade.

Introduction; N-of-1 studies (sometimes called single case studies) have been used for some time in clinical psychology, are increasingly used in medicine and have been advocated by the Medical Research Council in the UK as an important initial step developing and evaluating complex interventions. They are a critical aspect of testing that psychological theory applies to individuals. However n-of-1 studies are not common in behavioural medicine or health psychology perhaps because few researchers in these fields have been trained in the design, operationalisation and analysis n-of-1 studies. This workshop aims to modestly increase interested researchers’ and clinicians’ knowledge about n-of-1 and greatly increase their confidence that they can and should conduct n-of-1 research. Objectives; the objectives of this workshop are to consider: The importance of conducting n-of-1 studies in (a) the development and testing of theory, (b) the initial stages of treatment development and evaluation and (c) in clinical practice. The problems of establishing causality in n-of-1 studies and describe the main nof-1 designs that attempt to deal with this.The types of measures that can be used in n-of-1 studies including ecological momentary assessment/ experience sampling methods on PDAs, smart phones and similar devices. The main data analytic problems when analysing the time series data typical of n-of-1 studies in behavioural medicine. Current methods for understanding and analysing typical small sample n-of-1 data. Workshop activities; The workshop will include lectures, group discussion and demonstration of measurement and analytic procedures. Participants will be encouraged to describe their experience of n-of-1 methods and/or their future plans for n-of-1 studies. Description of the intended participants; The workshop is open to all with an interest in conducting n-of-1 studies. No previous experience of n-of-1 methods is required but some experience would be helpful. People with great expertise in n-of-1 methods might not gain much from the workshop, but we would surely gain from their attendance. Brief information on the presenters/facilitators Derek Johnston researches occupational stress, behavioural aspects of cardiovascular disease and physical activity. He has a particular interest in real time measurement of behaviour and physiological functioning and n-of-1 methodology. Marie Johnston conducts research on disability (theory, measurement and intervention) and on behaviour change in health and healthcare contexts using all appropriate methods including n-of-1 studies. Both are at

S2

Workshop #3 SELF REGULATION INTERVENTIONS AND A PROBLEM SOLVING IMPLEMENTATION IN HEALTH CARE PRACTICE Hennie Koelewijn & Nynke Kuipers PCC Health Promotion, The Netherlands Introduction.Self regulation can be defined as a circular process of human behavior, in which individuals give direction to their thoughts and feelings by means of self-reflection, in order to attain self set behavioral goals (Maes & Gebhardt, 2000; Maes & Karoly, 2005). The problem solving approach, as implementation framework of self regulation interventions, consists of four practical steps: 1) goal setting, 2) creating a plan of action, 3) feedback, process evaluation and control procedures and 4) goal evaluation and reformulation of (realistic) goals (Koelewijn, 2013). Research on self regulation interventions shows promising results in improving well-being of patients with chronic illnesses (Huisman et al., 2009; Janssen, 2012). Furthermore, pilot studies in health care practice settings also show that a problem solving approach with self regulation interventions is effective in improving wellbeing (depression, anxiety and burnout) for individual clients that participate in short-term care programs (Koelewijn, Kuipers & Van Mourik, 2012). Based on these research findings, we advice that health care providers receive more training to implement self regulation interventions. Objectives This interactive workshop will allow a venue for participants to exchange experiences applying self regulation interventions and a problem solving approach in short term care programs. In this workshop the following topics will be covered: Firstly, we will present an overview of the literature on interventions in experimental and clinical settings. Secondly, we will elaborate on thirteen intervention principles and the four implementation steps. Followed by a discussion on the opportunities and challenges in the application of self regulation and problem solving in practice. Workshop activities; the workshop will include an oral presentation on self regulation interventions and a problem solving implementation. Additionally, interactive exercises, case studies and discussions are facilitated.Description of the intended participants The workshop is at introductory level and is intended for participants to improve their implementation skills of self regulation interventions. Brief information on the presenters/facilitators: Last May Hennie Koelewijn received her doctorate after fifteen years of practicing and researching the problem solving implementation approach in health care settings. Furthermore, as director of four psychology practices she is an innovator in training and educating young health psychologists and other healthcare workers. Nynke Kuipers is currently Head Academy of PCC Health Promotion. She educates young health psychologists in learning to implement (selfregulation) intervention programs. She also teaches all kind of health care workers multidisciplinary and communication skills. Furthermore, she is a registered organizational and occupational psychologist and coaches individual clients, managers and teams in improving well-being, quality of work and personal development.

Workshop #4 APPLICATIONS OF LATENT VARIABLE MODELS IN BEHAVIORAL MEDICINE RESEARCH Maria M. Llabre, Ph.D., William Arguelles, PhD Department of Psychology and the Behavioral Medicine Research Center at the University of Miami, USA Latent variable models (LVM) are broad and flexible statistical models that specify the relations among a set of observed or manifest variables and their underlying unobservable constructs, the latent variables. Latent variables can be categorical or continuous, as is also the case with their resulting observed variables. Depending on the type of variable combination, these models are recognized by different names. For example, confirmatory factor analysis (CFA) is a LVM with continuous observed

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

and latent variables, whereas item response theory (IRT) models work with categorical indicators but continuous latent variables. Mixture models are those with categorical latent variables. Latent growth models (LGM) examine change across repeated measures by specifying parameters of change as latent variables. These models are useful in behavioral medicine research because they can accommodate measurement error, can reduce the number of observed variables, can facilitate the handling of some types of missing data, and can facilitate testing theoretical predictions. However, their application needs to consider the assumptions made. This workshop will introduce the basic concepts of LVM and illustrate the potential these models bring to different questions in behavioral medicine research. The concepts will be illustrated with several examples including both a CFA and a mixture model of metabolic syndrome, and applications of LGM to cardiovascular reactivity. Activities will include specification of the models and interpretation of results. This workshop is at an intermediate level. Participants should be familiar with linear models (regression analysis) and have a general notion of factor analysis. Maria M. Llabre is Professor of Psychology and teaches graduate statistics courses. William Arguelles is a Postdoctoral Fellow in Cardiovascular Behavioral Medicine.

Workshop #5 DESIGNING AND PROMOTING INTERVENTIONS IN PSYCHOSOCIAL CANCER CARE Barbara L. Andersen1, Robert T. Croyle2, Joost Dekker3, Michael A. Diefenbach4, Judith B. Prins5, Annette L. Stanton6 1 Professor of Psychology; Director, Livestrong Survivorship Center of Excellence; Ohio State University, USA 2 Director of the Division of Cancer Control and Population Sciences, National Cancer Institute, Bethesda, USA 3 Professor Allied Health Care, Department of Psychiatry and Department of Rehabilitation Medicine, VU University Medical Center, Amsterdam, the Netherlands 4 Director of Research, Department of Urology, Icahn School of Medicine at Mount Sinai, New York, USA 5 Chair of the Department Medical Psychology, Radboud 190 UniversityMedical Centre,Nijmegen, theNetherlands 6 Professor, Psychology & Psychiatry/Biobehavioral Sciences; Member, 192 Jonsson Comprehensive Cancer Center; Senior Research Scientist, Cousins 193 Center for Psychoneuroimmunology; UCLA, USA. Introduction ; this interactive workshop will address important challenges in designing and promoting effective interventions in psychosocial cancer care. World-leading experts in the field of psychosocial cancer care will present the state of the science and discuss the future development of the field. We welcome active participation of the audience, in a collaborative effort to improve psychosocial cancer care. The workshop is a collaborative activity of APA Division 38 (Health Psychology), Society of Behavioral Medicine (Cancer SIG), and International Society of Behavioral Medicine. Objectives; to present the state of the science of designing and promoting effective interventions in psychosocial cancer care. To involve the audience in a discussion, in a collaborative effort to improve psychosocial cancer care. Workshop activities; the state of the science is introduced in the following lectures: Designing, testing and implementing BREATH: a web-based self-management intervention for breast cancer patients, Judith B. Prins. Psychosocial cancer interventions on the internet: Are our patients ready?, Michael A. Diefenbach. Building Tests of Mechanisms into PsychoOncology Intervention Trials, Annette L. Stanton. Disseminating empirically supported psychological treatments for cancer patients, Barbara L. Andersen. Psychosocial Aspects of Quality Cancer Care, Robert T. Croyle. In the discussion following each lecture, the audience is invited to contribute their opinion and experience. Description of the intended participants; Health professionals, clinicians, researchers and policy makers involved in psychosocial cancer care.

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

S3

Wednesday 20 August 2014 ICBM Satellite Forum 09.00 – 16.00

&

FROM RESEARCH TO REALITY: LESSONS FROM DISSEMINATION AND IMPLEMENTATION RESEARCH ACROSS THE GLOBE Proposed by: Brian Oldenburg, Carina Chan (Chair, ISBM Early Career Network - INSPIRE) and K. Viswanath (Chair, Organizational Liaison)

&

Historically, researchers and policy makers have focused most of their attention on developing and testing empirically supported interventions. Much less attention has been paid to what is needed to implement and sustain these in a range of real-world settings. Many empirically supported interventions are less effective and/or fail to reach their target audiences when implemented within existing service structures and systems, and few are sustained over the long term. Clearly, if effective public health programs are not widely adopted and implemented, their potential to improve people’s health is significantly impaired. This challenge is exacerbated by the different values and perspectives that exist between practitioners, program implementers, policy makers and researchers. Practitioners often find evidencebased interventions difficult to conduct in community settings especially when there is such limited information about how to adapt programs to the local context. Furthermore, public health decision makers and program implementers are often reluctant to consider ‘new’ interventions when effectiveness has not been demonstrated in their particular setting or country. By contrast, researchers are usually more concerned with the internal validity of their programs rather than generalizability and external validity. “Dissemination” refers to the targeted distribution of information and intervention materials to a specific public health or clinical practice audience with the intent being to spread knowledge and the associated evidence-based interventions. “Implementation” refers to the use of strategies to adopt and integrate evidence-based health interventions and change practice patterns within specific settings. “Translation” refers to applying or adapting research findings or evidence to different community or population settings. This also includes ‘translation’ to other countries or cultures from where the original program was developed and evaluated. While there is a growing recognition of the importance of Dissemination and Implementation Research and a fast developing body of work, significant gaps remain. These gaps include a lack of case studies and evidence-base from low and middle income countries, continuing and vexing methodological challenges especially with study designs and measures, and more critical a lack of sustained dialogue between researchers and end-users on mechanisms of translation. The 2nd Satellite Forum of ICBM 2014 aims to follow on the conversations began at ICBM 2012 to address these gaps by bringing together key stakeholders from academy, policy and practice communities to continue the dialogue. This forum aims to draw on the knowledge and experience of speakers and participants from around the world:

& & & &

To learn about experiences from practice, policy and research communities on studying and implementing D&I research across the globe. To assess the state of the science in regard to theoretical models, research designs and measures on D&I research. To identify ways of working together in the future to further advance the evidence base for better implementation and translation of behavioural medicine interventions globally. To showcase exemplar D& I research studies from across the globe.

Topics to be covered in the 1-day forum will include:

& &

What are some of studies that best exemplify challenges and opportunities of D& I research across the globe? What are some of the needs of and challenges faced by policy makers to using research to promote changes in policies related to behavioural medicine?

&

&

What are the barriers to and preferences of practice communities in adopting, customizing and localizing effective interventions to their practice settings? What are some of the significant theoretical models and approaches being used? What is the state of evidence to support the models? What types of methods have been used for dissemination, implementation? What has been the experience in using these methods? What capacity, infrastructure and funding models are needed for more ‘scaled up’ dissemination and implementation program delivery and research?

These topics will be illustrated by case studies from different countries and settings. The speakers/discussants for the 1-day program will include ICBM attendees from both developed and developing countries who have expertise in research, policy and/or practice related to the dissemination and/ or implementation of behavioural medicine programs. A number of these individuals have already indicated their interest in being involved.

Workshop #6 DESIGN AND CONDUCT OF RANDOMIZED BEHAVIORAL CLINICAL TRIALS – SESSION 2 Peter G. Kaufmann, Ph.D.1, Lynda H. Powell, Ph.D.2, Kenneth E. Freedland, Ph.D.3 1 National Heart, Lung and Blood Institute, Bethesda, Maryland 2 Rush University, Chicago, Illinois 3 Washington University, St. Louis, Missouri Introduction: Randomized controlled trials are the standard objective method for evaluating efficacy and effectiveness of interventions. Their implementation in behavioral medicine poses specific challenges that warrant careful consideration. Speakers will draw on their substantial experience in conducting, reviewing, and monitoring clinical trials involving behavioral interventions to provide solutions for commonly encountered issues. Objectives: This workshop will address additional methodological issues associated with the design and execution of clinical trials involving behavioral interventions. These include: consideration of interventions dosage and its relationship to effect size; selection of appropriate control group options consistent with the objectives of the clinical trial in the context of existing literature; effect of inclusion and exclusion criteria on detecting a valid outcome; and, selection of appropriate primary and secondary outcome measures when the interventions can affect multiple targets or influence the expression of several symptoms. The presentations will address potential confounding of results by methods used to foster retention or with follow-up assessments themselves. Finally, the presentation will discuss effective publication of clinical trials. Workshop activities: The workshop will consist primarily of lectures, drawing on the contemporary clinical trials literature in behavioral medicine and related fields. Power point slides and selected readings will be distributed. We encourage questions submitted in advance of the workshop to [email protected] The faculty will be available for consultation after the workshop. Description of the intended participants: The session assumes no particular level of familiarity with clinical trials involving behavioral interventions as it should be of interest to individuals with a wide range of experience in clinical research. The selected topics build on those presented in Session 1. Maximum number of participants: There is no limitation on the number of participants. Presenters: All three presenters have been involved in numerous singleand multi-site behavioral clinical trials involving individual or grouprandomized designs, and have served on research grant review committees, data and safety monitoring boards, and taught clinical trials methods for more than a decade.

S4

Workshop #7 SPECIFYING THE CONTENT OF COMPLEX INTERVENTIONS TO IMPROVE HEALTH: USING A TAXONOMY OF BEHAVIOUR CHANGE TECHNIQUES (BCTS). Dr Wendy Hardeman (University of Cambridge) and Prof Marie Johnston (University of Aberdeen). Introduction; To date, there has been no shared language for describing the content, especially the ‘active ingredients’ of behaviour change interventions; by contrast, biomedical interventions are precisely specified. This limits the possibility of replicating effective interventions, synthesising evidence, and understanding the causal mechanisms underlying behaviour change. With the guidance of a multidisciplinary International Advisory Board of leaders in the field of behaviour change, we have developed a reliable method of specifying behaviour change techniques (BCTs), with consensually agreed labels and definitions (see www.ucl.ac.uk/health-psychology/BCTtaxonomy). Objectives; To familiarise participants with the need for the BCT Taxonomy, its method of development and its application. To give hands-on experience of using the BCT Taxonomy V1 to identify the active content of behaviour change interventions and write intervention descriptions. To encourage discussion between participants about their experiences of using BCT labels and definitions and future uses of the taxonomy. Workshop activities; Training tasks will teach participants a structured method for identifying the active content of behaviour change interventions and writing clear, replicable descriptions of intervention content using BCT labels and definitions in the taxonomy. The workshop will combine brief presentations, hands-on practice (individually and in groups) and structured discussion. Participants completing the workshop will be invited to join the Network of collaborators which will facilitate contact with others using and continuing to develop this resource. Description of the intended participants; The workshop is open to all participants with an interest in investigating, reviewing, designing or delivering behavioural interventions. No previous knowledge or experience of BCT taxonomies is required. Brief information on the presenters/facilitators; Wendy Hardeman and Marie Johnston are core members of the BCT Taxonomy V1 team. Wendy has developed and evaluated a range of behavioural interventions, with specification of interventions, fidelity, and process evaluation as key interests. Marie has also conducted RCTs of behavioural interventions with a focus on theory, measurement of outcomes and improving specification of the active ingredients. Acknowledgement: This workshop has arisen from initial findings of a UK MRC-funded study “Strengthening evaluation and implementation by specifying components of behaviour change interventions”, aka the BCT Taxonomy Project (Grant holders: Susan Michie, Charles Abraham, Martin Eccles, Jill Francis, Wendy Hardeman and Marie Johnston).

Workshop #8 TALK AND TEXT: QUALITYAND INNOVATION IN QUALITATIVE RESEARCH FOR BEHAVIOURAL MEDICINE Sally Sargeant Assistant Professor in Behavioural Sciences in the School of Medicine at Bond University, Australia Introduction; Qualitative research continues to enjoy increasing interest and use within a range of disciplines. In line with this increase, there are more demands to ensure that the quality of qualitative research is high. This means ensuring that such research is theoretically and conceptually supported; that analysis is thorough; and that it is well documented and disseminated appropriately. This workshop will focus on strengthening skills in qualitative research for behavioural medicine, for beginners and those who have some experience. In particular the focus will be on assessing the quality of

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

qualitative research and how to design effective studies that are matched to an appropriate research question. Attention is also paid to the process of analysis and how best to articulate this in written reports. Objectives

& & & &

To establish when it is appropriate to use qualitative methodology To understand data collection methods and analytic techniques and relevant distinctions between them To describe and critique writing qualitative reports To explain how to maximise quality in qualitative research

Workshop activities

& & & &

Designing and practising interview questions Practising qualitative analysis Documenting methods Personal reflections

Description of the intended participants; The workshop is suitable for participants who are new to qualitative research and those who wish to enhance their existing skills or learn about a new analytic technique. Brief information on the presenters/facilitators; Sally Sargeant is an Assistant Professor in Behavioural Sciences in the School of Medicine at Bond University, Australia. She has advised on many qualitative projects which have led to successful grant applications. She is an editorial board member of Qualitative Health Research and is committed to promoting good practice in qualitative data analysis.

Workshop #9 HOW TO IMPROVE EHEALTH INTERVENTIONS IN HEALTH PSYCHOLOGY AND BEHAVIORAL MEDICINE? Dr J van Gemert-Pijnen, Dr S Kelders Introduction: eHealth is gaining more and more ground in health psychology and behavioural medicine to support wellbeing, a healthier lifestyle or adherence to medications. Despite the large number of eHealth projects to date, the actual use of eHealth interventions is lower than expected. Many projects in Health fail to survive beyond the pilot phase and studies that investigate the effectiveness of eHealth applications most often do not show long-term effects. The ongoing concern in academic literature about these shortcomings situates eHealth as an immature research domain not capable to follow the golden standards of high quality research in health psychology and behavioral medicine. On the other side, in lifestyle or mental health interventions positive results with internet-based interventions can be observed and patients favor internet interventions. Therefore, we will discuss how to make eHealth interventions more enjoyable, feasible, and cost-effective. Objectives: In this workshop we demonstrate how eHealth interventions can be improved, using technology as a driver for adherence and behavior change. In our workshop we introduce a new approach to guide the development of eHealth interventions. The approach is based on an integration of theories form behavioural sciences and persuasive technology. The purpose of the workshop is first, to provide new opportunities to develop eHealth interventions that are user centred and that are implementable in practice, second to discuss an agenda for future research in eHealth for health psychology and behavioral medicine-interventions. Learning outcomes are insight in how to use technology to develop eHealth interventions that work, help and are cost effective, and knowledge about how to evaluate the effects of eHealth interventions on adherence and outcomes. Workshop activities: The workshop starts with a demonstration of examples of eHealth interventions to support self-care and safe-care. Interventions we focus on are blended therapy for depression, self-management applications to support patients with chronic diseases, and applications to

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

support medical professionals to increase safety at work. The examples will be discussed alongside the new approach for development and implementation to improve eHealth interventions. Persuasive Technology will be highlighted as a strategy to increase the adherence to eHealth interventions. The effects of persuasive technology on adherence and its’ consequences for design will be discussed with participants. Participants will evaluate the persuasiveness and user friendliness of eHealth interventions using the eHealthwiki.org (toolkit) and a persuasive model for system and content design. Finally, the workshop synthesizes views on innovative research methods to improve the uptake and impact of eHealth interventions. For the workshop we use our book, Improving eHealth(www.elevenpub.com). Description of the intended participants: participants with interest in design, implementation, evaluation of eHealth interventions. Brief information on the presenters/facilitators: Lisette van Gemert-Pijnen is associate professor at the University of Twente, at the University Medical Center Groningen, and adjunct professor at the University of Waterloo in Canada. Lisette founded and is the director of the first Center for eHealth Research and Disease Management (www.ehealthresearchcenter.nl). Her research group’s program focusses on persuasive designs (people-driven) and business modeling (valuedriven) for interactive eHealth interventions, in particular on how persuasive designs can increase trust, engagement, and adherence to eHealth interventions aimed at safe care and self- care. Lisette is the chief editor of the book Improving eHealth, which provides mixed methods for developing and implementing eHealth interventions aimed at behavior change. Saskia Kelders is assistant professor at the University of Twente. In 2012 she defended her PhD-thesis ‘Understanding adherence to web-based interventions’. Currently she teaches eHealth and eMental Health to Master students in psychology. She is involved in research projects about positive technology to change attitudes and behaviors in mental health. Her focus is on the development and evaluation of these technologies, with a special interest in persuasive technology, adherence and log-data.

Workshop #10 BEHAVIORAL MEDICINE IN CHINA: CHALLENGES AND OPPORTUNITIES Organizers: Bo Bai (Chair), Joost Dekker (Co-chair), Chengxuan Qiu (Coordinator), Feng Ji (Coordinator) Presenters: Qingshan Geng, Joseph Tak-Fai Lau, Peiyuan Lu. Discussants: Qihua Tan, Chengxuan Qiu Introduction/background; In the last three decades, along with rapid economic growth and improvement in living standards in China, there have been considerable changes in diets (e.g., meat consumption), lifestyle behaviors (e.g., smoking, alcohol consumption, sedentary lifestyle, and sexual behavior), and social environment, which significantly contribute to the epidemic of chronic non-communicable diseases (e.g., cancer, ischemic heart disease, and stroke) and sexually transmitted diseases (e.g., HIV) as well. In addition, health problems of rural-urban migrant workers have become increasingly important due to rapid urbanization. Research in the field of behavioral medicine will help to deal with the tremendous threats owing to these major public health issues and thus, contribute to the sustainable development of economy and society in China. Objectives; To present and discuss research on several relevant topics related to behavioral medicine in China. To promote collaboration and exchange on research in behavioral medicine between China and other nations. Workshop activities; Welcome speech (5 min): Prof Bo Bai, President of Chinese Society of Behavioral Medicine (CSBM), Chinese Medical Association, China. Scientific session (20 min/talk plus 10 min break): Prof Qingshan Geng, "Sexually transmitted diseases and education-based intervention in China" Prof Joseph Tak-Fai Lau, "Health of migrant workers from rural to urban areas in China"

S5

Prof Peiyuan Lu, "Stroke and vascular cognitive impairment in China" Dr Qihua Tan, “Metabolic factors, lifestyle behaviors, and health”. Dr Chengxuan Qiu, “Prevention of vascular cognitive impairment: an international perspective”. Concluding remarks (5 min): Prof Joost Dekker, President of International Society of Behavioral Medicine (ISBM), The Netherlands. Description of the intended participants; The intended participants of this workshop include (1) researchers from China and other Asian-Pacific countries, and (2) people who have traditional collaboration with China or who are interested in developing new collaboration with researchers in China. Brief information on the presenters/facilitators Prof Bo Bai, President of CSBM, and President of Jining Medical University, Shandong, China. Prof Joost Dekker, President of ISBM, and Professor of VU University Medical Centre, Amsterdam, The Netherlands. Prof Qingshan Geng, Deputy Head of Health Department of Guangdong Province, China. Prof Joseph Tak-Fai Lau, President of the Hong Kong Society of Behavioral Health and Director of the Centre of Health Behaviors Research, School of Public Health and Primary Care, Chinese University of Hong Kong, China. Prof Lau serves as a HIV consultant of several Provincial Centres for Disease Control and Prevention in China. Prof Peiyuan Lu, Dean of Department of Neurology, and Vice President of Hebei General Hospital, Hebei, China. Dr Qihua Tan, Associate Professor at Institute of Public Health, University of Southern Denmark, Denmark. Dr Tan’s research focuses on epigenetic epidemiology of cancer and aging phenotypes as well as genetic and environmental factors of metabolic disorders in eastern and western populations. Dr Chengxuan Qiu, Associate Professor at the Aging Research Centre, Karolinska Institutet, Sweden and Visiting Professor of Jining Medical University, China. Dr Qiu’s research focuses on epidemiology and prevention of dementia and vascular mechanisms of brain aging, cognitive, and physical dysfunction.

Workshop #11 USING SOCIAL MEDIA TO WRITE AND PROMOTE YOUR SCIENTIFIC PAPERS James C Coyne PhD Introduction; Social media provide a powerful set of tools for organizing research teams, conducting research, crowd sourcing expertise, and obtaining preliminary peer review of your scientific papers. Social media also provide powerful ways to publicize and disseminate newly published papers to a broad audience. Writing groups organized on the Internet increasingly publish papers in JAMA, PNAS and BMJ, as well as open access journals. Individual authors often develop their ideas in tweets, short and long read blog posts, and then solicit feedback on drafts. Authors and universities increasingly engage in social media campaigns to publicize and discuss recently published papers. High impact journals undertake their own such efforts. Powerful tools such as PubMed Commons provide increasingly respected post publication peer review. Altmetrics now quantify such activities. This workshop provides exposure and hands-on experience in use of these tools at all stages of writing and promoting scientific work. Objectives

1. Familiarize participants with tools such as Twitter, blogs, Rebel 2. 3.

Mouse, and Facebook to develop research and promote it after publication. Provide concrete case examples of successful application of these tools to both the knowledge production and dissemination. Engage participants in utilizing these tools with the projects they bring to the workshop.

Workshop activities; Didactic presentation of tools and concrete examples of their application will be followed by a highly interactive session in

S6

which participants will be invited to develop their ideas for organizing paper writing and subsequent promotion using social media. Description of the intended participants; The workshop will be of greatest interest to junior and senior researchers who are actively writing papers. Participants will be encouraged to bring ideas and works in progress. If they have forthcoming papers, they may use the workshop to construct social media strategies for promoting and disseminating their work. Brief information on the presenters/facilitators; The presenter has written hundreds of articles. His blog posts at PLOS Mind the Brain and Science Based Medicine each receive thousands of hits. A few years ago he turned to writing collectives organized on the Internet that have produced numerous high impact and award-winning articles. He has yet to meet all authors on some of these papers. He can present a number of personal examples of the progression of tweets to published papers to post publication promotion.

Poster Session A P1 PLANNING, TIMING AND QUIT SUCCESS: A RANDOMISED CONTROLLED TRIAL J.B. Balmford, Ron Borland, Elena Swift Cancer Council Victoria, MELBOURNE, Australia Introduction: Recent research has led to debate over the benefit of delaying the implementation of a decision to quit smoking in order to plan the attempt, and the role of planning. However, timing and planning need not be linked: planning can occur before a commitment to quit is made, after it is implemented, and in between. Method: 2x2 factorial RCT. Smokers (n=1,602) were recruited from Australian QuitCoach users who were open to the possibility of quitting, but not committed to quit within 2 days. In one arm participants were encouraged to implement a quit attempt without delay (Timing arm) and in the other were encouraged to use a set of planning resources added to the basic QuitCoach (Planning arm). Process measures of intervention uptake and short term outcomes were collected at 1 month, and 6-month sustained abstinence at 7 months. Results: 1-month follow-up data was obtained from 49% (n=789), of whom 42.8% had quit. There were significant positive main effects for timing and planning (both OR's 1.49, 95% CI 1.00-2.24) but no evidence for an interaction (OR 0.71, 95% CI 0.41-1.27). In the Timing arm, only 35.6% agreed to quit immediately. In the Planning arm, 52.9% used at least one of the planning tools. Results for 6-month sustained abstinence are being collected and will be reported at the meeting. Conclusions: Preliminary findings indicate that encouraging immediate implementation of a quit attempt, and the availability of planning resources, are associated with modest increases in abstinence. CORRESPONDING AUTHOR: Dr JB Balmford, Cancer Council, Victoria Australia, [email protected]

P2 DRUG CONSUMPTION ATTITUDES AT PENINSULA MUSIC FESTIVAL Z Abram University of Medicine and Pharmacy Tirgu Mures, TIRGU MURES, Romania Introduction. The study of drug consumption and health attitude among the youth makes it possible to know the risk factors that show correlation between health, general condition and drug use. Method. During a festival called Peninsula Music Festival Tîrgu Mureş we examined smoking, alcohol consumption and drug usage habits of

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

young participants using a questionnaire method. The collected data were compared with data obtained during other surveys. Results. The changes of both legal and illegal drug use habits led us to conclude that the same happens in our country as in the western EU countries. The frequency of drug trials grows among the youth, the habits of drug use change, cannabis derivatives and ethnobotanical drugs dominate while the proportion of opiate users is decreasing. The statistics are alarming among music festival participants: 83% of them are smokers, 64% have been drunk more than 10 times and the majority have tried at least one drug, about 10% of them being frequent users. Conclusions. Our study emphasises the importance of drug-prevention strategies and concludes with the importance of making more effort in order to involve the younger generation in health education programmes. CORRESPONDING AUTHOR: Prof Z. Abram, University of Medicine and Pharmacy, Tirgu Mures Romania, [email protected]

P3 MOBILE PHONE ADDICTION: A SOCIO-COGNITIVE PERSPECTIVE ON EXCESSIVE MOBILE PHONE USE H. Allahverdipour, Sara Poorrazavi, Mohammad Asgharijafarabadi Tabriz university of Medical Sciences, TABRIZ, Iran Introduction: The study examined the predictive ability of selected demographic and socio-psychological characteristics to explain excessive mobile phone use (EMPU) behavior and problematic use, in a sample of university students, on the basis of the social cognitive theory. Methods: The sample consisted of 476 randomly selected university students in Tabriz, Iran. The study was cross-sectional in nature. A researcher-designed questionnaire was used for the purpose of data collection. No causal inferences were drawn due to non-experimental nature of the study. Results: It was found that having boyfriend/girlfriend increases the likelihood of EMPU while self-efficacy to avoid EMPU decreases it. Selfefficacy to avoid EMPU, self-regulation, observational learning, selfcontrol, and attitude toward EMPU were predictors of mobile phone problematic use. The results provided a proper fit for a conceptual model of reciprocal determinism. Conclusion: Although social cognitive constructs may predict mobile phone problematic use, they are not useful in predicting the behavior of EMPU. CORRESPONDING AUTHOR: Prof. H Allahverdipour, Tabriz university of Medical Sciences, Iran, [email protected]

P4 BRIEF INTERVENTION PRACTICES AMONGST NURSES WORKING WITH PREGNANT WOMEN OR PARENTS IN NORTH-NORWAY S.C. Wangberg1, 2 22 1 Narvik University College, NARVIK, Norway 2 Regional Centre of Substance Abuse, University Hospital of North Norway, NARVIK, Norway Introduction: Brief interventions are an effective way of reducing risk use of alcohol use amonst pregnant women and parents. A training program is currently being implemented in Norway for nurses working with pregnant women or parents of pre-school children. This study assessed the current practices and attitudes amongst these nurses in North-Norway. Methods: A survey was conducted amongst all 385 nurses working with pregnant women or parents in North-Norway. Of these, 105 community nurses and 35 midwives responded (36%). Results: Most of the midwives (91%) ask the pregnant women about their alcohol use at their first consultation. Twenty-seven per cent of the community nurses do the same. Fathers are asked less. While the nurses

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

agreed that screening tools are useful, they are not always using them and they stated a need for more training. When risk use of alcohol is detected, 49% of the midwives and 34% of the community nurses follow up with an intervention of their own. Referring to the parent’s GP was even more common. Motivational Interviewing was well known and frequently used. Time constraints and lack of cooperation with other professionals were mentioned as barriers for brief interventions in response to an open question. Conclusion: Response rate limits conclusions to be drawn, but it seems that the nurses find brief interventions for alcohol use to be an important part of their job, but could use more training in the screening tools, and more time for following up parents. CORRESPONDING AUTHOR: Dr. SC Wangberg Narvik, University College, Norway, [email protected]

P5 PERSONAL AND PERCEIVED PEER USE OF TOBACCO AMONG UNIVERSITY STUDENTS IN SEVEN EUROPEAN COUNTRIES C.R. Pischke1, SM Helmer1, G Van Hal2, B Vriesacker2, R Dempsey3, J McAlaney4, BM Bewick5, YAkvardar6, F Guillén-Grima7, O Orosova8, F Salonna9, O Kalina8, C Stock10, H Zeeb1, R Mikolajczyk11 1 Leibniz-Institute for Prevention Research and Epidemiology - BIPS, BREMEN, Germany 2 Faculty of Medicine and Health Sciences, University of Antwerp, Antwerp, Belgium, ANTWERP, Belgium 3 Staffordshire University, STAFFORDSHIRE, United Kingdom 4 Bradford University, BRADFORD, United Kingdom 5 University of Leeds, LEEDS, United Kingdom 6 Marmara University, ISTANBUL, Turkey 7 Public University of Navarra, NAVARRA, Spain 8 PJ Safaric University, KOSICE, Slovak Republic 9 Palacky Univesity Olomouc, OLOMOUC, Czech Republic 10 University of Southern Denmark, ESBJERG, Denmark 11 Helmholtz Centre for Infection Research, BRAUNSCHWEIG, Germany Background: Previous research suggests that students tend to overestimate peer tobacco use. However, except for one Dutch study, the majority of studies on social norms regarding tobacco consumption were conducted in North America. The Social Norms Intervention for the prevention of Polydrug usE (SNIPE) study aims to assess rates of tobacco use and perceptions of peer tobacco use among European students. Methods: The SNIPE study was EU-funded and conducted in 7 European countries (Belgium, Denmark, Germany, Slovak Republic, Spain, United Kingdom, Turkey). In total 4,439 students (71% female) answered questions about their personal tobacco use and their perceptions on tobacco consumption rates in the majority of their peers at baseline. Results: Overall, the majority of male (60.8%) and female (72.5%) students were non-tobacco users (never use and no use in the past 2 months) at baseline. Tobacco use was least common in Belgium, where only 18.2% of students had used tobacco two or more times a week in the last two months, compared to higher rates of tobacco consumption in Turkey (43.3%) and Germany (40.6%). Across all countries, students reported higher rates of perceived tobacco consumption for their peers than for themselves. For example, more than 50% of students in each country perceived that the majority of peers of their own gender (>51%) consumed tobacco at least twice or more often a week. Conclusions: The SNIPE-study represents the first cross-national European collaboration indicating significant variations in tobacco use by country. Consistent with previous research, European students overestimated peer tobacco use. CORRESPONDING AUTHOR: Dr. Pischke, Leibniz-Institute for Prev e n t i o n R e s e a r c h a n d E p i d e m i o l o g y - B I P S , G e r m a n y, [email protected]

S7

P6 OVER 12 YEARS FOLLOW UP OF QUIT SMOKING PROGRAM PARTICIPANTS Y.T. Takahashi1, Hideshi Miura2 1 Nara Women's University, NARA, Japan 2 Quit Smoking Marathon, AGEO, Japan Introduction: Although achieving lifelong tobacco abstinence is an important issue, few studies have evaluated the long term effectiveness of the internetmediated programs. In 1997, the first internet-mediated smoking cessation program was provided in Japan named Quit Smoking Marathon (QSM). The program was composed of two step programs: step1 program is to start quitting (1 month) and step2 program is to continue lifelong abstinence. This is the first report of over 12 years follow up of QSM participants. Methods: A questionnaire asking the present status of smoking, duration of abstinence and participation of the step2 program of QSM was sent by e-mail to 1127 members participated the QSM in 1997-2001 (over 12 years since participated). Biochemical methods for determination of smoking status, such as expired carbon monoxide, were not adopted. Results: 108 e-mails were not delivered because of the e-mail address failure. 118 members answered to the questionnaire. 108 members quit at the end of step1 program and 82 of 108 (76.0%) had continued absolute abstinance. Of 118 members, 58 still join the step2 program (55 quitters and 3 failures). 56 retired the program (45 quitters and 11 failures). Significant difference was found between the quitting rate in step2 program members and in those who had left (p<0.05). Conclusion: This study indicate that, among QSM participants, step2 program was suggested to be one of the the important factors to continued abstinance. CORRESPONDING AUTHOR: Y.T. Takahashi, Nara Women's University, Japan, [email protected]

P7 PERSONALTRAITS OF DRUG USERS IN TAIWAN CORRECTIONAL SERVICES L.H.Y. Yang, C.W. Wu Hsuan Chuang University, HSINCHU, Taiwan Purpose: Taiwan's Agency of Corrections data showed that drug users reached 10,947 at the end of 2013. This number accounted for 19% of the total inmates in Taiwan. This study was to explore the relationships among the personal traits and three types of illegal drugs (narcotic, psychotropic, and mixed). Method: This study was cross-sectional, and research design based on the five-factor model that revised by McCrae & Costa (2008). The questionnaire was developed for measuring the personal traits including five dimensions (OCEAN): openness, conscientiousness, extraversion, agreeableness, and neuroticism. Samples were collected from two prisons (Hsinchu and Tainan) from January in 2012 to July in 2013, and there were 1701 male drug users participated in this study. Results: The descriptive statistics indicated that the users inclined towards young (ages from 30 to 40), less educated and single. The main reason for drug use was to refresh. The users had higher scores on agreeableness and conscientiousness than on neuroticism, extraversion, and openness. The personal traits of agreeableness and neuroticism preferred to choose narcotic drugs significantly. There was no significant relationship between three personal traits (conscientiousness, extraversion, and openness) and three types of the drugs (narcotic, psychotropic, and mixed). Conclusions: Personal traits affected the choice of type of drug; however, the relationship between personal traits and drug types need further studies, especially among female users. CORRESPONDING AUTHOR: Prof LHY Yang Hsuan, Chuang University, Taiwan, [email protected]

S8

P8 SMOKING CESSATION BRIEF INTERVENTION IN INDONESIA: WHO REALLY WANT TO QUIT AND DOES IT WORK? Y.P. Prabandari1, R.S. Padmawati1, J.D. Rahaju2, T. Istiani2, E. Pujiastuti2 1 Faculty of Medicine, Universitas Gadjah Mada, YOGYAKARTA, Indonesia 2 Quit Tobacco Indonesia, Faculty of Medicine, Universitas Gadjah Mada, YOGYAKARTA, Indonesia Introduction: Quitting smoking for people living in the third most populous country in the world is challenging, since the limitations of smoking cessation services and the lack of tobacco control policy. Further, smoking is regarded as a means of hospitality and cultural friendship. The objective of this paper is to assess the implementation of a smoking cessation clinic at the hospital, lung clinic and special events in Yogyakarta. Method: A descriptive and qualitative method was used to analyze the smoking cessation brief intervention record of 248 clients who visited district hospitals, lung clinics and a smoking cessation stall at an exhibition in Yogyakarta. Subsequently, in-depth interviews were carried out among the smoking cessation service counselors. Results: Most clients were patients who visited the hospital or clinic, were male and ranged in age from 17 to 87 years old. The majority of clients in the district hospital visited smoking cessation services voluntary, as well as the clients who visited smoking cessation stalls at the exhibition. More than half patients in the lung clinic were referred by doctors and a quarter of clients have tried, but failed, to stop smoking without assistance. Older clients had more intention to quit than the young. The main reason to quit is health concern, and view of them were asked by the doctor to quit because they suffered from disease. Only limited clients returned for follow up, although a few of them sent texts to the counselors that they have quit. CORRESPONDING AUTHOR: Associate Prof YP Prabandari, Faculty of Medicine, Universitas Gadjah, Mada Indonesia, [email protected]

P9 DRUG ABUSE PATTERNS AMONG CLIENTS OF CENTERS FOR DRUG ABUSE TREATMENT IN TABRIZ, IRAN A. Shaghaghi, Reza Nikanfar, Akram Mirzaei Tabriz University of Medical Sciences, TABRIZ, Iran Introduction: Iran is confronted with growing illicit drug abuse especially among its young aged population due to being in the transit route of drugs trafficking from main producing countries. Preferences of the Iranian drug abusers to select certain drugs over others and their administration routes was envisaged in this cross-sectional study to expand research evidence. Method: Random sampling was applied to select 760 active files from a list of 3510 registered drug abusers in the 20 centers for drug abuse treatment in Tabriz the capital city of the East Azarbaidjan province, North West of Iran. Proportions and their 95% confidence intervals (CI) were applied to report the type(s) of abused drugs and their route of administration. Results: Speculation of the records revealed that 20.0% (95% CI: 17.323.0%) of the addicts were using crystal meth, 55.4% (95% CI: 51.858.9%) opium and 9.3% (95% CI: 7.5-11.6) heroine. Among the 43.4% (95% CI: 35.8-51.4) of crystal meth users multi drug use with at least one extra drug was also prevalent. The administration route was oral in 52.2% (95% CI: 48.7-55.8%) and injection in only 0.7% (95% CI: 0.3-1.5%) of the studied cases. Having a full time job was reported for 64.0% (95% CI: 60.5-67.3%) of the patients. Conclusions: Transition from abuse of conventional drugs to relatively new generation drugs was evident. Some addicts being in full time work means low productivity and represents an additional burden on the economy of the country in addition to the direct health care costs.

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

CORRESPONDING AUTHOR: Dr A Shaghaghi, Tabriz University of Medical Sciences, Iran, [email protected]

P10 UNDERSTANDING HARD TO MAINTAIN BEHAVIOUR CHANGE: INSIGHTS FROM SMOKING CESSATION R. Borland Cancer Council Victoria, MELBOURNE, Australia Many core challenges for behavioural medicine involve changing the lifestyles of people to be more conducive to their long term health. This includes behaviours like smoking, taking other drugs, excessive use of alcohol, inadequate exercise and inappropriate eating patterns. People typically make multiple failed attempts to change these behaviours, and some never succeed in doing what they know is desirable. Understanding these hard to maintain behaviours is critical to making progress in improving health. Unfortunately for those who believe it all can be done by changing social indicators, the reality, at least for smoking, is that it can’t, environmental factors play an important role but sometimes more is needed. This paper reconceptualises the problem through a dual process model of human behaviour and one that accepts that the determinants of wanting to do something and of succeeding are in some respects fundamentally different. This model helps explain why some behaviours are hard to maintain, and also identifies similarities and differences in the challenges of getting people to do more of desirable things, less of undesirable ones, and to eliminate some behaviours altogether. It provides an explanation of why some puzzling, seemingly paradoxical effects occur, such as in the west, the more a person wants to quit smoking the less likely they are to succeed on any given attempt, and the finding that even though quitting smoking relieves financial stress, those who are financially stressed are less likely to successfully quit, even though they are just as motivated, or more so. CORRESPONDING AUTHOR: Dr R Borland, Cancer Council, Victoria Australia, [email protected]

P11 ENHANCING ADHERENCE, ENGAGEMENT AND RETENTION IN HIV CARE IN ARGENTINA P.L. Jones1, L. Bofill1, D. Cecchini2, O. Sued3, S.M. Weiss1, D. WaldropValverde4, M.R. Lopez1 1 University of Miami Miller School of Medicine, MIAMI, United States of America 2 Helios Salud, BUENOS AIRES, Argentina 3 Fundacion Huesped, BUENOS AIRES, Argentina 4 Emory University, ATLANTA, United States of America Background: Treatment engagement, retention and adherence are required to optimize HIV treatment outcomes. This study examined the most challenging patients, those not engaged in care in public and private HIV healthcare in Buenos Aires. Methods: Patients (n = 60) prescribed antiretrovirals and disengaged from treatment were assessed on adherence, knowledge, motivation and attitudes regarding medications and randomized to intervention or control conditions. Results: Participants were similar between sites (age, x = 39.22 SD+ 7.92) and drug/alcohol use (33%); over 52% of public patients reported depression and 60% had discontinued medications (p<.005); private patients reported higher adherence (85%) than public patients (52%). Those not understanding their health status (35%) were less likely to know their CD4/VL status (X2=7.67, p = .005) and to have more questions (X2 = 14.81, p = .002). Overall, 52% were taking medications less than 80% of the time. Treatment motivation was similar between sites but most patients had low motivation. Patients not engaged immediately following diagnosis were less likely to take medications; those confused about their health (67%)

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

more often discontinued treatment (p = .03); self-reported adherence to medication was associated with VL (r = .35, p = .01). Attendance was higher in the intervention (p=.02), though 33% failed to attend any sessions. Healthcare barriers were addressed in both clinics. Conclusions: Opportunities exist for educational, motivational and mental health interventions. Preliminary results highlight challenges to patient engagement and disparities between public and private sites. Enhanced healthcare systems may improve engagement. Funded by NIMH grant R34MH097609 CORRESPONDING AUTHOR: Prof. PL Jones, University of Miami Miller School of Medicine, United States of America, [email protected]

P12 USING COMPUTER-TAILORING TO IMPROVE PRATICE NURSES' ADHERENCE TO SMOKING CESSATION GUIDELINES: A RESEARCH PROTOCOL D.F.L. de Ruijter, E.S. Smit, H. de Vries, C. Hoving Maastricht University, MAASTRICHT, Nederland Introduction: Smoking cessation advice from general practice staff, like practice nurses, effectively increases patient cessation rates. Yet, practice nurses’ adherence to evidence-based smoking cessation guidelines is often suboptimal. Computer-tailoring can successfully influence predictors of behavior, but this promising technology has so far not been applied to improving intermediaries’ guideline adherence. This project aims to develop and test a computer-tailoring intervention to improve practice nurses’ adherence to STIMEDIC, a recently disseminated but underused Dutch smoking cessation guideline. Method: A randomized controlled trial will be applied to test the effect of the web-based computer-tailoring intervention against no intervention among 272 practice nurses. The primary outcome measure is STIMEDIC guideline adherence, measured via self-report guideline checklists, via two objective score reports from randomly audiotaped consultations and via web-based questionnaires filled out by the patients. The secondary outcome measure is the patient’s point prevalence smoking abstinence (yes/no) and the number of quit attempts during the trial period. A trialbased economic evaluation subsequently examines the intervention’s cost-effectiveness and cost-utility. Potential for improvement is identified by performing a process evaluation. Results: Results are expected to demonstrate that the intervention is (cost-)effective in significantly improving STIMEDIC guideline adherence rates among practice nurses and increased smoking cessation rates among their patients. The results of the needs assessment interviews with practice nurses are expected halfway 2014 and will be presented. Conclusions: Computer-tailoring interventions for practice nurses to improve guideline adherence have not been developed yet. If successful, the intervention could be considered for wider implementation in the general practice. CORRESPONDING AUTHOR: DFL de Ruijter, Maastricht University, Nederland, [email protected]

P13 EXPERIENCES OF PARTICIPATING IN AN ANTIRETROVIRAL TREATMENT ADHERENCE CLUB S.A. Kagee, R. Dudhia Stellenbosch University, STELLENBOSCH, South Africa Introduction: In an effort to streamline the management of large numbers of patients receiving antiretroviral therapy (ART) in South Africa, adherence clubs were introduced over the last five years. We sought to document the experiences of patients attending adherence clubs and health care workers at clinics where clubs were operating.

S9

Methods: Participants were six ART adherence club members and seven health care workers, which included HIV nurses, medical doctors, pharmacists and counsellors. Data in the form of one-on-one interviews were collected at the Infectious Diseases Clinic of a large district hospital in a prei-urban area of South Africa. The interviews covered ART users’ experiences of the clubs, advantages and challenges that arose in the context of the club-based method of providing treatment, and the concerns facing ART users and health care workers (HCW’s) with regard to the clubs. Results: The data were analysed using thematic analysis. The themes that emerged from the data were: Rationale for the adherence clubs, Benefits of adherence clubs, Cohesion among patients, Different patient identity, Problems associated with adherence clubs, Confidentiality, Need for more clubs, Non-adherent club members, and Consequences of poor adherence among club members. Conclusions: There were clear benefits to the introduction of adherence clubs, most importantly the reduced amount of time ART users needed to spend at the clinic. Despite the benefits, various problems also emerged, the most important one being the logistical problems associated with the timely and correct delivery of drugs. CORRESPONDING AUTHOR: Prof. SA Kagee Stellenbosch University South Africa [email protected]

P14 THERAPEUTIC COMPLIANCE TO ANTIHYPERTENSIVE MEDICATION: A QUALITATIVE EXPLORATION FROM PATIENTS' PERSPECTIVE S.P. Pati1, S.P. Pati Sanghamitra2 1 Unit 9 ,Zonal Dispensary, BHUBANESHWAR, India 2 Indian Institute of Public Health, BHUBANESWAR, India Introduction- Globally hypertension is the leading cause of mortality and disability. Despite the availability of effective therapy, it is poorly controlled worldwide.1 In developing countries poor patient compliance and non adherence challenges the effective management of hypertension.2Hence it is critical to understand the reasons for poor compliance among hypertensives and identify the modifiable factors. Method- 30 hypertensives including 15 compliant and 15 non compliant or poorly compliant ( both self reported), attending primary health dispensary were randomly selected. In depth interviews were conducted with every participant to explore their knowledge, belief, care seeking behaviour, attitude and possible barriers and enabling factors in compliance towards therapy. Result- Interviewee were aware of hypertension and its complications. Family support emerged as a major motivator for adherence. Adequate counselling by the physicians regarding the importance of medications and perceived severity of complications if they discontinued contributed to compliance. Availability of free antihypertensive medications at government hospitals also enhanced compliance. Majority of non-compliants perceived they should not take medicine daily as they ``feel good”and don`t have any overt symptoms. Besides, fear of side effects of medicines ``taking medicine daily may harm me’ , some felt hypertension is a temporary disorder and self limiting. Use of multiple pills “polypharmacy”,poor counselling by the prescribing physician and costly medicines were demotivators. Conclusion-Family and society support, patient education on importance of continuing therapy in hypertension by physicians, cost effective medications are some of the modifiable factors that should be looked into to enhance treatment compliance. CORRESPONDING AUTHOR: Dr. SP Pati, Unit 9, Zonal Dispensary India, [email protected]

S10

P15 UNDERSTANDING LOW-INCOME PATIENTS' HEALTH EXPERIENCES TO IMPROVE ADHERENCE TO HYPERTENSION MANAGEMENT PLANS E.P. Polander1, V.L. Shalin2 1 Applied Research Associates, BEAVERCREEK, United States of America 2 Wright State University, FAIRBORN, United States of America Individuals’ personal health experiences and understanding of these experiences (including misconceptions) provide part of the foundation for their decisions regarding adherence to hypertension management plans. While individuals may not always explicitly communicate every aspect of their understanding, they can provide additional insight into their beliefs through recounting personal health experiences and decisions. The current study aimed to identify facets of patient understanding underlying decisions to adhere by soliciting their specific health experiences, interpretations of experiences, health decisions, and explicit hypertension knowledge. Individual, in-depth interviews were conducted with 15 low-income, hypertensive patients at a free health clinic in a U.S. metropolitan area. The interviews focused on soliciting personal health experiences and decisions regarding adherence to hypertension management plans and perceived implications of their decisions. Constant comparative analysis was used to identify important trends in patient understanding. Several interesting trends were noted across interviews and were obtained through both explicit belief statements and accounts of specific health experiences and management decisions. Important findings included a lack of concern for accumulating damage from everyday blood pressure spikes, misattribution of symptoms in the presence of other health conditions (e.g., diabetes), a lack of concern for possible consequences of non-adherence in the absence of perceptible feedback (e.g., physical symptoms), confusion regarding appropriate water and salt consumption with diuretics, and a general lack of awareness or disregard for the importance of potassium levels. These trends in patient understanding likely contribute to adherence decisions and should be explicitly addressed by hypertension treatment plans and interventions. CORRESPONDING AUTHOR: E.P. Polander, Applied Research Associates, United States of America, [email protected]

P16 IMPROVING MEDICATION ADHERENCE AMONG U.S. HYPERTENSIVE BLACKS IN A FAITH-BASED SETTING: A PILOT STUDY L.L. Lewis1, W.L. Taylor2, M.K. Tyler2, L. Patterson-Tyler2, B. Shelton-Dunston3 1 University of Pennsylvania, PHILADELPHIA, United States of America 2 Mother Bethel African Methodist Episcopal Church, PHILADELPHIA, United States of America 3 Philadelphia Black Women's Health Alliance, PHILADELPHIA, United States of America Introduction: Our study compared a spiritually-based education program with standard patient education on antihypertensive medication adherence among hypertensive black church members. Methods: We randomized hypertensive black church members (n=60) to a spiritually-based education program or standard patient education. We assessed antihypertensive medication adherence with self-report (Morisky Scale) at baseline and 3-month follow-up; and with electronic monitoring (EM) at 3 months for both groups. We compared the impact of the spiritually-based education program and standard patient education on self-reported adherence using a mixed between-within subjects ANOVA. To compare the 3-month objective EM adherence, we conducted an independent-samples t-test for the two groups.

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

Results: Participants were female (65.9%), 68.0 + 9 years and nonadherent (69.7%). We found no significant interaction between the two types of interventions and time. There was a substantial main effect for time (Wilks Lambda = .78, F (1, 47) = 13.26, p = .001, partial eta squared = .22) with both groups showing an improvement in adherence. The main effect comparing the two interventions was not significant (F (1, 47) = .95, p = .33, partial eta squared = .02). The spiritually-based education group had a higher EM adherence (M = 84.4%, SD = 23.61) compared to the standard patient education group (M = 79.0%, SD = 28.97) but the difference in scores was not significant (t (41) = .67, p = .50). Conclusions: In a faith-based setting, standard patient education may produce similar improvement in adherence to antihypertensive medication when compared to a spiritually-based education program. CORRESPONDING AUTHOR: Dr. L.L. Lewis, University of Pennsylvania, United States of America, [email protected]

P17 A STUDY ON BLOOD PRESSURE AND ORTHOSTATIC DYSREGULATION COMMUNITY-DWELLING ELDERLY PEOPLE SR Sato Saitama Prefectural University, SAITAMA, Japan Introduction: We have conducted this study to clarify the orthostatic dysregulation in the elderly and their state of health using a measurement of blood pressure. Method: The subjects were elderly people, aged 60 years and above, living in the Tokyo suburban area who were able to perform the activities of daily life independently, and come to a public meeting hall using public transportation. The subjects' blood pressure was measured in three positions (sitting, rising, and standing). Blood pressure and state of health were subjected to a t-test and chi-square test. Subjects received an explanation of the research and gave consent. The permission of the research ethics committee of the university was obtained. Results: There were 108 subjects, 28 men (25.9%) and 80 women (74.1%).The mean age was 73.7 years old. (minimum age: 60 years, maximum age: 87 years, standard deviation: 5.62). When they changed position, the results of blood pressure measurement showed that 33 subjects (30.6%) had orthostatic dysregulation, and 75 had normal blood pressure (69.4%). A significant relationship was observed between orthostatic dysregulation and hypertension. Conclusion: It was found that for elderly people living in an urban area: hypertension was related to orthostatic dysregulation. It was supposed that the reactivity of the vascular decreased for arteriosclerosis. Regarding health management of the elderly, they themselves need to practice selfcare in control of blood pressure and prevention of arteriosclerosis. CORRESPONDING AUTHOR: S.R. Sato, Saitama Prefectural University, Japan, [email protected]

P18 TRENDS AND BURDEN IN DISABILITY IN ACTIVITITES OF DAILY LIVING AMONG CHINESE OLDER ADULTS Y. Liang1, C. Qiu2 1 Jining Medical University, JINING, China 2 Aging Research Center, Karolinska Institutet-Stockholm University, STOCKHOLM, Sweden Introduction: There are inconsistent results on functional disability trend of older adults across countries. We seek to investigate the trends in disability of activity of daily living (ADL) among Chinese adults during 1997-2006. Methods: This study included 7845 participants in the China Health and Nutrition Survey in the years of 1997, 2000, 2004, and 2006 who were

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

aged ≥60 years. Data of demographics, lifestyles, medical history, and ADL were collected through interviews and examinations.ADL disability was defined as having difficulties in at least 1 of the 5 items: bathing, dressing, toileting, feeding, and transferring. We used multiple logistic model to test the trends in ADL disability and linear regression model to estimate the annual changes in prevalence of ADL disability. Results: The prevalence of ADL disability was significantly higher in older age group (0.31%, 0.39%, and 1.32% for adults aged 60-69 years, 70-79 years, and ≥80 years, respectively, P<0.001), in women than men (0.40% vs. 0.33%, P<0.001), and marginally higher in rural residents than urban ones (0.41% vs. 0.31%, P=0.099). After controlling for potential confounders, the prevalence of ADL disability decreased significantly from 1997 to 2006 in total sample and in all subgroups by age, sex, and region (Ptrend<0.05). The estimated number of cases with ADL disability remained relatively stable from 2000 to 2020. Conclusions: The prevalence of ADL disability had decreased from 1997 to 2006 in Chinese older adults. However, as the aging population increases in China, the burden of ADL disability remains a huge challenge. CORRESPONDING AUTHOR: Dr. Y Liang, Jining Medical University, China, [email protected]

P19 NUTRITIONAL STATUS AND A BIOCHEMICAL AND SOCIOBEHAVIORAL ANALYSIS OF INDIAN URBAN ELDERLY R.K. Nigam Rajeev Gandhi College, BHOPAL, India Introduction:The elderly population is the fastest growing population globally and is gaining very sizable proportion Ageing is an issue of global concern and of immense importance in developing countries like India where they constitute 6.7% of population. Aging causes physical, physiological, and psychological changes which leads to changed needs Looking at the specialized needs of the elderly the study of the urban elderly population was done to asses their nutritional status and psychosocial behavior Methods:85 senior citizensof both genders were selected from various areas through geriatric health camps in Bhopal a city of Central India. To assess nutritional status , a mini nutritional assessment chart and anthropometrics measures like BMI, height, weight and waist circumference were used. Biochemical parameters like serum protein, cholesterol, serum iron and hemoglobin were assessed and correlated with anthropometric and behavioral measures (Geriatric Depression Scale). Results:Results were analyzed in line with social framework of Indian Lifestyle and dietary habits shows 50% having normal BMI and 4.3% are underweight and the rest are obese. Waist Circumference of only 17.39% is normal. Anemia of different grades was present in 97.8% of the elderly. Hypo-proteinemia was found in 56.5% the of elderly and more prominent in females (72%). Hypercholesterolemia was found in 63 % of the sample both in males and females. Mild to moderate depression (21.7%) was found to be more prevalent in females (44%) as in males (4%).Conclusion:High incidence of central obesity,hypercholestrolemia and mild to moderate depression in Indian urban elderly CORRESPONDING AUTHOR: R.K. Nigam, Rajeev Gandhi College, India, [email protected]

P20 FACTORS ASSOCIATED WITH WATCHING OF SOCIALLY ISOLATED ELDERLY PEOPLE Y.K. Kanzaki, Emiko Saito Tokyo Metropolitan University, TOKYO, Japan

S11

Introduction Prevention of social isolation is a significant challenge for support specialists of community-dwelling elderly people in Japan. As health and medical specialists, nurses at Community General Support Centers (CGSCs) occasionally watch elderly people but the actual level of support provided is unclear. The purpose of this study was to clarify the characteristics and working conditions of CGSC nurses, and the characteristics of the elderly population associated with nurse watching of socially isolated elderly people. Methods This was a cross-sectional study using an anonymous self-administered questionnaire mail survey. Participants included approximately 6,642 nurses of CGSCs throughout Japan. Data collection took place in July 2013. Logistic regression analysis was used to analyze the data. Results The responses of 1,098 CGSC nurses were analyzed. CGSC nurses spent a mean of 3.7 hours per week watching elderly people. Multivariate logistic regression analysis showed that watching is significantly associated with the opportunities to interact with other CGSC nurses (odds ratio (OR) = 1.54, 95% confidence interval (CI) [1.04-2.29]) and availability of staff for planning specific preventive long-term care (LTC) support (OR = 1.41, 95% CI [1.01-1.97]). Conclusions We clarified time spent by CGSC nurses in watching socially isolated elderly people and associated factors. This study suggests that if nurse watching of the elderly is a CGSC priority, it will be necessary to review CGSC workloads and the adequacy of personnel, and to create opportunities for interaction with others in the same occupation having the same expertise. CORRESPONDING AUTHOR: Y.K. Kanzaki, Tokyo Metropolitan University, Japan, [email protected]

P21 EPIDEMIOLOGICAL STUDY OF FACTORS ASSOCIATED WITH SUBJECTIVE WELL-BEING OF COMMUNITY-DWELLING ELDERLY PERSONS S.M. Murata Kyoto Tachibana University, KYOTO, Japan Introduction: This study aimed to examine factors associated with perceived health of community-dwelling elderly persons, including their self-evaluation of “forgetfulness”. Methods: Subjects were 3,357 community-dwelling elderly persons (mean age: 74.5 ± 6.6 years old, male: 1,364, female: 1,993), and their basic attributes (age, gender, height, weight), perceived health, experience of a fall, the index of competence, motorfunction, level of being homebound, and forgetfulness were investigated using aself-administered questionnaire. Results: Regarding perceived health, a total of 2,349and 1,008 elderly answered that they were “healthy”and “unhealthy”,respectively, and no significant difference was observed in the sex ratio. As the results of logistic regression analysis using perceived health as a dependent variable, significant odds ratios were observed for age, experienceof a fall, the TMIG (Tokyo Metropolitan Institute of. Gerontology Index of Competence; TMIG) index of competence, motor function, level of being homebound, and forgetfulness. Conclusion: Results indicated that maintaining a good motor function and physical ability, and keeping themselves active outside their home are factors that they can recognize themselves as being healthy. Awareness of forgetfulness was also clarified as a factor affecting their perceived health, suggesting that the self-evaluation of forgetfulness and its support may contribute to promoting the perceived health of elderly persons. CORRESPONDING AUTHOR: SM Murata, Kyoto Tachibana University, Japan, [email protected]

S12

P22 AGING IN PRECARIOUS TIMES: PSYCHOSOCIAL RESOURCES FOR POSITIVE AGING IN MIDDLE-AGED GERMANS C. Craciun, U Flick Freie Universität Berlin, BERLIN, Germany Introduction: Nowadays demographic and economic changes affect the health and pension system, contributing to rising health and social inequalities in old age. The present study represents preliminary results of a larger ongoing research. It aims to identify resources that help middleaged people become proactive in preparing for old age in a precarious social context, characterised by financial (i.e. insecure job opportunities) and psychological insecurity (i.e. not being able to make long term plans). Methods: A mixed method design was applied, using qualitative and quantitative data (DEAS survey). Semi-structured interviews were conducted with middle-aged people (10 with and 10 without pension insurance). Quantitative data from a German representative sample (N=2072, 50,6% women, 49,4% men) of individuals, aged 40-54, were analysed. Results: Thematic analysis revealed self-efficacy, positive images of aging, planning, social networks and relations as resources for positive aging in both insured and uninsured participants. Hierarchical regression analysis showed social networks and positive images of aging as best predictors of health and positive emotions in both gender and social security categories. Conclusions: Findings show the importance of positive images of aging and the development of good social networks as resources for positive aging in a precarious social context. Implications for behaviour change interventions and policy are discussed. CORRESPONDING AUTHOR: Dr C. Craciun, Freie Universität Berlin, Germany, [email protected]

P23 PATIENT RELEVANT OUTCOMES IN EVALUATING SENIOR COMMUNITY CENTERS S. Lippke, JP juliane Paech Jacobs Center, BREMEN, Germany Introduction: Behavioral medicine in terms of health care of older people is of central importance in times of demographic change. Senior community centers may be an imperative part of prevention and health promotion. Methods: With the aim to evaluate the potentials of senior community centers with patient relevant outcomes, N=105 individuals (46-95 years old; M=70 yrs; 67% female) were assessed regarding their user”behavior and their age. They were asked regarding their calendar age, as well as regarding their perceived physical and mental age. Results: Results indicated that individuals who visited a senior community center regularly were older at the mean level (M=76 years) than individuals who did not use senior community centers at all (M=63 years). All study participants perceived themselves physically 10 years younger than they actually were regarding their calendar age. Cognitively, they felt younger, too; however, significant differences transpired between the groups: The non”users perceived themselves as being 13 years younger, the infrequent users 18 years and the regular users 23 years younger than they were in regard to their chronological age (F (82, 2) =5.10; p<.05). Conclusion: The findings direct towards the potentials of senior community centers and especially towards the mental health promotion. The importance of patient relevant outcomes in behavioral medicine and health care research also becomes apparent: Subjective age can serve as an outcome which indicates advantages especially crucial in times of demographic change. CORRESPONDING AUTHOR: Prof. S. Lippke, Jacobs Center, Germany, [email protected]

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

P24 FALL RISK FACTORS IN PSYCHOGERIATRIC RESIDENTS NM Kosse1, MH de Groot2, T Hortobágyi1, CJC Lamoth1 1 University of Groningen, University Medical Center Groningen, GRONINGEN, Nederland 2 Department of Geriatric Medicine, Slotervaart Hospital, AMSTERDAM, Nederland INTRODUCTION Falls in psychogeriatric residents represent a costly but unresolved safety issue. To develop an effective fall prevention program fall risk factors must be identified. Therefore, we aimed to examine the relationship between patient characteristics (factors) and fall rate in psychogeriatric residents. METHODS Twenty nursing home residents of a psychogeriatric ward diagnosed with dementia (80±11 years; 60% male) participated. Based on patient records, we identified 65 factors and extracted the number of falls (115 falls; 5.1±6.7 falls/person years) over 19 months. The factors represented seven domains: demographics, ADL, mobility, cognition and behaviour, vision and hearing, medical conditions, and drug-use. We evaluated the relationship between factors and fall rate using multivariate Partial Least Squares regression. A variable importance in projection (VIP) value of >1 indicated important factors in the model. RESULTS The 65 factors included in the model explained 88% of the variance in fall rate. Factors in the domains mobility, ADL, cognition and behaviour, medical conditions and drug-use contributed most to the model. Walking difficulties, indicators of disinhibited behaviour, and inactivity were strongly related to fall rate (VIP>1). Demographics, vision and hearing were less important (VIP<1). CONCLUSION Dementia is considered as fall risk factor. However, our results showed that especially cognitive impairment related to disinhibited behaviour was associated with a high fall rate in psychogeriatric residents. Moreover, being immobile and inactive seems to decrease fall rate. The analytic approach provided a detailed view of the fall risk factors and their interactions in psychogeriatric residents and enables more effective fall prevention. CORRESPONDING AUTHOR: NM Kosse University of Groningen, University Medical Center Groningen Nederland [email protected]

P25 THE IMPACT OF WORK MOTIVATION AND JOB SEARCH BEHAVIOR ON REEMPLOYMENT AMONG UNEMPLOYED AGED 45+ N. Velterop1, H.K.E. Oldenhuis2, J.F. Ybema3, J.J.L. Klink, van der1, L Polstra2, S. Brouwer1 1 University Medical Center Groningen, GRONINGEN, The Netherlands 2 Hanze University of Applied Sciences, GRONINGEN, The Netherlands 3 Netherlands Orgnisation for Applied Scientific Research, HOOFDDORP, The Netherlands Introduction: The aims of this study were to (1) examine the associations of intrinsic and extrinsic work motivation with preparatory and active job search behavior and (2) investigate how intrinsic and extrinsic work motivation as well as preparatory and active job search behavior predict reemployment among older unemployed individuals (aged 45+) at one-year follow-up. Method: We investigated a sample of 502 unemployed Dutch citizens aged 45-64 years that completed questionnaires within the longitudinal Study on Transitions in Employment, Ability and Motivation (STREAM). From the original database we included persons in three ways: (1) persons who were unemployed at T1 (n=128); (2) persons who became unemployed at T2 (n=99); (3) persons who were unemployed at T1 and T2 (n=275). Two time-waves were conducted: (1) T1-T2 and (2) T2-T3. These two time-waves were merged. Logistic regression analyses were performed. All analyses were adjusted for age, gender, education level, duration of unemployment, financial situation and time-wave.

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

Results: Intrinsic work motivation was related to preparatory as well as active job search behavior. Extrinsic work motivation was not related to any type of job search behavior. Work motivation was not related to reemployment. Active job search behavior was related to reemployment. Conclusion: Intrinsic motivation to work was predicting preparatory and active job search behavior and active job search behavior was important for getting reemployed. This study suggests that it is important to focus on job search strategies and not so much on financial incentives in supporting older unemployed to re-enter the labour market. CORRESPONDING AUTHOR: MSc. N Velterop, University Medical Center Groningen, The Netherlands, [email protected]

P26 TRENDS IN DEPRESSION AND WELL-BEING IN HUNGARIAN CANCER PATIENTS. A LONGITUDINAL STUDY E.K. Kallay1, D.L. Cs Degi L. Csaba1, K.M. Kopp Maria2, K.E. Kallay Eva1 1 Babes-Bolyai University, CLUJ-NAPOCA, Romania 2 Semmelweis University, BUDAPEST, Hungary Hungary has the highest rate of cancer-deaths in Europe, but longitudinal studies investigating psychosocial aspects of cancer-related distress are unknown. Our study aims to investigate major tendencies in cancer rate, trends in depression and well-being among Hungarian cancer patients. Our study is based on data obtained from Hungarostudy Epidemiological Panel, a nationally representative follow-up survey assessing healthrelated functioning in the Hungarian adult population. Self-report data were collected in three waves: 2002, 2006, 2013. Our results are based on the complete results of 2000 participants, assessing well-being (WHO-Well-being Index) and depression (Beck Depression Inventory). Statistical analyses used non-parametric tests. The number of diagnosed male cancer patients has significantly increased 2002-2006, and slightly decreased in 2013. In female participants, after the 2006 increase, cancer rates stagnated. After a significant increase from 2002 to 2006, depression decreased in male patients, while in female patients depression slightly decreased. In male participants well-being significantly decreased from 2002 to 2006, and than significantly increased in 2013. In female patients well-being constantly increased from 2002 to 2013. High cancer mortality rates were confirmed as significant factors associated with psychosocial burden of cancer. Depression contributes to adverse health effects over time, though cancer patients might counterbalance these effects with a “response shift”adaptation processes, reflected in favorable changes in subjective well-being. The practical value of our results consists in the delineation of tendencies, and identification of specific groups that may be more vulnerable to changing trends in cancer-distress. [Grant supported by CNCS - UEFISCDI (PN-II-RUTE2012-3-0011)]. CORRESPONDING AUTHOR: Lecturer EK Kallay, Babes-Bolyai University, Romania, [email protected]

P27 EFFECTS OF PSYCHOLOGICAL INOCULATION VERSUS SUPPORTIVE LISTENING ON OPTIMISM AND HOPELESSNESS IN BREAST CANCER PATIENTS Y. Gidron1, A.G. Gomes2, I.L. Leal2 1 Free University of Brussels (VUB), BRUSSELS, Belgium 2 ISPA, LISBON, Portugal Introduction: Studies have shown that helplessness accelerates tumors in animals and that helplessness-hopelessness (HH) predicts poor cancer prognosis in cancer patients. However, few studies tried to specifically

S13

reduce levels of HH in cancer patients. This study preliminarily tested the effects of “psychological inoculation”(PI) on implicit optimism and explicit HH. Methods: Portuguese women with breast cancer completed online measures of implicit optimism (a word-search task) and a scale of HH (from the mini-MAC), before and immediately after being randomized to view one of two brief films. In the implicit task, women had to find in 1min as many words as possible, where 5 were optimistic, 5 pessimistic. This scale previously predicted cancer survival. One film depicted a patient rejecting “challenging sentences”which reflected cognitive distortions underlying HH (PI film). The other film depicted the same patient speaking about her concerns and receiving supportive listening and empathy (control film). Results: The PI film led to significant increases in implicit optimism, while the control film led to significant reductions in explicit HH. Explicit HH was not correlated with implicit optimism scores in the present study. Conclusions: This study found preliminary evidence for the effects of PI on an indirect measure of optimism, and demonstrates the feasibility for possibly influencing patients’ psychological prognostic factors by merely viewing a PI-method film. The study also reveals the complexity of psychological assessment. Future studies need to test the long-term impact of these interventions and consider combining them to maximize effectiveness on reducing HH. CORRESPONDING AUTHOR: Prof. Y. Gidron, Free University of Brussels, (VUB), Belgium, [email protected]

P28 DEPRESSION AND ITS CORRELATES IN HUNGARIAN CANCER PATIENTS E.K. Kallay1, D.L. Cs Degi L Csaba1, K.E. Kallay Eva1, K.M. Kopp Maria2 1 Babes-Bolyai University, CLUJ-NAPOCA, Romania 2 Semmelweis University, BUDAPEST, Hungary The diagnosis and prolonged treatment with cancer have a profound impact on the patients’ life, which induce increased, debilitating levels of distress. Our study aims to investigate the major correlates of cancerinduced depression in Hungarian cancer patients. Our study is based on data obtained from the 2013 Hungarostudy Epidemiological Panel, a nationally representative follow-up survey assessing health-related functioning in the Hungarian adult population. Our results are based on the complete results of 2000 participants from the general population, 49 of which were diagnosed with different forms of cancer. We assessed demographic variables, depression (Beck Depression Inventory), well-being (WHO Well-being Index), sense of coherence, self-efficacy, and anomie. Statistical analyses used non-parametric tests. Our results indicate that in 2013 cancer patients were significantly more depressed, had significantly lower levels of well-being, and self efficacy than the general population. Within the group of cancer patients, we found strong positive correlations between depression and anomie, and strong negative correlations between depression and well-being, sense of coherence, self-efficacy. Regression analyses indicate that after controlling for age and gender, anomie, sense of coherence, self-efficacy and well-being explain 51% in the variance in depression within the group of cancer patients. Depression in cancer patients significantly impacts both the progression of the illness, and the quality of life of the patients. The practical value of our results consists in the identification of those areas of functioning where interventions may have positive effect. [Study was supported by CNCS - UEFISCDI (number PN-IIRUTE-2012-3-0011)]. CORRESPONDING AUTHOR: Lecturer EK Kallay, Babes-Bolyai University, Romania, [email protected]

S14

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

P29 AGE AT SEXUAL ONSET AND ILNESS PERCEPTION AS PREDICTORS OF SELF CARE IN CERVICAL CANCER A.C. Peuker, Priscila Lawrenz, Natália Britz de Lima, Fernanda Romeiro, Elisa Kern de Castro UNISINOS, PORTO ALEGRE/RIO GRANDE DO SUL, Brazil

P31 ILLNESS PERCEPTIONS AMONG PATIENTS WITH HEAD AND NECK CANCER AND THEIR CAREGIVERS: RELATIONSHIPS WITH Q.O.L. A.E.R. Richardson, R. Morton, E. Broadbent University of Auckland, AUCKLAND, New Zealand

Introduction: The scarce financial resources, limitations of health services and the quality of professional attention in Brazil, undermine the possibility of broad screening of cervical cancer and turn the disease control into a major challenge. Economically, self-care plays a key role, serving as bridge between scarcity and demand of health services. The aim of the study was to test a predictive model of self-care, considering the dimensions of illness perception and risk factors (age at sexual onset and number of sexual partners). Methods: In this cross-sectional study, 120 women (mean age of 40.38 years; SD= 12) were enrolled consecutively, from a primary care unity in southern Brazil. It used a sociodemographic and clinical data form, the Revised Illness Perception Questionnaire for Healthy People and the Questionnaire of preventive self-care behaviors. Results: Data analyses included linear regression (stepwise) and found that 'treatment control' and 'Age of first sexual intercourse”were predictors of self-care, 9.3% and 4.2% of the variance respectively. This model showed better self-care in women with higher perception that treatment controls illness and latter sexual initiation. The tested model explained 13.5% of the variance in self-care. Conclusion: The data indicates that psychological factors and behavioral aspects should also be considered in the design of targeted actions to prevent and control cervical cancer.

Introduction: This study investigated illness perceptions in patients with head and neck cancer (HNC) and their caregivers. We investigated whether these were associated with patient health-related quality of life (HRQL) and caregiver depression and anxiety. Method: Ninety-eight newly diagnosed HNC patients and their caregivers (n=78) completed questionnaires assessing depression, anxiety, and illness perceptions. Patients also completed a HRQL questionnaire. Results: The illness perceptions of caregivers were significantly more negative than patients with respect to: consequences, timeline, treatment, and the emotional impact of HNC. Caregivers also reported a better understanding of HNC than patients. Patient perceptions of worse consequences, longer timeline, stronger identity, and greater emotional impact of HNC were significantly associated with poor HRQL, while higher control perceptions were associated with better HRQL. Caregiver perceptions of worse consequences of HNC for their own life were significantly associated with caregiver anxiety. Low caregiver perceptions of the patient’s control, as well as their own control, were significantly associated with caregiver depression. The interaction between patient and caregiver perceptions of consequences, concern, treatment, identity, and emotional impact of HNC explained additional variance in patient HRQL, above and beyond patients’ own illness perceptions. Conclusion: Caregivers’ illness perceptions tend to be more negative than patients’ perceptions and are associated with their own depression and anxiety. Caregivers’ perceptions interact with patients’ perceptions to influence patient HRQL. Both patients and caregivers should be included in psychological interventions for HNC to improve their psychological outcomes and patient HRQL.

CORRESPONDING AUTHOR: PhD A C Peuker, UNISINOS, Brazil, [email protected] P30 'EXPERIENCES OF SHELTERING' AS THE GOAL OF THE ONCOLOGY CARE SETTING C. Laranjeira Piaget Institute, VISEU, Portugal Introduction. Numerous conceptual models offer a wide array of definitions and insights into the environmental patterns that promote health within the individual. The main purpose of this study was to understand what factors of the healthcare environment are perceived as being important to patients in oncology care setting. Method. The sample was 10 patients with different cancer diagnoses in an oncology ward at a Portuguese District Hospital. A qualitative design was adopted using focus group interviews. The verbatim transcripts of the interviews were analysed using manifest content analysis, which is a wellknown method for descriptive analysis of a text. Results. Analysis of the patients’ perceptions of the environment indicated a complex entity comprising several aspects. These came together in a structure consisting of three main themes: feelings of at-homeness, supportiverelationships with the staff, and physical space. The care environment is perceived as a complex entity, made up of several physical and psychosocial aspects, where the physical factors are subordinated by the psychosocial factors. It is clearly demonstrated that the patients wanted opportunities for good encounters with staff, fellow patients, and family members, supported by a good physical environment; and the patients valued highly a place to withdraw and rest. Conclusions. This study presents those attributes that are valued by cancer patients as crucial and important for the support of their wellbeing and functioning. The results show that physical aspects were subordinate to psychosocial factors, which emerged strongly as being the most important in a caring environment. CORRESPONDING AUTHOR: Prof. C. Laranjeira, Piaget Institute, Portugal, [email protected]

CORRESPONDING AUTHOR: A.E.R. Richardson, University of Auckland, New Zealand, [email protected]

P32 SCREENING FOR DISTRESS AND CONCERNS IN LUNG CANCER: AN INTERDISCIPLINARY TASK S. Simard1, B. Fournier2, L. Fradette2, C. Gagné2, L. Tremblay3 1 Laval University, QUÉBEC (QC), Canada 2 Institut universitaire de cardiologie et de pneumologie de Québec, QUEBEC, Canada 3 Medecine, Laval University, QUÉBEC (QC), Canada Introduction: Lung cancer is often diagnosed late in the disease trajectory and associated with a rapid progression. Recently, several Canadian cancer centres have implemented systematic distress screening programs during their cancer trajectory to ensure the most appropriate care services. Objective: To explore the distress of lung cancer patients within an ambulatory oncology clinic setting. Methods: A total of 806 lung cancer patients, at different phases of their illness, completed a distress screening tool (DST) as part of their clinical routine care. The DST includes the Distress Thermometer (DT), an adaptation of the Canadian Problem Checklist (CPC) and the Edmonton Symptom Assessment System (ESAS). Results: The mean DT score was 3.4 and 37% of patients expressed a clinical level of distress (score ≥ 5). The mean ESAS score was 21.4 and shortness of breath, fatigue as well as appetite were the highest symptoms reported. The mean number of CPC was 8.6 (range 044) and was related to DT (r = 0.54) and ESAS (r = 0.56). The most frequent problems reported were fatigue, dyspnea, fear, pain, coping with cancer and sleep. One third of patients wished to receive help. Conclusion: Routine screening for distress is necessary to adequately address the

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

patients’ needs and to better organized interdisciplinary services in order to achieve better person-centered care. CORRESPONDING AUTHOR: Dr. S Simard, Laval University, Canada, [email protected]

P33 PERSONALITY CORRELATES OF BREAST CANCER PATIENTS S.K. Kreitler, M.M. Kreitler Tel Aviv University, TEL-AVIV, Israel Background: Previous studies about personality correlates of cancer patients led to inconclusive results Studies by Kreitler et al. (2002) showed that focusing on theoretically-relevant factors provides new insights into personality correlates of cancer patients. The study was done in the framework of the Cognitive Orientation (CO) theory of health behavior and wellness which enables identifying relevant factors in cancer patients. Purpose: The goal was to examine whether personality tendencies grounded in the CO theory can be identified in breast cancer patients. Method: The participants were 250 breast cancer patients and 180 matched healthy controls. They were administered a questionnaire of demographic and medical information, and the CO questionnaire of breast cancer assessing personality dispositions identified in pretests as relevant for breast cancer. Results: Discriminant and logistic regression analyses showed that patients and controls differed significantly in most personality dispositions, including concern with controlling oneself and others, dependence on others’ evaluations, emotional blocking, perfectionism, and conflicts about self identity and giving to others. Some of these variables were related to medical features, none to demographic ones. Conclusions: There exists a relevant set of psychological correlates of breast cancer patients that could serve as basis for psychological interventions accompanying medical treatments and needs to be examined in other cultural settings. CORRESPONDING AUTHOR: Prof SK Kreitler, Tel Aviv University, Israel, [email protected]

P34 PREVALENCE AND IMPACT OF SEVERE FATIGUE IN PATIENTS WITH GASTROINTESTINAL STROMAL TUMORS (GIST) H. Poort, M.F. Gielissen, J.A. Custers, R. Tielen, W.T. van der Graaf, H. Knoop Radboud university medical center, NIJMEGEN, Nederland INTRODUCTION: GIST is a rare sarcoma of the gastrointestinal tract. Since the first oral tyrosine kinase inhibitor (TKI) imatinib was introduced, the outcome of patients with metastatic GIST has improved from less than 1 year to median 5 years. The prolonged survival and the chronic use of TKIs are related with new disease- and treatment induced problems, which can compromise quality of life. One class effect of TKIs is fatigue, but this has not been studied in depth yet in GIST patients. This study determines the prevalence and impact of severe fatigue compared to a group of matched healthy population-based controls. METHOD: Eighty patients treated with surgery and/or a TKI for locally advanced or metastatic GIST were asked to participate. Participants filled out questionnaires including the Checklist Individual Strength (fatigue), EORTC-Quality of Life Questionnaire, Fatigue Catastrophizing Scale, Self-Efficacy Scale, and Hospital Anxiety and Depression Scale. RESULTS: 54 patients (age 22-84) returned questionnaires. The prevalence of severe fatigue was significantly higher in GIST patients (31.5%) than in 160 matched healthy controls (11.9%). Fatigued patients reported lower quality of life, more impairment on functional domains, and experienced more physical symptoms. A multiple regression showed that use of a TKI, fatigue catastrophizing and psychological distress were associated with fatigue.

S15

CONCLUSIONS: One in three patients with GIST is severely fatigued. These patients experience more functional impairments and symptoms. Besides the use of a TKI, psychosocial factors were also associated with fatigue. Behavioral interventions aimed at these psychosocial factors to reduce fatigue can be considered. CORRESPONDING AUTHOR: H Poort, Radboud university medical center, Nederland, [email protected]

P35 THE CONTRIBUTION OF BEHAVIOURAL, CANCER-RELATED AND PSYCHOLOGICAL DETERMINANTS TO HEALTH BEHAVIOURS OF CANCER SURVIVORS I.M. Kanera1, C.A.W. Bolman1, R. A. Willems1, I. Mesters2, A.A.J.M. Beaulen1, J.M.J. Geurts1, L. Lechner1 1 Open University of the Netherlands, HEERLEN, Nederland 2 Maastricht University, MAASTRICHT, Nederland Introduction. Healthy lifestyle is important for cancer survivors. To develop lifestyle interventions for them, insight into the predictors of vegetable-, fruit-, alcohol consumption and smoking behaviour, is needed. The study aims to describe lifestyle risk behaviours of cancer survivors and to analyse the contribution of demographic, cancer-related, psychological and social cognitive factors in explaining these behaviours. Method. Participants were 279 cancer survivors (68.8% females; 31.2% males; mean age 60), treated with surgery, chemotherapy and or radiation. Participants were invited through 8 Dutch hospitals to participate in this crosssectional study. Validated questionnaires (e.g. EORTC-QLQ-C-30, MAC, IPQ-SF, HADS) were administered to identify cancer-related, psychological and social cognitive determinants and to assess the lifestyle behaviours. Preliminary results. A fifth of the participants smoked, a fifth did not meet the recommendations for alcohol use and one third did not meet the fruit intake recommendation. However, 70% did not meet the vegetable intake norms. Explained variance of lifestyle behaviours ranged between 40-56 %. Except in alcohol consumption, attitude and self-efficacy were the strongest predictors. Their influence decreased by adding intention to the model. Alcohol use was associated with anxiety, several EORTC-QLQC-30 subscales and illness perception next to self-efficacy and intention. Education, gender and BMI were minimally related to most behaviours while mental adjustment was only associated with fruit consumption. Conclusions. Our study confirmed that lifestyle behaviours in cancer survivors are eligible for improvement. Especially vegetable consumption is poor. Because the behaviours differ in their determinants, interventions need to be tailored to the target behaviour. CORRESPONDING AUTHOR: MSc I.M. Kanera, Open University of the Netherlands, Nederland, [email protected]

P36 THE TIRED STUDY: TWO INTERVENTIONS FOR FATIGUED ADVANCED CANCER PATIENTS, A RANDOMISED CONTROLLED TRIAL H. Poort, CAVerhagen, ME Peters, MM Goedendorp, WT van der Graaf, H Knoop Radboud university medical center, NIJMEGEN, Nederland INTRODUCTION: Fatigue is an important symptom lowering the quality of life (QoL) in patients with advanced cancer (AC). Graded exercise therapy (GET) and cognitive behaviour therapy (CBT) have shown to be effective in reducing fatigue in cancer survivors, but the effectiveness for patients receiving systemic palliative treatment has not been demonstrated. We started a study to test the effects of both interventions in patients with AC. Mediators of the expected reduction in fatigue will be determined.

S16

METHOD: TIRED is a randomized controlled multicentre intervention trial with three conditions (GET, CBT, and care as usual). GET consists of weekly sessions of two hours resistance and aerobic training with a physical therapist during 12 weeks. CBT consists of ten individual one-hour sessions with a cognitive behavioural therapist over a period of 12 weeks. Both interventions are designed to reduce fatigue in patients with advanced cancer. Secondary endpoints are functional impairment and QoL. RESULTS: 219 patients diagnosed with advanced breast or colorectal cancer will be recruited. All patients who have been severely fatigued for at least two weeks without known and treatable somatic cause (other than cancer related), and scheduled to receive first line of palliative cancer treatment are eligible. Post-treatment measures are expected to be completed in December 2015. CONCLUSIONS: TIRED will provide insight in the effectiveness of GET and CBT specially designed to target severe fatigue in patients with AC. Identifying the mediating factors for both interventions will enable us to improve future interventions for fatigue in this patient group. CORRESPONDING AUTHOR: H Poort, Radboud university medical center, Nederland, [email protected]

P37 ENGAGING CARDIOLOGISTS IN TOBACCO CONTROL M.B. Aghi Advocacy Forum for Tobacco Control, NEW DELHI, India Introduction Long term smoking of tobacco material has been observed to result in heart ailments. This finding has been supported by large scale studies. It has also been found that people who smoke or chew tobacco are prone to sudden heart attacks as they have a ‘plaque rupture’, which is instant and which results in the heart losing its ability to pump blood. Objectives All health care professionals should advice his patients not to smoke or chew tobacco. It is imperative that this suggestion be specially implemented by cardiologists Methodology Cardiologists are best placed to advice their patients not to use tobacco. They should not only advise but also work towards prevention. It has been revealed that their beliefs are a major barrier in their working towards eliminating tobacco use among their patients and they do not feel that it is their call to interfere in the health style practices of people. They need to recognize that their professional responsibility extends beyond treatment of tobacco-caused disease. Their treatment should include prevention and cessation. Results It has been demonstrated by many major researches that quitting tobacco use reduces the risk of heart attack and death to the levels of people who have never smoked. Prevention efforts can achieve much more. A cardiologist may save more lives by lobbying for tobacco-control laws for a few hours than by treating cardiac ailments lifelong. Conclusion There is no room for speculation. Let them respond to the wake-up call. CORRESPONDING AUTHOR: Dr MB Aghi, Advocacy Forum for Tobacco Control, India, [email protected]

P38 EFFECT OF PERITONEAL DIALYSIS ON LIFE QUALITY IN PATIENTS WITH REFRACTORY END STAGE HEART FAILURE PA Alivanis, Eirini Tsampikaki, Christos Paliouras, Georgios Ntetskas, Fotini Lamprianou, Costantin Roufas, Nikolaos Karvouniaris, Georgios Aperis General Hospital of Rhodes, RHODES, Greece

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

AIM Improvement of self-perceived quality of life in in patients with congestive heart failure (CHF) on peritoneal dialysis (PD). METHODS Thirteen patients with NYHA class IV cardiac failure were enrolled. None of the patients suffered from primary renal disease, but 11 out of 13 had evidence of cardiorenal syndrome. After initial haemofiltration the patients were converted to PD. Before and one month after initiation of the treatment all subjects completed the 36-item short form generic questionnaire (SF-36 v2). Answers were evaluated comparatively in the eight parameters measuring respective health concepts. RESULTS During a one year follow-up pre and post dialysis, both frequency and duration of hospitalization was reduced significantly (from 3.2 ±1.7 to 1.6 ±1.3 and from 45.5 ±32.4 to 20.2 ±33.6 days respectively, p<0.05). Except for the bodily pain, there was a significant improvement in the health status in all other parameters evaluated before and after the peritoneal dialysis implementation (Physical Functioning 17±2.5 vs 31.4±10.5, Role Physical 27.9±0.2 vs 45.1±12.1, General Health 21.2±3.5 vs 31.6±9.1, Vitality 30.7±10.1 vs 44.7±14.6, Social Functioning 27.7±10.8 vs 40.1±12.7, Role Emotional 31.2±11.7 vs 41.8 ±13.2 and Mental Health 19.9±13.6 vs 36.5±19.3, p<0.005 to 0.05). Thus, overall physical and mental component scales were also improved significantly (from 29.5±5.5 to 37.8±5.8 and from 28.1±14.6 to 41.8±16.5 respectively, p<0.02) CONCLUSION In patients with CHF without ESRD, the use of PD not only result in an improvement in symptomatology and reduction in hospitalization, but also improves the quality of life which remains the ultimate goal of health care. CORRESPONDING AUTHOR: PA Alivanis, General Hospital of Rhodes, Greece, [email protected]

P39 SYMPTOM AND COURSE HETEROGENEITY OF DEPRESSION FOLLOWING MYOCARDIAL INFARCTION AND THE ASSOCIATION WITH LONG-TERM MORTALITY A.M. Roest, K.J. Wardenaar, P. De Jonge University Medical Center Groningen, GRONINGEN, Nederland Introduction: Depression following myocardial infarction (MI) is associated with adverse prognosis. Previous studies suggested that both the longitudinal course and type of depressive symptoms predict cardiac outcomes. However, no studies have combined both aspects. The aim of this study was to assess whether MI patients can be empirically classified based on course trajectories of cognitive/affective (CA) and somatic/affective (SA) symptoms, and whether these classes differentially predict long-term mortality. Method: Patients with acute MI (n=437) were recruited between 1997 and 2000. Patients provided Beck Depression Inventory (BDI-I) data at four time-points (baseline, 3, 6, 12 months) and were followed up until 2007 for all-cause mortality. We used dualtrajectory growth mixture modeling to identify subgroups based on their simultaneous temporal patterns on CA and SA symptoms. Results: Four subgroups were identified: ‘No depression’: consistently low CA and SA (n=298); ‘Somatic depression’: consistently low CA and high SA (n=100); ‘Moderate depression’: consistently high CA and SA (n=27); ‘Severe depression’: high and increasing CA and SA (n=12). There were significant differences between classes, including on the Global Registry of Acute Coronary Events (GRACE) risk score (p=0.001). ‘Somatic depression’ (HR:1.82;95% CI:1.14-2.90;p=0.01), but not ‘moderate depression’ (HR:1.35;95%CI:0.48-3.77;p=0.57), or ‘severe depression’ (HR:1.26;95% CI:0.30-5.24;p=0.75), was an independent predictor of mortality compared with ‘no depression’.Conclusions: MI patients differed on severity, course, and type of depressive symptoms. Only a subgroup characterized by merely persistent somatic depressive

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

symptoms predicted mortality. The increased mortality risk associated with depressive symptoms following MI probably reflects somatic complains in general instead of actual depression. CORRESPONDING AUTHOR: A.M. Roest, University Medical Center Groningen, Nederland, [email protected]

P40 ACHIEVING DIET AND ACTIVITY CHANGE: MODIFYING SOCIAL-COGNITIVE FACTORS AMONG CORONARY HEART DISEASE PATIENTS (STUDY DESIGN) C. Thomas1, G.W.B. Näring1, M.O.M. Van deVen11, H. Ginting2, E.S. Becker1 1 Radboud University / Behavioural Science Institute, NIJMEGEN, Nederland 2 Maranatha Christian University, BANDUNG, Indonesia Background: The 2007 Indonesian Basic Health Research Survey revealed that cardiovascular disease is the leading cause of mortality in Indonesia, accounting for 30% of total deaths. Two important modifiable risk factors are sedentary lifestyle and having an unhealthy diet. Despite of the fact that these factors are preventable and modifiable, it is difficult for patients to change their health behaviour. Purpose: Applying the Health Action Process Approach (HAPA) model as a theoretical basis, the study investigates the effectiveness of an intervention to modify social-cognitive factors, barriers and resources for patients to change eating and exercise habits. Method: The study is designed as a randomized controlled trial. Patients will be randomly assigned into three groups: the care as usual group, the experimental group that receives lifestyle change treatment, and a group that receives social media peer support in addition to the lifestyle change treatment. Behaviour change techniques used in the sessions will include goal setting, action planning, setting graded tasks, barrier identification, prompting self-monitoring of behaviour, and providing feedback on the participant’s performance. Measurement of HAPA social-cognitive constructs, clinical outcomes, negative emotions, and eating and physical exercise habit will be conducted before and after the treatment, and 3 months after the intervention. Results will be analysed using repeated measures ANOVA. Significance: The findings will provide tools to change dietary and physical exercise behaviour, particularly among patients living in a tropical climate and Southeast Asian, a setting which has not been adequately explored in previous studies. CORRESPONDING AUTHOR: C. Thomas, Radboud University / Behavioural Science Institute, Nederland

P41 ASSESSMENT OF COPING IN A CHRONIC HEART FAILURE PATIENT POPULATION: THE BRIEF COPE RESTRUCTURED C. Nahlén Bose1, G. Björling2, M.L. Elfström3, H. Persson4, F. Saboonchi2 1 Karolinska Institutet, STOCKHOLM, Sweden 2 Karolinska Institutet and The Red Cross University College, STOCKHOLM, Sweden 3 Mälardalen University, Academy of Health, Care and Social Welfare, ESKILSTUNA/VÄSTERÅS, Sweden 4 Karolinska Institutet, Department of Clinical Sciences Danderyd Hospital, STOCKHOLM, Sweden Introduction:Individuals with chronic heart failure (CHF) need to cope with both the physical limitations and the psychological impacts of the disease. Whereas some coping strategies are beneficial, other coping strategies are linked to increased mortality and worse health-related quality of life. Assessment of coping strategies provides valuable information to health care professionals. Brief COPE measures 14 coping strategies and has been previously used in the context of CHF. The

S17

purpose of this study was to examine a theoretical second order 4-factor model of Brief COPE consisting of Problem focused coping, Avoidant coping, Socially supported coping and Emotion focused coping. Method:Patients (n=183) with CHF were recruited via an out-patient heart failure clinic or at a cardiac ward and filled in Brief COPE. Confirmatory Factor Analysis was performed on the 4-factor model of Brief Cope. Results: A 4-factor model of Brief COPE showed fairly good model-fit to the data, the fit indices were : c2 = 88.89, df = 49, p < 0,001, CFI = 0,914, RMSEA = 0,067. Avoidant coping did not correlate significantly to any of the other second-order factors. Conclusion: Establishing this model facilitates the assessment of coping in CHF patient populations in both clinical and research settings. Emergence of Avoidant coping as an independent second-order factor warrants further investigation of its potential unique role in adverse outcomes. CORRESPONDING AUTHOR: C. Nahlén Bose, Karolinska Institutet, Sweden, [email protected]

P42 DYADIC COPING AND PARTNERSHIP QUALITY IN CHRONIC OBSTRUCTIVE PULMONARY DISEASE I. V. Vaske1, D. Keil1, W. Rief1, K. Kenn2, N. Stenzel3 1 Philipps-University of Marburg, MARBURG, Germany 2 Schoen Clinic Berchtesgadener Land, SCHOENAU AM KOENIGSSEE, Germany 3 University of Leipzig, LEIPZIG, Germany Introduction: Dealing with a chronic and unpredictable stressor such as chronic obstructive pulmonary disease (COPD) challenges coping resources on individual and on partnership level. Dyadic coping has shown to be relevant for partnership quality in healthy couples. The purpose of this study is to determine the influence of dyadic coping on partnership quality in COPD. Method: The sample consisted of N = 71 patients with COPD. In accordance with the Global Initiative for Chronic Obstructive Lung Disease GOLD, the disease was classified as mild (n = 1), moderate (n = 10), severe (n = 30), and very severe (n = 30). Questionnaires assessing dyadic coping and partnership quality were administered online. Gender differences in partnership quality were tested by analysis of variance. Regression analyses for partnership quality were conducted, with GOLD stage, age, and dyadic coping scales as predictors. Results: There were no gender differences in partnership quality (F(1;48) = .18; p = .68). Partnership quality was predicted by patients’ stress communication (beta = -.30; p ≤ .05), patients’ delegated dyadic coping (beta = -.37; p ≤ .05), and supportive dyadic coping of partners (beta = .46; p ≤ .05). GOLD stage and age explained no significant variance in partnership quality. Overall, the model explained 75.9% of variance (F(11;38) = 10.86; p ≤ .001). Conclusions: Dyadic coping seems to be important with regard to partnership quality in people suffering from COPD. In order to maintain partnership quality despite the progressive course of COPD, appropriate dyadic coping strategies should be promoted. CORRESPONDING AUTHOR: I. V. Vaske, Philipps-University of Marburg, Germany, [email protected]

P43 FATIGUE RELATED PSYCHOLOGICAL FACTORS IN PATIENTS WITH CARDIOVASCULAR DISEASES A.N. Nagy1, E. Szabados2, A. Simon3, B. Mezey2, B. Sándor2, Á. Csathó1 1 University of Pecs, Medical School, PECS, Hungary 2 1st Department of Internal medicine, PÉCS, Hungary 3 State Hospital for Heart Diseases, BALATONFÜRED, Hungary Introduction.The positive effects of exercise training in cardiovascular diseases are well known and used widely in cardiac rehabilitation.

S18

Evidences indicate that exercise improves physical activity and overall cardiorespiratory fitness. However, less is known about the relations of exercise training to psychological factors, like vital exhaustion and chronic fatigue as highly predictive risk-factors of myocardial infarction and sudden cardiac death. The pathophysiological mechanisms are not yet well understood but fatigue might be partly related to disrupted cardiorespiratory values. Methods. Patients with ischemic cardiovascular diseases entered the study (n=227; mean age=63.30±9.89). All patients underwent both standard ergometric and psychological test procedures. In addition, 72 of these patients (mean age= 64.88±5.59) entered an exercise-based cardiac rehabilitation program and were followed for three months. Results and Conclusions. Our results indicate the positive effects of rehabilitation on both cardiovascular and psychological fitness. In addition, we found strong associations between the ergometric and psychological factors. More specifically, the fatigue related psychological parameters were found to show significant associations with maximum systolic blood pressure during exercise, metabolic equivalent and maximum heart rate. CORRESPONDING AUTHOR: A.N. Nagy University of Pecs, Medical School, Hungary, [email protected]

P44 INTERVENTION AGAINST LIFESTYLE AND METABOLIC FACTORS TO ACHIEVE HEALTHY AGING IN CHINA: A LIFE-COURSE PERSPECTIVE Qiu1, Y. Liang2, Z. Yan3 1 Karolinska Institutet, STOCKHOLM, Sweden 2 Jining Medical University, JINING, SHANDONG, China 3 Jining First People's Hospital, JINING, SHANDONG, China TEST. Life expectancy at birth in China has increased dramatically, from less than 40 years in the early 1960s, ~63 years in the 1970s, and ~72 years in the 2000s to ~75 years in the early 2010s. It is projected that the pace of further increase in life expectancy will largely depend on the extent to which the major chronic non-communicable diseases such as coronary heart disease, stroke, and cancers can be effectively prevented or postponed; it is well known that unhealthy lifestyles and metabolic risk factors play a critical role in determining the risk and epidemic of these diseases. In addition, in the last decade evidence from epidemiological research has been accumulating that lifestyle and metabolic risk factors in middle age and late life could significantly contribute to the development and progression of functional deterioration (e.g., cognitive impairment, dementia, and physical disability) later in life. In recent years, epidemiological studies have shown that, along with miracle economic and societal transition and rapid urbanization in China, major unhealthy lifestyle behaviors and metabolic risk factors have become increasingly common among young, middle-aged, and older people, whereas these factors have been poorly controlled for. Thus, as population aging accelerates in the next a couple of decades in China, aging-related disorders will pose a serious threat not only to health care service system, but also to the sustainable development of the nation’s economy and society, unless proper preventative intervention programs are implemented. Fortunately, unhealthy lifestyle behaviors (e.g., smoking, sedentary lifestyle, and excessive alcohol consumption) and metabolic risk factors (e.g., obesity, hypertension, diabetes, and dyslipidemia) represent the major targets for intervention to achieve physical and functional health in aging. Therefore, the community intervention programs to promote healthy lifestyle behaviors and to effectively control of major metabolic risk factors among young adults and middle-aged people are likely to help to achieve healthy aging by improving and maintaining the physical health and functional status in late life, i.e., living a longer and healthier life. CORRESPONDING AUTHOR: Dr. Qiu, Karolinska Institutet, Sweden, [email protected]

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

P45 TYPE D PERSONALITY, PSYCHOLOGICAL ADAPTATION STRA TEGIES AND CHRONIC FATIGUE IN PATIENTS WITH CHRONIC HEART DISEASE I. T. Tiringer1, A. S. attila Simon2, N. F. nora Faubl1, A. Cs. arpad Csatho1 1 Pecs University Medical Faculty, PECS, Hungary 2 State Hospital for Cardiology, BALATONFÜRED, Hungary Introduction: Type-D personality is an independent prognostic factor in coronary artery disease, and it has been found to be associated with increased vital exhaustion. This finding suggests that Type-D individuals generally experience increased fatigability. Chronic fatigue is known to have a negative consequence on patients’ quality of life. Thus, it is a clinically relevant question whether chronic fatigue is more frequent in patients with Type-D personality. Method: 182 patients with chronic heart disease participated in a crosssectional study. Chronic fatigue was assessed by the Multiple Fatigue Inventory and the Fatigue Impact Scale. These self-report instruments measure different dimensions of fatigue (e.g. physical and mental dimensions). Type-D personality was measured by the DS14. In addition, the following questionnaires were also applied: Vital Exhaustion, Perceived Stress, Beck Depression Inventory, SF-36, as well as the TenFlex-30 (Tenacious Goal Pursuit and Flexible Goal Adjustment). Results: Twenty-one percent of the patients were identified as Type-D. The Type-D and non-Type-D patients didn’t differ significantly in objective cardiac parameters (LVEF, MET value), in the summary measure of physical health (SF-36), as well as in the level of perceived stress. In contrast, Type-D personality was found to be significantly associated with increased fatigue in each dimension, and with a higher level of depression. In addition, both adaptation strategies measured by TenFlex-30 had a significantly lower level in Type-D patients. Conclusions: Type-D personality might significantly contribute to an increased fatigue level in patients with chronic heart disease. The possible mediating role of adaptation strategies needs clarification in further studies. CORRESPONDING AUTHOR: I. T. Tiringer, Pecs University Medical Faculty, Hungary, [email protected]

P46 SOMATIC-AFFECTIVE AND COGNITIVE-AFFECTIVE SYMPTOMS OF DEPRESSION ARE NOT ASSOCIATED WITH MORTALITY IN PATIENTS WITH COPD J.N. de Voogd1, J.B. Wempe2, G. Maters3, R. Sanderman4 1 UMCG, HAREN, Nederland 2 Pulmonary Medicine, UMCG, GRONINGEN, Nederland 3 University of Groningen, UMCG, GRONINGEN, Nederland 4 Health Psychology, Medical sciences, University of Groningen, UMCG, GRONINGEN, Nederland Introduction: Depressive symptoms are prevalent in patients with COPD and these symptoms have been related to long-term prognosis. Whether different symptom dimensions of depression are related to prognosis is unclear. We examined the relationship between somatic-affective and cognitive-affective symptoms dimensions of depression in patients with mild to severe COPD and all-cause mortality. Method: Patients (n=323), 58 (11) years, 52 % female, FEV1 %predicted 42 (18), were recruited at the department of pulmonary rehabilitation, Center for Rehabilitation, UMCG, the Netherlands between 2011 and 2013. Measurements included demographic data, pulmonary function, functional status and the Beck Depression Inventory-II (BDI-II). Endpoint was all-cause mortality. Results: 27 patients died. Neither of the two components; the somaticaffective (OR: 0.97; 95% CI: 0.90-1.03)or cognitive-affective (OR: 0.97; 95% CI: 0.86-1.09)symptom dimension were related to mortality . Multivariate showed a significant association for age and FEV1 %predicted.

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

Conclusion: In this study somatic-affective and cognitive-affective symptoms of depression are not associated with mortality. This study did not replicate previous findings that depressive symptoms were associated with prognosis in patients with COPD. CORRESPONDING AUTHOR: J.N. de Voogd, UMCG, Nederland, [email protected]

P47 COMPARISON OF HEALTH RELATED QUALITY OF LIFE IN CARDIAC PATIENTS UNDERGOING INVASIVE TECHNIQUES C.F. Figueroa, O.G. Ornelas Grisel, G.A. Gonzalez Anahi, R.B. Ramos Bertha, R.V. Rocha Victor National Autonomous University of Mexico, MEXICO CITY, Mexico Introduction: A disturbance in heart rhythms can lead to different treatments, including invasive techniques, radiofrequency ablation (G1) or the use of cardiac pacing devices (G2). These techniques are used to maintain the patient's life, but at the same time can have a negative impact on their quality of life. In order to make these techniques function well, some changes are necessary in lifestyle (diet, physical activity, adherence to treatment and medical examinations). Unfortunately, some patients continue to report physical and emotional symptoms. Method: A Health-Related Quality of Life (HRQoL) questionnaire was given to the 17 patients adscribed to the Arrhythmia Unit of the Laboratory of Experimental and Clinical Electrophysiology in the General Hospital of Mexico. Results: Compared to G2, G1 reported severe discomfort related to the symptoms caused by heart disease, reported a regular health as well as a greater deterioration in HRQoL especially in the areas of self-efficacy. G1 also reported less discomfort related to changes in life style compared to G2. Conclusion: The HRQoL of cardiac patients is diminished after an invasive technique; however, cardiac ablation patients reported a better health related quality of life. CORRESPONDING AUTHOR: Dr. CF Figueroa, National Autonomous University of Mexico, Mexico, [email protected]

P48 IMPLANTABLE CARDIOVERTER DEFIBRILLATOR PATIENTS: RELATIONSHIP BETWEEN POST-TRAUMATIC STRESS AND SELF-EFFICACY C.F. Figueroa, M.M. Morales Melina, M.R. Mora Rocio, R.B. Ramos Bertha, R.V. Rocha Victor National Autonomous University of Mexico, MEXICO CITY, Mexico Introduction: Patients with Implantable Cardioverter Defibrillators (ICD) who have experienced shock therapy may develop psychological implications. Post-traumatic Stress (PTS) is a delay response to an event or situation. Perceived self efficacy of patients ICD can be affected by PTS, causing the perception of self-efficacy to their own abilities to face it is diminished. Method: The sample consisted of 27 mexican ICD patients with a mean age of 57.11 in the range of 27 to 82 years old, patients with shock therapy. Instruments used were adapted the Scale Stressor Impact and the questionnaire-adapted specific Health Related Quality of Life. Results: Only 18% of the ICD patients have post-traumatic stress, the self-efficacy mean was 74.07 (range 0 - 100). The Spearman test was performed to determine the correlation between PTS and self-efficacy, resulting in -.316 (sig. .044). Conclusion: The results endorse the hypothesis that less PTS means greater self-efficacy. This study is a pioneer in self-efficacy assessessment of a situation of PTS in ICD patients. The results of this research may

S19

have important implications for the design and implementation of selfefficacy interventions in order to improve the quality of life of ICD patients. CORRESPONDING AUTHOR: Dr. C.F. Figueroa National Autonomous University of Mexico, Mexico, [email protected]

P49 PREDICTORS OF ALL-CAUSE MORTALITY IN PATIENTS WITH STABLE COPD: MEDICAL COMORBID CONDITIONS OR DEPRESSIVE SYMPTOMS G.A. Maters, J.N. Voogd de, R. Sanderman, J.B. Wempe UMCG, GRONINGEN, Nederland Introduction: Comorbid conditions are frequently found in patients with COPD. We evaluate the association of comorbidities with mortality, in stable COPD. Methods: 224 patients, mean age 61.2 (±10.00), 48.2% female, mean FEV1 1.1 (±0.5) liters, median follow-up time 4.2 years, participated. Medical comorbidities were scored according to the Charlson Comorbidity Index (CCI). Depressive symptoms were assessed with the Hospital Anxiety and Depression Scale (HADS) and Symptom Checklist-90 (SCL-90). The Cox proportional hazard model was used for survival analyses. Results: In our sample, 70% of all patients have a comorbid medical condition or high depressive symptoms. During follow-up 51% of all patients died, and those with heart failure have the highest mortality rate (75%). Age, fat-free mass and exercise capacity were predictive factors, contrary to CCI-scores and high depressive symptoms. An unadjusted association between heart failure and survival was found. Conclusion: Although the presence of comorbidities, using the CCIscore, is not related to survival, heart failure seems to have a detrimental effect on survival. Higher age and lower exercise capacity or fat-free mass predict mortality. CORRESPONDING AUTHOR: G.A. Maters, UMCG Nederland, [email protected]

P50 TYPE D PERSONALITY: IN DECLINE OR THRIVING? J.C. Coyne, J.N. De Voorgd UMCG, PHILADELPHIA, United States of America Introduction: After an unbroken series of studies demonstrating a Type D Personality/Mortality link, there had been a series of null findings, both within the original investigative group and by independent investigators. Moreover, there is a decline in serious conceptual, statistical, and methodological issues that have been raised in describing Type D Personality. However, the original investigative group has been responding to criticism and now claims the link between Type D Personality that is replicated in both the standardized scoring and interaction effects in multiple regression of continuous variables proposed by critics. Method: Updated systematic review and reanalysis of published data. Results: One new study of Type D Personality and mortality was found. Published results indicated a significant prediction of mortality from the standardized scoring of the DS14, as well as multiple regressions involving continuous NA and SI variables and their interaction. However, graphing of the results with standardized scoring reveals an anomalous result for the low/low NA/SI quadrant versus the high/high, with the two quadrants both being higher than the quadrants high in either NA or SI, but not both. Furthermore, graphing of the interaction effect with continuous variables does not fit the predictions of the original Type D Personality model.

S20

Conclusion: There is not yet a demonstration that analyses involving standardized scoring and continuous multiple regression converge to show a link between Type D personality and mortality. The criticism stands that results are not comparable in the two ways of analyzing DS14 data. CORRESPONDING AUTHOR: Professor J.C. Coyne, UMCG, United States of America, [email protected]

P51 ADOLESCENT NECK AND SHOULDER PAIN - THE ASSOCIATION WITH DEPRESSION, PHYSICAL ACTIVITY AND SCREEN-BASED ACTIVITIES S.M. Myrtveit 1 , Børge Sivertsen 2 , Jens Chri Skogen 2 , Lisbeth Frostholm3, Kjell Mor Stormark4, Mari Hysing4 1 Department of Clinical Science, BERGEN, Norway 2 Division of Mental Health, Norwegian Institute of Public Health, BERGEN, Norway 3 The Research Clinic for Functional Disorders, Aarhus University Hospital, AARHUS, Denmark 4 Regional Centre for Child and Youth Mental Health and Child Welfare, Uni Health,, BERGEN, Norway Introduction: Neck and shoulder pain is frequent in adolescents and multiple factors seem to affect the risk of such symptoms. We aimed to investigate the prevalence of neck/shoulder pain in Norwegian adolescence, and to examine whether behavioral and emotional factors were associated with the risk of neck/shoulder pain. Finally we aimed to investigate whether neck/shoulder pain was related to use of health services. Methods: Data from the population based study ung@hordaland was used. Participants were asked how often during the last six months they had experienced neck/shoulder pain. The association between frequent neck/shoulder pain and physical activity, symptoms of depression and screen-based activities was evaluated using logistic regression analyses stratified by gender. The relative risk of visiting health services when reporting neck/shoulder pain was calculated using multiple logistic regression analyses. Results: Frequent neck/shoulder pain was reported by 20.0% (1,797 of the total 8,990) and more often by girls than boys (p<0.001). A high score of depressive symptoms was the strongest risk factor for neck/shoulder pain in both boys and girls (OR=6.14 (95%CI: 4.48-8.42) and OR=3.10 (95%CI: 2.63-3.67) respectively). Frequent screen-based activities slightly increased the risk while physical activity was protective. Individuals reporting neck/shoulder pain more often visited their GP (47.1% vs. 31.8%) and school health services (24.6% vs. 13.5%). Conclusion: Frequent neck/shoulder pain was reported in 20% of Norwegian adolescents. Symptoms of depression and screen-based activities increased the risk of neck/shoulder pain while physical activity was protective. Individuals reporting neck/shoulder pain visited health services more frequently than others.

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

Method: 817 children participated in the study. Children’s ages ranged 9 to 15(boys 423 people, girls 394 people). In the study, the authors examined the reliability of the correlations within the CSR by using Cronbach’s alpha coefficient. The authors also looked at the validity of the CSR by studying the correlations between the CSR and PSI. Result: 817 children participated in the study of those 809 samples were acceptable for analyzing. In the study, the authors conducted a factor analysis. In doing so, the found three factors. The first was named “Helplessness”. The second was named “Irritation”. The third was named “DepressionPhysical response”. The Cronbach’s alpha for the first factor was 0.83, the second factor was 0.87, and the third factor was 0.74. When the domains were compared with established outcome measures, the correlations were extremely to strong (0.51 to 0.85). There is therefore a high degree of reliability and validity for the PSI, which means that the CSR has a high degree of clinical usability. Discussion: As to future research, it is assumed that by using the CSR, the stress of children can be simply and appropriately assessed. Thus, leading to rapid treatment and addressing the most important issues. Further, it was suggested that the authors make use of this study of the CSR by formulating a stress management program for children. CORRESPONDING AUTHOR: RM Matsuo, Okinawa University, Japan, [email protected]

P53 DEVELOPMENT OF THE STRESS MANAGEMENT PROGRAM FOR CHILDREN (2) -SCHOOL-BASED STRESS MANAGEMENT PROGRAM FOR CHILDRENS.T. Takeda1, Risa Matsuo2, Maki Ota1 1 Tottori University Graduate School of Medical Sciences, YONAGO, Japan 2 Okinawa University Department of Child Studies, NAHA, Japan

P52 DEVELOPMENT OF THE STRESS MANAGEMENT PROGRAM FOR CHILDREN (1) R.M. Matsuo1, M. Ota2, S. Takeda2 1 Okinawa University, OKINAWA,, Japan 2 Tottori University, TOTTORI, Japan

The present study evaluated the effects of school-based universal program for stress management for elementary school students. The program consisted of teaching students about cognitive distortions and training them by cognitive restructuring. The participating children (aged 11-12 years) were assigned either to an intervention group (n=234) or to a control group(n=43). The program involved one 45-minutes session during school classes. In this program, slides are used as teacher can conduct their stress management class to their children so that they can carry out their class with equal quality. Cognitive distortions were characterized so that children can understand it easily. In order to assess the effects of the program, students were asked to complete the Children's Stress Response (Matsuo et al., 2014) and five questions about self-efficacy of cognitive restructuring before and after the program was implemented. Two factors ANOVA, with groups and phases as independent variables, and stress responses and self-efficacy of cognitive restructuring as a dependent variable, was conducted. The results were as follows: (a) stress responses decreased (P < 0.001), (b) self-efficacy in the awareness of their own feelings and thinking improved (P < 0.001) , (c) understanding of the cognitive model was prompted, (d) self-efficacy to review their thinking improved when they feel uncomfortable (P < 0.001), and (e) self-efficacy to change their negative thinking to positive thinking improved (P < 0.001) in the intervention group. These results suggest that the program was useful for reducing stress responses and improving self-efficacy in cognitive restructuring of children.

Introduction: The aim of this study examined the reliability and validity of the CSR test.

CORRESPONDING AUTHOR: Ph.D ST Takeda, Tottori University Graduate School of Medical Sciences, Japan, [email protected]

CORRESPONDING AUTHOR: S.M. Myrtveit, Department of Clinical Science, Norway, [email protected]

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

S21

P54 A MULTIVARIATE LONGITUDINAL STUDY OF GENDER DIFFERENCES IN U.S. CHILDREN'S INTERNALIZING AND EXTERNALIZING PROBLEMS G. Z. Zeng1, P. Fu2 1 Texas A&M University-Corpus Christi, CORPUS CHRISTI, United States of America 2 Case Western Reserve University, CLEVELAND, United States of America

inexperienced participants, attitudes and the gender-by-perceived risk interaction term represented significant influences. Conclusions: The results suggest that interventions to improve condom use intentions and behaviour amongst adolescents should consider the differential impact of gender and sexual experience.

Introduction: This study investigates gender differences in the developmental trajectories of internalizing and externalizing problems among children in the United States from kindergarten to fifth grade, while controlling for other sociodemographic factors that may be attributed to such differential trajectories. Presently, there is a paucity of research on the development of internalizing and externalizing problems among young children on a national scale. Methods: Data on a U.S. nationally representative sample of 9,796 kindergarteners from the Early Childhood Longitudinal Study were analyzed using bivariate mixed-effects models to simultaneously model the trajectories of both internalizing and externalizing problems. Results: Gender differences are found in the developmental trajectories of both internalizing and externalizing problems as evidenced by their differential slopes. Specifically, boys displayed more symptoms of both internalizing and externalizing problems than girls at school entry, and that such elevated levels of problems persisted into fifth grade. Higher socioeconomic status and parental report of child’s ability to pay attention are found to be protective factors for both internalizing and externalizing problems. Conclusion: This study fills a major gap in the literature and will contribute to the future research on the etiological factors of these behavioral problems of children. Additionally, these findings are informative for school-based early intervention efforts.

P56 UNUSUAL BEHAVIOR OF AUTISTIC CHILDREN IN BANGLADESH S.M.S. Kabir University of Chittagong, CHITTAGONG, Bangladesh

CORRESPONDING AUTHOR: Dr. G. Z. Zeng, Texas A&M UniversityCorpus Christi, United States of America, [email protected]

CORRESPONDING AUTHOR: Dr AR Rich, City University / Atlantis Healthcare, Australia, [email protected]

Introduction: The aim of the present study was to assess unusual behavior of autistic children and to explore the degree of autism present in children in Bangladesh. Method: 230 autistic children aged from 3 to 16 years old were selected by purposive sampling technique from eighteen different autistic and special schools of four different cities (Dhaka, Rajshashi, Khulna and Chittagong). The Autistic Diagnostic Check List (Banerjee, 2007) was used as instrument. There were 60 items which were categorized in six sub-groups, namely general observation, cognition, emotion, social, communication, sensory deficiency. Results: It was shown that Cronbach Alpha of the total scale was .92, and that the internal consistency was excellent. Validation of the scale was assured by convergent validity. Among the 230 children, 156 autistic children were in normal range and 74 children were in severe range. Among them 130 were boys and 100 were girls and their mean scores for six sub-groups did not differ significantly. The study showed that mean score was highest in general observation (38.13) and lowest in emotion (23.54). Conclusion: The six sub-groups are uniformly important to the understanding of autism symptoms. CORRESPONDING AUTHOR: Mr. S.M.S. Kabir, University of Chittagong, Bangladesh, [email protected]

P55 ROLE OF GENDER, SEXUAL EXPERIENCE AND THE THEORY OF PLANNED BEHAVIOUR IN PREDICTING CONDOM INTENTIONS A.R. Rich1, B.A. Mullan2, K. Sainsbury3, A.R. Kuczmierczyk4 1 City University / Atlantis Healthcare, SYDNEY, Australia 2 Curtin University, PERTH, Australia 3 The University of Sydney, SYDNEY, Australia 4 Hertfordshire Partnership Foundation NHS University Trust, HERTFORDSHIRE, United Kingdom Introduction: The aim of this study was to examine how the prediction of condom-related cognitions, intentions, and behaviour amongst adolescents may differ according to gender and sexual experience within a theory of planned behaviour (TPB) framework. Methods: Adolescents (n = 306) completed questionnaire measures of sexual experience, condom use, TPB variables, perceived risk, and safe sex knowledge. Results: Significant differences in TPB variables, perceived risk, and knowledge were found such that males and sexually experienced participants were generally less positive about condom use. Twenty percent of the variance in attitudes was accounted for by four variables such that female gender, no previous sexual experience, better safe sex knowledge, and greater risk perceptions were associated with more positive attitudes. The prediction of intentions separately amongst sexually experienced (R2 = .468) and sexually inexperienced (R2 = .436) participants revealed that for the sexually experienced participants, attitudes and subjective norms were the most important considerations. In contrast, amongst the

P57 INTELLIGENCE STRUCTURE AND ATTENTION PROPERTIES IN CHILDREN WITH ATTENTION DEFICIT HYPERACTIVITY DISORDER S.W. Wang, Y. Yang, L. Ma The Third Affiliated Hospital of Soochow, CHANGZHOU, China To investigate the intelligence structure and attention properties and to explore the relationship between them in children with attention deficit hyperactivity disorder of the predominantly inattention type (ADHD-I), predominantly hyperactive-impulsive type (ADHD-HI) and combined type (ADHD-C). Children with ADHD-I (N=24), ADHD-HI(N=12) and ADHD-C (N=54) aged between 7-10 years completed Chinese Revised Wechsler Intelligence Scale for Chilren (C-WISC) and integrated visual and auditory continuous performance test (IVA-CPT). Comparison of cognitive profiles of C-WISC scores between subtypes of ADHD was performed. Additionally, the relationship between attention/response control quotients and cognitive profiles was investigated. (1) Intelligence structure: Verbal intelligence (VIQ) of ADHD-HI (110.08±10.64) was higher than that of ADHD-C (101.13±13.20) and ADHD-I (94.71±11.11). Full scale intelligence quotient (FIQ) of ADHD-HI (104.33±9.63) was higher than that of ADHD-I (94.38±10.48). (2) Attention properties: Full scale attention quotient of ADHD-C (90.26±11.67) was lower than that of ADHD-I (98.17±18.03) and ADHD-HI (99.25±15.58). Auditory attention quotient of ADHD-C (89.94±14.16) was lower than that of ADHD-I (99.00±18.66). (3) Relationship between intelligence and

S22

attention: The Pearson correlations of the auditory attention quotient and auditory control quotient with the PIQ were 0.24 and 0.29. The correlations of the visual attention quotient with the PIQ and FIQ were 0.21 and 0.25. The correlations of the full scale attention quotient with PIQ and FIQ were 0.27 and 0.24. Each of these correlations reaches the level of significance at the 5% level. Cognitive profile and attention properties analysis reveal children with different subtypes of ADHD have distinct weakness in intelligence structure. CORRESPONDING AUTHOR: Dr. S.W. Wang, The Third Affiliated Hospital of Soochow, China, [email protected]

P58 THE MENTAL HEALTH OF ADOLESCENTS: WHAT DIFFERENCES EXIST? J. Dray1, M. Freund2, J. Bowman3, E. Campbell2, R. Hodder2, J. Wiggers2, K. Gillham4 1 Hunter New England Population Health/University of Newcastle, WALLSEND, Australia 2 Hunter New England Population Health;The University of Newcastle;HMRI, NEWCASTLE, Australia 3 Hunter New England Population Health; University of Newcastle, WALLSEND, Australia 4 Hunter New England Population Health; HMRI, NEWCASTLE, Australia Introduction: Within Australia, the mental health of adolescents is a major area of health concern. However little recent data describes the mental health of Australian adolescents and how it varies across subgroups of that population. Method: Web-based surveys were undertaken with 7935 adolescents across 32 disadvantaged secondary schools in one New South Wales local health district, Australia. The survey included the student Strengths and Difficulties Questionnaire as well as student demographic items (age, gender, Aboriginal and/Torres Strait Islander status, other cultural background and postcode). Level of socio-economic disadvantage was determined by the SEIFA index using student postcodes. Results: The presentation will report the level of risk of mental health problems of the student sample, and examine differences in risk of mental health problems by gender, Aboriginal or Torres-Strait Islander status, other cultural background, age and socioeconomic disadvantage. A significant difference in risk of mental health problems was found when examining the factors of Aboriginal and/or Torres Strait Islander status (X2(2, N=7935 = 48.59, p<.0001) and gender (X2(2, N=7935) = 38.38, p<.0001). Conclusions: The mental health of young people is linked to many short and long term health outcomes. The findings of this research will add significantly to the understanding of mental health and young people in Australia and has the potential to inform targeted interventions for adolescent subgroups to increase positive mental health and reduce inequity amongst adolescent subgroups.

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

particularly making friends and instant chatting via the Internet, implicates adverse psychosocial concerns such as lower self-esteem and increased levels of psychological distress. However, existing research has neglected the effect of timing of online communication on psychosocial development. The present study thus builds upon Belsky’s evolutionary perspective to address timing of online communication, and examine influence of early age involved at online communication on trajectories of psychological distress in family contexts. We further posit an interaction of timing of online communication x adolescent self-esteem which influences psychological distress and may well vary across family contexts. Methods. Data are from longitudinal surveys of Taiwanese adolescents at age 16 with several follow-ups (n=1,715). Psychological distress was measured by 12-item Symptom Checklist. Results. Multivariate analyses from growth curve models indicate that increased levels of psychological distress in late adolescence are significantly associated with early age at online communication (β=2.68, p<0.05 for online communication at 15 or younger). And, individuals with early age at online communication and lower adolescent self-esteem are at a particularly increased risk for psychological distress in late adolescence. In contrast to changes in psychological distress, the effect of early age at online communication is diminished. Conclusion. Results suggest joint effects of timing of online communication and adolescent self-esteem in the development of psychological distress. Implications of our findings aimed at promoting psychological well-being are discussed. CORRESPONDING AUTHOR: Prof. C. Chiao, National Yang-Ming University, Taiwan, [email protected]

P60 SUICIDAL IDEATION AND ITS CORRELATES AMONG JUVENILE DELINQUENTS IN SOUTH KOREA S. Kim1, H. Kim1, D. C. Seo2, S. D. Kim1, H. Cho3 1 Korea Association of Heath Promotion, SEOUL, South-Korea 2 Ewha Womans University, SEOUL, South-Korea 3 Korea Association of Health Promotion/Seoul National University, SEOUL, South-Korea

P59 EARLY AGE AT ONLINE COMMUNICATION INTERACTS WITH ADOLESCENT SELF-ESTEEM TO INFLUENCE PSYCHOLOGICAL DISTRESS IN ADOLESCENTS C. Chiao, Yunyu Chen National Yang-Ming University, TAIPEI, Taiwan

Introduction: Suicide is the first leading cause of death among Korean adolescents and young adults, which becomes a major social issue in South Korea. Juvenile delinquents have been pointed out as a high-risk group for suicidal thoughts and attempts but there is a paucity of data on their prevalence and correlates. Purpose / Methods: This study investigated suicidal ideation and its correlates among juvenile delinquents in Korea. Suicidal ideation, psychological health status and health-related behaviors were assessed using a self-administered questionnaire in 1,682 juvenile offenders aged between 15 and 18 years in 2012. Results: The prevalence of suicidal ideation in juvenile delinquents was 15%. Girls were more likely to report suicidal thoughts than boys (30.3% vs. 12.7%). Suicidal ideation was more common among adolescents who were not living with their parents before entering detention centers (22.6% vs. 13.2%) than their counterparts. The likelihood of suicidal ideation was significantly associated with unhappiness (OR = 6.76), high stress perception (OR = 5.36), poor perceived health status (OR = 3.09), drug use (OR = 2.6) and physical inactivity (OR = 1.67). Conclusions: The present study provides evidence for an association between suicidal ideation and psychological health and health risk behaviors among juvenile delinquents. It also highlights the importance of mental health and behavioral interventions for this population to prevent suicidality.

Introduction. Global literature documented prevalent online communication in the past decade. Studies demonstrated that online communication,

CORRESPONDING AUTHOR: S Kim, Korea Association of Heath Promotion, South-Korea, [email protected]

CORRESPONDING AUTHOR: J. Dray, Hunter New England Population Health/University of Newcastle, Australia, [email protected]

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

P61 'ADULTS DON'T LIKE CHILDREN TO PLAY OUTSIDE'. SOCIAL DETERMINANTS OF LEISURE IN CHILDREN WITH DISABILITIES N. Kolehmainen1, M. Bult2, C. Missiuna3, M. Ketelaar4 1 Newcastle University, NEWCASTLE, United Kingdom 2 University of Utrecht, UTRECHT, Nederland 3 McMaster University, HAMILTON, Canada 4 University Medical Center Utrecht De Hoogstraat Rehabilitation, UTRECHT, Nederland Introduction: Participation in leisure, including physical play, is important for all children’s development and health. We identified perceived determinants of participation in leisure, especially physical play, in children with motor impairments. Methods: Two parallel studies. Study 1: Systematic reviewing methods to identify, select and appraise qualitative evidence on parents’ views about the determinants. Framework synthesis was used, with the Theoretical Domains Framework (TDF) as the initial framework. Study 2: Children’s (6-8 years) views on physical play participation were explored using semi-structured, child-friendly interviews, audio recorded and analysed for emerging themes. Results: Study 1: 16 papers (of 2333 identified) reported parents’ views. Of the determinants discussed by parents, 56.4% (356/631) related to behaviours, cognitions and emotions of those around the child (e.g. parents worrying about the child, peers ostracising the child, and leisure instructors not knowing how to include the child). Study 2: The children’s (n=17) narratives also emphasised the social context. Children expressed strong perceptions that adults and older siblings regulate their physical play. A particularly strong theme emerged of adults discouraging children’s outdoor play due to adults not liking the consequences of such play (e.g. children getting wet/dirty, adults being hit by snow balls). Conclusion: The evidence indicated that, similarly to non-disabled children, social factors are potentially important determinants of leisure participation in children with disabilities. Future work will seek ways to: (i) integrate behaviour change strategies to leisure interventions for children with disabilities, and (ii) extend generic leisure interventions to include children with disabilities. CORRESPONDING AUTHOR: N. Kolehmainen, Newcastle University, United Kingdom, [email protected]

P62 EXPLORING SALIVARY CORTISOL AND RECURRENT PAIN IN MID-ADOLESCENTS LIVING IN TWO HOMES E. Fransson, L. Folkesson., M. Bergström, V. Östberg, P. Lindfors CHESS, STOCKHOLM, Sweden Introduction: Every year, around 50.000 children in Sweden experience a separation between their parents. Joint physical custody (JPC), where the child alternates homes between the parents for about equal amount of time, has become a common living arrangement after parental separation. Children living in two homes can benefit from everyday contact with both parents and access to both parents’ financial resources. However, children can also experience stress from constantly moving and from exposure to any parental conflict. Yet, research on JPC and stress-related biological functioning is limited. The aim of this study was to investigate how living arrangements (intact family/JPC) relate to HPA-axis activity and recurrent pain in mid-adolescents. Methods: Mid-adolescents (106 girls and 51 boys) provided demographic details, self-reports of recurrent pain (headache, stomachache, neck/shoulder and back pain) and salivary samples. Salivary cortisol samples were collected: 1) immediately at awakening, 2) +30 minutes, 3) +60 minutes, and 4) at 8 p.m.

S23

Results: Hierarchical regressions showed that living arrangements did not predict morning cortisol levels, the diurnal cortisol rhythm or? recurrent pain. However, sex was significantly associated with both morning cortisol and recurrent pain. Conclusion: Living arrangements were not linked to HPA-axis activity or recurrent pain in this group of well-functioning mid-adolescents. Although this is the first study investigating how living arrangements relate to HPA-axis functioning, which means that additional research is needed, the findings suggest that these mid-adolescents have adapted to their living arrangements and that other factors seem more pertinent for HPA-functioning and subjective health complaints. CORRESPONDING AUTHOR: Dr L.S. Fransson, CHESS, Sweden, [email protected]

P63 A LONGITUDINAL CASE-CONTROL STUDY ON GOAL ADJUSTMENT IN ADOLESCENTS WITH CANCER E. Sulkers, M. Janse, A. Brinksma, P.F. Roodbol, W.A. Kamps, W.E. Tissing, R. Sanderman, J. Fleer University Medical Center Groningen, GRONINGEN, Nederland Introduction: Severe illnesses may disturb the attainment of personal meaningful goals. Adjusting one’s goals to what is possible is an adaptive way to deal with goal disturbance. This study examined whether: (1) the goals of adolescents with cancer (3 months post-diagnosis) differed from those of healthy controls with regard to value-orientation and abstraction level, (2) the value-orientation and abstraction level of the goals of adolescents with cancer changed over time. Method: Thirty-three adolescents with cancer and 66 matched controls completed the Personal Project Analysis Inventory. After nine months the adolescents with cancer completed the measure again. All goals were coded by two independent raters on goal content and abstraction level. Results: Compared to controls, adolescents with cancer showed an intrinsic rather than extrinsic value-orientation. They also reported their goals on a lower level of abstraction than controls. Despite small changes, there were no significant differences in patients’ goals over time. Conclusions: Group differences in value-orientation and goal abstraction indicate that adolescents with cancer use the flexible structure of the goal system to deal with changing circumstances to goal pursuit. The lack of change over time suggests that goal adjustment begins early in the disease trajectory and continues over time. More research with a longer follow-up is needed to determine whether these shifts in value-orientation and goal abstraction level are permanent or not. CORRESPONDING AUTHOR: E Sulkers, University Medical Center Groningen, Nederland, [email protected]

P64 COMBINED USE OF NON-NUTRITIVR SUCKING, SUCROSE, AND TUCKING ATTENUATES INFANTS' BEHAVIORAL CHAN GES DURING HEEL-STICK PROCEDURES Jen-Jiuan Liaw1, Y.T. Yin2 1 National Defense Medical Center, TAIPEI, Taiwan 2 Nursing Department, Tri-Service General Hospital Song Shem Branch, TAIPEI, Taiwan Purpose: The study purpose was to describe the effects of using combined nonpharmacological interventions [(non-nutritive sucking (NNS), oral sucrose, and facilitated tucking (FT)] on the infant behaviors after adjusting infant positioning, non-treatment NNS, infant state and characteristics.

S24

Method: A convenience sample of 110 infants (gestational age 26.4-37 weeks) needing heel sticks were randomly assigned to five combinations of non-pharmacological treatments using Clinstat block randomization: (1) routine care, (2) NNS + FT, (3) sucrose + FT, (4) NNS + sucrose, and (5) NNS + sucrose + FT. The outcome variables were the occurrences of infant disorganized behaviors, grimace and regulatory behaviors. Results: Infants receiving NNS + sucrose + FT experienced less frequent limb and trunk extension or squirming than those receiving routine care . Frequency of grimace in infants receiving NNS + sucrose and sucrose + FT was .65 and .61 times significantly lower, respectively, than in those receiving routine care across all assessment phases. The combined use of NNS, sucrose, and FT helped reduce disorganized behaviors during heel sticks. Conclusions:The combined use of NNS + sucrose + FT effectively reduced the frequency of infants’ limb and trunk extension or squirming that are common disorganized behaviors in infants. Using sucrose + FT and NNS + sucrose effectively reduced the frequency of infant grimace, a highly sensitive pain indicator in infants. Combinations of NNS, oral sucrose, and FT helped reduce preterm infants’ disorganized behaviors during heel stick. Our results guided clinicians to develop atraumatic heel-stick procedures to minimize infants’ pain during procedures. CORRESPONDING AUTHOR: Professor Jen-Jiuan Liaw, National Defense Medical Center, Taiwan, [email protected]

P65 FACTORS RELATED TO PSYCHOSOMATIC COMPLAINTS AND SELF-ESTEEM AMONG JAPANESE JUNIOR HIGH SCHOOL STUDENTS Y.T. Takata University of Tsukuba, TSUKUBA, Japan Purpose: To clarify factors related to psychosomatic complaints and selfesteem among Japanese junior high school students, and the relations among these factors. Method: A questionnaire was administered to 7th- through 9th-graders at X junior high school in April 2012 and April 2013. School nurses asked homeroom teachers to have their classes complete a health questionnaire. Homeroom teachers distributed the questionnaire to their classes, and the students completed. Multiple logistic-regression analysis was conducted with the scale score for psychosomatic complaints as the dependent variable and sex, grade, and other relevant factors as independent variables. Result: Completed questionnaires with usable data were obtained from 557 boys and 592 girls. As a result, girls had lower scores for “selfesteem”as compared to boys (p=.01). Girls had more psychosomatic complaints than boys (p=.066). Multiple logistic-regression analysis adjusted for sex and age with the psychosomatic complaints scale score as the dependent variable revealed that children with high scores for had fewer psychosomatic complaints. In addition, meal of moderate eating, well-balanced meal, those who wake up shapely in the morning, children who went to bed between 8:00 and 11:00 p.m., those who ate vegetables every day, and those with a high Self-esteem score had fewer psychosomatic complaints than other children. Conclusion: The present results indicate that self-esteem and lifestyle are related to psychosomatic complaints among 7th- to 9thgrade junior high school students. These findings suggest that support to help children raise their self-esteem and improve their lifestyle is required. CORRESPONDING AUTHOR: Prof. Y.T. Takata, University of Tsukuba, Japan, [email protected]

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

P66 DO INTERVENTIONS FOR CHILDREN WITH MOTOR IMPAIRMENTS INTEGRATE BEHAVIOURAL THEORY AND METHODS? A SYSTEMATIC REVIEW N. Kolehmainen1, J. McAnuff2, A. Tissen3 1 Newcastle University, NEWCASTLE, United Kingdom 2 Leeds Community Healthcare NHS Trust, LEEDS, United Kingdom 3 Zorginstellingen Peter van Foreest, HAGUE, Nederland Introduction: Interventions for children with motor impairments (e.g. cerebral palsy, developmental co-ordination disorder) have scarce evidence of effectiveness, and there is little consensus about their ‘active ingredients’ or pathways to change. We aimed to identify any patterns in the outcomes, ingredients and pathways to change of these interventions by reviewing the existing literature. Methods: Two electronic databases specialising in two common intervention types (physical and occupational therapy [PT/OT]) were searched using option ‘child/paediatric/adolescent’. Papers published in 2000-2011 were screened for inclusion using the following criteria: the participants were children (0-18yrs) with motor impairments (defined using explicit criteria); the study included intervention(s) delivered by PTs/OTs; AND the primary outcome related to the child. Data were extracted on the study characteristics, primary outcome, intervention descriptions, and hypothesised pathways to change; and summarised descriptively. Two independent reviewers completed all inclusion/exclusion screening and data extraction. Results: 109 full-texts (of 913 papers identified) were included. Of the primary outcomes, 70 were behaviours (e.g. hand writing, gross motor movements), 33 were body functions or structures (e.g. bone density, muscle strength), 5 were diagnosis-related (e.g. torticollis symptoms), and 1 was psychological (self-concept). The interventions were primarily described by general therapy labels (e.g. task oriented upper limb training) rather than specific ingredients, and the hypothesised pathways to change were primarily biomedical/neurological. Conclusion: Interventions for children with motor impairments frequently target behavioural outcomes; there is substantial scope for integrating behaviour change theory and methods into how interventions are described and hypothesised to result in change. CORRESPONDING AUTHOR: N. Kolehmainen, Newcastle University, United Kingdom, [email protected]

P67 MATERNAL SMOKING IN EARLY PREGNANCY AND CHILDREN'S AUTONOMIC NERVOUS SYSTEM ACTIVITY AT AGE 5-6 YEARS T.G.M. Vrijkotte1, M.G.J. Gademan1, S.R. Rooij, de1, T.J. Roseboom1, M.B. Twickler2 1 AMC, AMSTERDAM, Nederland 2 Academic Hospital Antwerpen, ANTWERPEN, Belgium Background: Maternal smoking during pregnancy might have long lasting effects on the cardiovascular system in the offspring, but evidence is inconclusive. We studied the association between maternal smoking during early pregnancy and cardiac autonomic nervous system (ANS) activity in the offspring at young age. Method: Data were collected within the ABCD-study, a multi-ethnic prospective birth-cohort in Amsterdam. Smoking was self-reported. Indicators of ANS activity were: heart rate (HR), pre-ejection-period (PEP; sympathetic activity) and respiratory sinus arrhythmia (RSA; parasympathetic activity), measured in resting supine position. Results were adjusted for maternal and child characteristics. Results: 3102 mother-child pairs were included. 74.1% of the mothers were non-smokers, 16.9% stopped smoking, 6.2% smoked 1-5 cigarettes/ day and 2.8% > 5 cigarettes/day. Maternal smoking was associated with decreased HR (compared to non-smokers: stopped smoking: -0.30 bpm,

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

1-5 cigarettes/day: -1.05 bpm, >5 cigarettes/day: -2.53 bpm (p-value 0.006)) and increased RSA (stopped smoking: 0.17 msec, 1-5 cigarettes/day: 5.52 msec, >5 cigarettes/day: 22.36 msec (p-value 0.008)). Results were not mediated by fetal growth restriction. Smoking was not associated with PEP. Conclusion: Smoking during early pregnancy is associated with increased cardiac parasympathetic activity in according with a decreased heart rate in the offspring at young age. This observation is puzzling as it suggests that intensity of mothers´ smoking behavior is not negatively affecting the offspring's vagotonus; hitherto, an acknowledged risk factor in premature cardiovascular disease and diabetes later in life. Future research regarding effects of high prenatal nicotine exposure on development of brain structures determining the vagal balance is interesting. CORRESPONDING AUTHOR: T.G.M. Vrijkotte, AMC, Nederland, [email protected]

P68 'WE UNDERSTAND BUT WE FORGET IT' DERMATOLOGY HEALTH PROFESSIONALS' UNDERSTANDING AND MANAGEMENT OF PSORIASIS A.C. Chisholm1, P.A. Nelson1, C.J. Pearce1, C. Keyworth1, C.E.M. Griffiths2, L. Cordingley3, C. Bundy1 1 Dermatology Research Centre, University of Manchester, MANCHESTER, United Kingdom 2 Dermatology Centre, Salford Royal NHS Foundation Trust and MAHSC, MANCHESTER, United Kingdom 3 Dermatology Research Centre, Health Psychology,University of Manchester, MANCHESTER, United Kingdom Introduction Psoriasis, a long-term inflammatory skin condition, is associated with significant psychological and physical disability. Illness beliefs or ‘personal models’ underpin psoriasis patients’ mood, self-management, healthcare seeking behaviours, and daily functioning. Little is known about health professionals’ illness beliefs, how such beliefs inform clinical management or how practitioners who manage people with psoriasis conceptualise the condition. This study aimed to identify health professionals’ personal models about psoriasis. Method Semi-structured interviews were conducted with 23 primary and secondary care practitioners working with people with psoriasis. Purposive sampling was used to achieve maximum variation regarding practitioner discipline, level of experience, gender and age. Data analysis was informed by Leventhal’s Illness Representations Model, using principles of Framework Analysis to generate key issues. Results Practitioners often held incongruent personal models about psoriasis; while commonly aware of the condition’s complexity, they described more linear and narrowly focused approaches to managing psoriasis. They described psoriasis as complex and long-term, while simultaneously reporting skin-focused, episodic management of patients. Conflicting beliefs were also evident regarding causes, consequences and control of psoriasis. Positive affect (e.g. job satisfaction/calmness) was described by practitioners holding more elaborate models, but frustration and worry were expressed by those with incongruent models. Conclusions Despite awareness of the long-term, complex nature of psoriasis, practitioners often held conflicting beliefs and reported single-focus management behaviour. Some practitioners however, held more sophisticated models of psoriasis and reported patient-focused rather than skin-focused care. CORRESPONDING AUTHOR: Dr. A.C. Chisholm, Dermatology Research Centre, University of Manchester United Kingdom, [email protected]

S25

P69 THE ADAPTIVITY OF MONITORING AND SENSITIVITY TO BODILY SIGNALS: THE ROLE OF MINDFULNESS AND CATASTROPHIZING N.T. Tsur, K. Ginzburg, R. Defrin Tel-Aviv University, TEL-AVIV, Israel Introduction: Monitoring and sensitivity to bodily signals are often used interchangeably, yet recent findings indicate that they represent different processes. Two studies examined the adaptivity of these constructs, and the mechanisms that underlie it. More specifically, Study 1 examined the relations of monitoring and sensitivity with positive and negative affect. Study 2 examined the intervening roles of catastrophizing and mindfulness in these processes. Method: 217 young adults (Study 1) and 179 undergraduate students (Study 2) completed self-report questionnaires. Linear regression analyses were conducted with positive affect, negative affect, and anxiety, as predicted outcomes. PROCESS procedures were used to examine moderation and mediation effects. Results: Study 1 demonstrated that monitoring was associated with negative affect (β=.19; p<.05), and sensitivity was correlated with positive affect (β=.17; p<.05). Study 2 indicated that catastrophizing mediated the association between monitoring and anxiety (significant indirect effect through catastrophizing: Bootstrap CI=0.03, 0.11), and mindfulness moderated it (β=-.21; p<.01). As indicated by the Johnson-Neyman significance region, in high levels of mindfulness, monitoring significantly predicted lower levels of anxiety, whereas in low levels of mindfulness, monitoring significantly predicted higher levels of anxiety. Further, the association between sensitivity and positive affect was moderated by mindfulness (β=0.18; p<0.05); in higher levels of mindfulness, higher levels of sensitivity significantly predicted higher levels of positive affect. Conclusions: While monitoring of bodily signals is mostly maladaptive, sensitivity to these signals is a salutogenic factor. These processes are conditioned and mediated by the orientation towards subjective experiences. CORRESPONDING AUTHOR: N.T. Tsur, Tel-Aviv University, Israel, [email protected]

P70 THE TEMPORAL ORDER OF CHANGE IN DAILY MINDFULNESS AND AFFECT DURING MINDFULNESS-BASED STRESS REDUCTION E. Snippe1, I. Nyklícek2, M.J. Schroevers1, E.H. Bos1 1 University Medical Center Groningen, GRONINGEN, Nederland 2 Tilburg University, TILBURG, Nederland Introduction: Increases in mindfulness are assumed to be responsible for improvements in psychological wellbeing during Mindfulness-Based Stress Reduction (MBSR), but compelling evidence is lacking. This longitudinal diary study examines whether within-person changes in mindfulness precede or follow changes in negative affect (NA) and positive affect (PA) during MBSR. Also, inter-individual differences are investigated. Method: Mindfulness, NA and PA were assessed on a daily basis during an 8-week MBSR program in 83 individuals from the general population. Multilevel autoregressive models were used to investigate the temporal order of day-to-day changes in mindfulness and affect. Results: Day-to-day changes in mindfulness predicted subsequent day-today changes in NA and PA. Reverse associations did not emerge. The magnitude of the effects differed substantially between individuals. Conclusions: The findings suggest that changes in mindfulness precede rather than follow changes in affect. As evidenced by our results, enhancing mindfulness indeed seems to be a beneficial means to improve psychological wellbeing. CORRESPONDING AUTHOR: E. Snippe, University Medical Center Groningen, Nederland, [email protected]

S26

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

P71 BLENDING ONLINE AND FACE-TO-FACE THERAPY FOR DEPRESSION: OPPORTUNITIES AND PRECONDITIONS ACCORDING TO PATIENTS AND THERAPISTS R. van der Vaart, J.E.W.C. Van Gemert-Pijnen University of Twente, ENSCHEDE, Nederland

date-ness, accessibility (login problems) and comprehensibility. Questionnaires show average scores per construct from 2.75 (speed) to 3.37 (navigation), measured on a 0-4 scale (higher scores are more positive).ConclusionsParticipatory development led to: clarification of complex nurse tasks and centralized information in an easy to use app for patient safety.

Background: Blending online and face-to-face therapy for major depression offers perspectives to increase the (cost)effectiveness of treatment, while still providing patients the support they need. Our aim was to gain insight into the optimal implementation of blended care, according to patients and therapists. Methods: A two-step Delphi method was performed. In step 1 an survey was used in which patients’ and therapists’ preferences and opinions about online psychotherapy were questioned. In step 2, data from step 1 was used in interviews with therapists to investigate how blended therapy protocols could be set up and what essential preconditions would be. Findings: Twelve therapists and nine patients completed the survey. Overall, blended therapy was positively perceived, especially to enhance patients’ self-management. Practical components of therapy such as assignments, diaries and psycho-education could be provided via online modules according to most respondents. Process-related components, such as introduction, evaluation and discussing thoughts and feelings should be supported face-to-face. The interviews showed that individual tailoring of the treatment is essential in secondary mental health care, due to the complexity of the problemacy. Amount and ratio of online modules should be adjusted according to the patient’s problems and personality characteristics, and therapists should develop skills to apply online therapy in a useful way. Discussion: Blending online sessions with face-to-face therapy is innovative and challenging in secondary mental health care. This study provides starting points to create online modules and screening methods to tailor blended therapy according to needs and skills of individual patients.

CORRESPONDING AUTHOR: J. Wentzel, University of Twente, Nederland, [email protected]

CORRESPONDING AUTHOR: Dr R. van der Vaart, University of Twente, Nederland, [email protected]

P72 TASK SUPPORT FOR NURSES IN ANTIBIOTIC STEWARDSHIP VIA INFORMATION APP J. Wentzel, J.E.W.C. van Gemert-Pijnen University of Twente, ENSCHEDE, Nederland Introduction Nurses experience high task-load and need to rely on good information supply, especially in antibiotic stewardship (prudent antibiotic use). Nurses prepare and administer drugs and monitor the patient during treatment. To increase patient safety, we developed the Antibiotic App, which centralizes and re-structures antibiotic-related information into one location. In a pilot study we evaluated the app, before and after its introduction.MethodScenario-based tests were performed in a before (n=16) and after design, with post-intervention measurement randomization into intervention (n=17) or control group (n=17). Nurses were presented with three scenarios they needed to resolve by searching for information, without or with the app. Task-success, speed and problems encountered were assessed. Interviews were held to discuss app satisfaction. A questionnaire (n=30) measured perceived app quality (Website Evaluation Questionnaire) and persuasiveness (Persuasive Systems Design Model).ResultsAnalysis of scenario tests and interview results are ongoing. Baseline measurements show problems concerning: localizing , interpreting and understanding information, and usability of available information sources. To solve a scenario, nurses need from zero (ready knowledge, no need for a search) to up to eight minutes.Interview results include appreciation of app design and structure (clarity). Some concerns include completeness (not all antibiotics are included in the app), up-to-

P73 STRONG POINTS FOR THE IMPLEMENTATION OF PERSONAL HEALTH RECORDS IN PRIMARY CARE S.M. Akkersdijk, F. Sieverink, L.M.A. Braakman-Jansen, J.E.W.C. van Gemert-Pijnen Universiteit Twente, ENSCHEDE, Nederland Introduction: My Health Platform and e-Vita are two of the largest Dutch personal health records (PHRs) for improving self-management of patients with chronic diseases. Functions include insight in and monitoring of clinical data, self-management support and patient-provider communication. The University of Twente conducts two PhD projects to improve the use and persuasiveness of the PHRs (with 10.500 potential end-users in primary care).To date, there is only a small reach of participants (11-12%). Therefore, we conducted a study to gain insight in the factors that influence the uptake and impact of PHRs and the possibilities to improve the content. Method: Semi-structured interviews were conducted (recorded and transcribed verbatim) among general practitioners (n=1), practice nurses (n=15) and patients (n=10) to explore their reasons to implement a PHR and the encountered or expected facilitators and barriers for implementation in primary care. Results: Despite respondents’ enthusiasm, PHRs have a rather low reach due to a lack of information about the purpose, content and use of the PHRs. Also, little thought has gone towards the integration of PHRs with other health care systems and the integration of the PHRs with national guidelines for the treatment of chronic diseases in primary care. Conclusions: Strong points for the implementation of PHRs in primary care are involving end-users in formulating guidelines for implementing these technologies into their healthcare settings. Therefore, to improve the integration in healthcare setting it is important to provide guidelines and to involve all stakeholders during all stages of the development of PHRs. CORRESPONDING AUTHOR: S.M. Akkersdijk, Universiteit Twente, Nederland, [email protected]

P74 QUALITY OF LIFE PROFILES ACROSS FIVE CHRONIC DISEASES: INPUTS FOR INNOVATIONS IN CHRONIC CONDITION MANAGEMENT I.N. Nagyova1, S.B. Baird2 1 Safarik University, KOSICE, Slovak Republic 2 Ecole des Hautes Etudes en Santé Publique, RENNES PARIS, France Background Chronic diseases are the main reason for poor health, restricted activity and poor quality of life (QoL), affecting over one third of Europe’s population and calling urgently for innovations in health systems. The objective of this study was to compare QoL profiles across five chronic diseases with aim to find similarities and to identify potentials for healthcare improvements. Methods Data on 1566 patients were collated from five chronic disease datasets within the LORIDIS project (APVV-0220-10): rheumatoid arthritis, endstage renal disease, Parkinson’s disease, multiple sclerosis, and coronary heart disease. QoL was measured using the Short Form Health Survey 36

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

(SF-36), from which both a Mental and a Physical Component Summary (MCS, PCS) were calculated. QoL profiles were developed to visualise similarities across diseases. Kruskal-Wallis tests and General Linear Modelling (GLM) were used to statistically compare the diseases and control for socio-demographic factors and co-morbidity. Results Mental quality of life (MCS) was found to be more similar across the five diseases than physical quality of life (PCS). Moreover, controlled GLM analyses showed that MCS (F=1.175, p=0.320), role-emotional functioning (F=0.607, p=0.658) and role-physical functioning (F=1.271, p=0.279) were similar across all five diseases. Pairwise comparisons identified additional similarities in every SF-36 dimension. Conclusions The results of this study show that QoL, particularly mental QoL, is similar across chronic diseases. This supports the idea that patients with different chronic conditions may benefit from similar care. The outcomes may contribute to the improved designs, a wider applicability and more effective implementation of integrated care innovations. CORRESPONDING AUTHOR: Dr. I.N. Nagyova, Safarik University, Slovak Republic, [email protected]

P75 COGNITIVE AND BEHAVIORAL CHARACTERISTICS ASSOCIATED WITH OBESITY IN JAPANESE POPULATION T.M. Tanaka1, T.J. Tayama2, S.H. Shinkawa3, T. Tomiie3 1 Hirosaki University, HIROSAKI CITY, Japan 2 Nagasaki University, NAGASAKI CITY, Japan 3 Health Sciences University of Hokkaido, SAPPORO CITY, Japan Introduction: It is known that cognitive behavioral approach for obesity is effective. However, there is little evidence what kinds of cognitive and behavioral characteristics are associated with obesity in Japan. In this study, we compared irrational dieting beliefs, abnormal eating behavior, and exercise related variables between individuals with thinness, normal weight, and obesity. Methods: We conducted a cross-sectional web-based survey. 1040 individuals (mean age=44, SD=13.61, Range=20-69) were recruited from the database of Macromill INC in Japan. They completed the questions about their demographic features, the Irrational Dieting Beliefs Scale (IDBS; Shinkawa et al, 2011), the Eating Behavior Questionnaire (EBQ; Yamasaki et al, 2013), the Self-Efficacy for Exercise Scale (SE-ES; Oka, 2003), and the Decisional Balance for Exercise Scale (DB-ES; Oka et al, 2003). Participants were divided into three groups (thinness, normal weight, and obesity) based on BMI score. Results: Scores of IDBS all subscales (associations with emotion, concern for mistakes, and excessive justification) except for poor sense of over control, EBQ all subscales (eating rhythm abnormalities, sense of hunger and eating behavior, eating style, meal content, cognition of constitution, substitute eating and drinking, and feeling of satiety) were significantly higher obese group than normal weight group. Scores of SE-ES and DBES were significantly lower in obese group than normal weight group. Conclusion: Our findings suggested that there might be obesity-specific cognitive and behavioral characteristics in Japanese population. CORRESPONDING AUTHOR: T.M. Tanaka, Hirosaki University, Japan, [email protected]

P76 THE WILMA (WEIGHT LOSS MAINTENANCE IN ADULTS) TRIAL: A QUALITATIVE EVALUATION E. Randell1, Yvonne Moriarty1, Sharon Simpson1, Rachel McNamara1, Chris Shaw2 1 Cardiff University, CARDIFF, United Kingdom 2 University of Glamorgan, CARDIFF, United Kingdom

S27

Introduction Weight loss maintenance interventions have had limited effectiveness with weight regain common. We are conducting a trial to assess a 12 month individually tailored intervention based on two key features; Motivational Interviewing (MI) and self-monitoring. As part of the process evaluation of this trial we interviewed study participants at 6 and 12 months. We aimed to interview participants about their participation in the trial, their views of the intervention and longer term weight loss maintenance challenges. Methods We interviewed 14 participants at 6 months and 30 at 12 months. The data were analysed using a framework approach, supported by NVIVO. Results Key themes included weight loss history, weight loss strategies, the intervention and research processes. Individuals described long struggles with their weight, their attitudes to and relationship with food, factors motivating them to lose or keep weight off including realising the health risks associated with obesity. They described barriers to weight maintenance including environmental influences, life events and coping with relapse. Generally the face to face sessions of MI were seen as helpful whilst the phone sessions less so. Some participants described how the MI sessions helped them understand the reasons for their eating patterns and helped them make plans. Planning and self-monitoring were seen as key to weight loss maintenance. Conclusions Participants in the study described long histories of struggles with weight and strategies that have used to manage their weight. They were generally positive about taking part in the study and they found the intervention helpful. CORRESPONDING AUTHOR: E. Randell, Cardiff University, United Kingdom, [email protected]

P77 DEPRESSION AND DIABETES DISTRESS AS RISK FACTORS FOR THE TRANSITION TO INSULIN K.J. Smith, N. Schmitz McGill University, MONTREAL, Canada Background: The transition from oral hypoglycemic medication to insulin therapy is a major life event in people with type 2 diabetes. Traditional risk factors for this transition include difficulty maintaining blood glucose levels and increased physical complications. These risk factors are associated with indicators of psychological status in people with diabetes such as depression and diabetes distress. However, to our knowledge there is no large community-based study that has looked to prospectively ascertain if psychological status can predict the transition to insulin after controlling for important confounders. Methods: A total of 1,200 people who took part in the baseline and first follow-up of the Evaluation of Diabetes Treatment study (2011-2012) were assessed for this analysis. All participants were insulin-naïve at baseline and completed telephone interviews where questions about sociodemographic factors, diabetes complications, self-care behaviours, depression (PHQ-9) and diabetes distress (DDS) were asked. Results: A total of 62 people transitioned from oral hypoglycemic medication to insulin treatment between baseline and follow-up. Logistic regression analyses adjusted for age, sex, diabetes complications, chronic conditions and self-care revealed that those people who transitioned to insulin were more likely to have depression (OR 3.16, 1.33-7.54) or diabetes distress (1.62, 1.05-2.51). Conclusions: This study provides the first evidence that depression and diabetes distress both predict the transition from oral hypoglycemic medication to insulin therapy in people with type 2 diabetes even after controlling for important confounders. CORRESPONDING AUTHOR: Doctor K.J. Smith, McGill University, Canada, [email protected]

S28

P78 DIETARY CHOICES WITHIN MULTIGENERATIONAL AUST RALIAN FAMILIES: DOES THE MOTHER STILL PLAY THE MOST IMPORTANT ROLE? K. RHODES, C. Wilson, I. Prichard, A. Hutchinson, J. Coveney, P. Ward Flinders University, ADELAIDE, Australia Introduction Obesity is a risk factor for a number of chronic diseases. At the individual level, it's important to consider the key behavioural factors involved in making the dietary improvements necessary for disease prevention. Family food choice is complex and a number of people within the family context may share this responsibility. This paper presents a qualitative study that, through family interviews, explored food purchasing, preparation and consumption, and examined the bi-directional influences that occurred between family members within three generations. Method Participants were 13 Australian families (N = 57); each comprising at least one grandparent, parent and child aged 7-17 years. Semi-structured family interviews were conducted. Thematic analysis identified the prevalence of relevant themes and these were compared between selected participant groups. The influence of generation (grandparent, parent, child) and role (grandmother, grandfather, mother, father, daughter, son) were considered. Results Grandmothers and mothers were shown to dominate family food choice decisions including those households where fathers were most responsible for the preparation of family meals. The women in each generation influenced fruit and vegetable consumption by purchasing control, insisting on consumption, monitoring and reminding, conditional treats, food rules and restricting others food choices. Grandparents and children shared a relationship that skipped the parent generation and influenced dietary behaviors bi-directionally. Men, more so than women, preferred unhealthy foods and relaxed the family food rules. Conclusions There are implications for the delivery of dietary health messages used in disease prevention interventions designed to successfully reach all members of multigenerational families. CORRESPONDING AUTHOR: Ms K. RHODES, Flinders University, Australia, [email protected]

P79 MINDFULNESS AND EATING BEHAVIOUR STYLES IN MORBIDLY OBESE MEN AND WOMEN A.A.J.J. Schiffer1, M.A. Ouwens2, L.I. Visser1, N. Raeijmaekers1, I. Nyklícek2 1 TweeSteden hospital, TILBURG, Nederland 2 Tilburg University, TILBURG, The Netherlands Introduction Morbid obesity is highly prevalent and comes with a high risk of various diseases and high financial costs. Knowing the determinants of the disturbed eating behaviours that are characteristic for many morbidly obese persons is important for developing interventions aimed at their eating patterns. Based on results from mindfulness-based intervention trials, one could hypothesize that mindfulness might be one such determinant. However, a paucity of studies examined the correlational association between mindfulness and eating behaviour and, to the best of our knowledge, no study examined this for the morbidly obese. Therefore, the aim of the current study was to examine whether mindfulness is associated with eating behaviours in persons with morbid obesity. Methods 154 persons (81.2% female) with morbid obesity who were candidates for bariatric surgery filled out the Dutch Eating Behaviour Questionnaire (DEBQ), the Freiburg Mindfulness Inventory (FMI) and the Hospital Anxiety and Depression Scale (HADS). Information on

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

demographic variables, weight and length were obtained from medical records. Results Mindfulness was negatively associated with emotional (β= -.24, p< .004) and external (β = -.38, p<.001) eating behaviour, independent of sex, age, educational level, BMI and symptoms of anxiety and depression. Mindfulness was not significantly associated with restrained eating. Conclusion Mindfulness is negatively associated with risky eating styles in morbidly obese persons, above and beyond affective symptoms. Future studies, establishing the causal direction of the association, are needed. CORRESPONDING AUTHOR: Dr. A.A.J.J. Schiffer, TweeSteden hospital, Nederland, [email protected]

P80 THE RELATIONSHIP AMONG BODY IMAGE, NUTRITIONAL STATUS AND SPIRITUALITY I.S.K. Kakeshita, S.S.A. Almeida, Sebastião de Sousa University of São Paulo, RIBEIRÃO PRETO, Brazil Introduction: Body image is an important component of physical, psychological and social health. The hypothesis is that spiritual beliefs can determine behaviors which are conducive to human health, as the experience of Adventists’ lifestyle, although there is little scientific evidence. This study aimed to assess body image perception and body (dis)satisfaction of Seventh-day-Adventists, in relation to spirituality and nutritional status. Method: The experimental sample included 71 Adventists and the control group included 69 participants, aged between 18 and 29 years, of both sexes. The volunteers had their weight and height measured and their body mass index (BMI) calculated. They were sociodemographically characterized and oriented to fill the Spiritual Wellbeing Scale (SWBS). The body image perception was measured using the Brazilian Figure Rating Scale (BFRS). Results: Preliminary analysis showed significant differences between the groups according to SWBS’s religious [F(1,136)=27.99, p<0.001] and existential subscales [F(1,136)=7.2, p<0.01], as well as the effect of sex for both subscales, even considering the effects of group and sex for the religious subscale [F(1,136)=4.07, p<0.05]. Qui square test showed significant difference of body dissatisfaction according to actual BMI [F(2,137)=25.06, p<0.001]. The results of Adventist women were not different from their peers in the control group neither to the body dissatisfaction nor to the inaccuracy of body perception. However, post-hoc orthogonal contrasts showed that Adventists men were significantly less dissatisfied than Adventist women and men from the control group. Conclusions: The nutritional status and the body image perception may be influenced by spirituality, especially in males. CORRESPONDING AUTHOR: Dr I.S.K. Kakeshita, University of São Paulo, Brazil, [email protected]

P81 SELF-MONITORING OF BLOOD GLUCOSE DATA ANALYSIS TECHNOLOGY WITH AN ELECTRONIC MEDICAL RECORDING SYSTEM H.K. KITAOKA, H. OZU, M. OH Seikeikai Hospital, SAKAI CITY, Japan Aims: SMBG data analysis technology with an electronic medical recording system was examined. Methods: The MEQNET SMBG Viewer, a technology developed for analyzing SMBG data, was synchronized with an electronic medical recording system (Fujitsu EGMAIN FX). The graphical representations were then used in our medical practice. A questionnaire concerning the usefulness of this system was administered to 111 diabetic patients (61.8 ± 13.6 years; HbA1c, 7.8 ± 1.3%).

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

Results: (1) This system was ranked as useful compared with a handwritten note (85%). (2) The most useful features were the graph according to time zone ( 39.6%), the graph according to time series (31%). (3) Regarding the ease of understanding changes in blood glucose, this system was rated as “became easy”(64%). (4) This system was described as “easily understood”regarding daily blood glucose readings (47.7%), blood glucose changes before and after meals (16.2%). (5) Blood glucose management with this system was rated as useful with regard to assisting in nutrition therapy (58.6%), managing injections and medications (26.1%). (6) This system was rated as useful for self-management with respect to making an effort toward better glycemic control (69.4%), improvement in blood glucose levels (18.0%). Conclusions: The mean and median values of SMBG data acquired through digital analysis are thought to be important factor. Our study also revealed that this system is useful for improving patient monitoring of blood glucose changes and increasing the motivation for selfmanagement by graphing the SMBG data through the software in the electronic medical recording system. CORRESPONDING AUTHOR: Dr. H.K. KITAOKA, Seikeikai Hospital, Japan, [email protected]

P82 THE RELATIVE ASSOCIATION OF ENVIRONMENTAL AND INDIVIDUAL FACTORS WITH THE DIETS OF MOTHERS C. Black1, G. Abbott2, G. Ntani1, K. Ball2, C. Cooper1, G. Moon3, J. Baird1 1 MRC LEU, University of Southampton, SOUTHAMPTON, United Kingdom 2 CPAN, Deakin University, MELBOURNE, Australia 3 Geography and Environment, University of Southampton, SOUTHAMPTON, United Kingdom Introduction Dietary quality is a determinant of health and illnesses such as obesity and cardiovascular disease. Food intake is recognised as a complex behaviour of multi-factorial origin, whereby individual and environmental factors interact to influence what people eat. This study explored the roles of environmental and individual factors on the dietary quality of a population sample of mothers with young children. Methods A total of 787 mothers from Hampshire, UK, reported complete data in a cross-sectional survey about dietary patterns, demographic and psychological characteristics, and perceptions of the local nutrition environment. Observational data describing the in-store environment of supermarkets and the community nutrition environment (density and diversity of food stores), and staff reported data about the nutrition environment of Sure Start Children’s Centres were collected. A theoretical model predicting dietary quality of mothers with young children was tested using structural equation modelling. Results The model revealed a strong positive relationship between dietary quality and psychological factors supporting healthy eating, but no association with perceived local nutrition environment. Dietary quality was directly associated with the community nutrition environment but not the nutrition environment of mothers’ Sure Start Children’s Centre. The association between the in-store environment of mother’s main supermarket and diet was mediated by psychological factors supporting healthy eating. Conclusion These results indicate that environmental factors are directly and indirectly associated with the dietary quality of mothers with young children. Our results provide support for multifactorial interventions to improve the dietary behaviours of young mothers, targeting both environmental and psychological factors. CORRESPONDING AUTHOR: C. Black, MRC LEU, University of Southampton, United Kingdom, [email protected]

S29

P83 COGNITIVE BEHAVIORAL THERAPY AND WEIGHT LOSS AFTER EATING BEHAVIOR CHANGES INDUCED BY GLP-1 ANALOGUE ADMINISTRATION H.O. Ozu, M. Oh, H. Kitaoka Seikeikai Hospital, SAKAI CITY, Japan Aim. Recently, cognitive behavioral therapy has been considered beneficial for body weight control in patients with type2 diabetes. We investigated changes in eating behavior in patients taking a glucagon-like peptide-1 receptor agonist (GLP-1), which may lead to weight loss and metabolic improvement, and also evaluated the intervention point of cognitive behavioral therapy to make diet therapy more effective. Methods. A questionnaire on eating behavior was given to 10 participants with type2 diabetes who were taking GLP-1. We assessed the questionnaire score as well as the weight and HbA1c level of the participants before and after GLP-1 administration. Multiple regression analysis was performed to identify factors that were significantly associated with the improvement of weight and HbA1c. Results. There was a significant decrease in HbA1c level (8.4 ± 1.5 to 7.0 ± 1.9%) and weight (73.2 ± 20.4 to 69.6 ± 19.6 kg) after 1 year administration of GLP-1 (both p < 0.05). Multiple regression analysis revealed several factors that were associated with an improved HbA1c: “understanding for the cause of obesity”(p < 0.01), “hunger or eating motivation”(p < 0.05), and “eating manner”(p < 0.01). The factors associated with a decrease in body weight were “hunger or eating motivation”, “eating manner”, and “food preference”(respectively, p < 0.01). Conclusion. The findings suggest that stability and improvement in diet therapy can be achieved by directing intervention related to emotion and cognition at “understanding for the cause of obesity,””hunger or eating motivation,””eating manner,”and “food preference.” CORRESPONDING AUTHOR: H.O. Ozu, Seikeikai Hospital, Japan, [email protected]

P84 UNSUPPORTIVE SOCIAL INTERACTIONS AMONG DIABETES PATIENTS ARE ASSOCIATED WITH WORSE DIABETES OUTCOMES O. Baron-Epel University of Haifa, HAIFA, Israel Introduction: Social negativity may be a source of negative health outcomes as opposed to social support. We developed a measure of unsupportive social interactions including two constructs: interference and insensitivity, based on the concept of social negativity. Unsupportive social interactions may be a factor hindering diabetic patients from taking care of themselves.The objectives of this study were to measure the association between unsupportive social interactions among diabetes patients and health outcomes.Methods: A telephone survey was administered to 731 Israelis adult diabetics, insured by Maccabi Healthcare Services. The questionnaire included questions about unsupportive social interactions (including the constructs insensitivity and interference), social support, self reported health, self reported satisfaction with management of diabetes and socioeconomic variables. Information regarding HgA1C was obtained from Maccabi Healthcare Services computerized database.Results: HgA1C was significantly correlated with unsupportive social interactions, mainly interference, however the correlation was low. In a linear regression only interference and not insensitivity was associated with HgA1C, after adjusting for age . Self-reported health and selfreported satisfaction with management of diabetes were also associated with unsupportive social interactions and with HgA1C in a linear regression model, age did not explain this association.Conclusions: This study supports the hypothesis that unsupportive social interactions, mainly interference, may impede on diabetic patients to take care of themselves

S30

and result in worse self-reported health and higher levels of HgA1C. Society needs to understand the needs of patients and help them cope with self-care and not be unsupportive of patients. CORRESPONDING AUTHOR: Prof O. Baron-Epel, University of Haifa, Israel, [email protected]

P85 BEING TEENAGER AND ATHLETE: A RISK FACTOR FOR EATING DISORDERS? C.A. Pruneti, C. Cosentino University of Parma, PARMA, Italy Introduction: In adolescence, researches demonstrated that the majority of young athletes display attitudes similar to subjects with eating disorders, as the attention to the body image and to dieting practice. According to Martinsen et al. (2012) the categories of athletes at higher risk are the elite ones practicing sports where body mass has an strong impact on performance. The aim of this study is to analyse dysfunctional eating habits and body misperception in a sample of young athletes, as a possible risk factor for the development of eating disorders.Methods: A sample of 120 subjects, 65 males (54.11%) and 55 females (45.83%) aging 15 to 24 were consecutively recruited in four gyms and in a high school focused on sports. They fulfilled a questionnaire for the evaluation of body selfperception and eating habits, Pisa Survey for Eating Disorders (PSED). The BMI was calculated for every subject. The statistical analysis Χ2 test Wilcoxon test and Mann-Whitney-U-Test were performed. Results: Data regarding effective weight and height were compared with referred weight and height by the individual and a statistically significant differences were found between estimated and objective data. Body perception distortions and maladaptive eating behaviours were also found with a significant prevalence in female.Conclusions: Young athletes have a highly distorted perception of their bodies, eating disorders detrimental to good sports results, and remarkable risk factors for the onset of complete Eating Disorders. These data suggest the relevance of a multidisciplinary approach to this individuals with a personalized assessment and counselling. CORRESPONDING AUTHOR: Prof. C.A. Pruneti, University of Parma, Italy, [email protected]

P86 CHANGES IN NUTRITIONAL STATUS IN CHILDHOOD CANCER PATIENTS: A PROSPECTIVE COHORT STUDY A. Brinksma, P.F. Roodbol, E. Sulkers., W.A. Kamps, E.S.J.M. De Bont, A.M. Boot, J.G.M. Burgerhof, R.Y.J. Tamminga, W.J.E. Tissing University Medical Center Groningen, GRONINGEN, Nederland Introduction: Under- and overnutrition are linked to adverse outcomes during and after childhood cancer treatment. Therefore, understanding the timing of weight loss and weight gain and their contributory factors is essential for improving outcomes. We aimed to determine in which period of treatment changes in nutritional status occurred and which factors contributed to these changes. Methods:A prospective cohort study of 133 newly diagnosed cancer patients with hematological, solid, and brain malignancies was performed. Anthropometric data and related factors were assessed at 0, 3, 6 and 12 months after diagnosis. Results: Despite initial weight loss at the beginning of treatment in patients with hematological and solid malignancies, body mass index (BMI) and fat mass (FM) increased within 3 months with 0.13 SDS (P<0.001) and 0.05 SDS (P=0.021) respectively. Increase continued during the following months and resulted in a doubling of the number of overnourished patients. Fat free mass (FFM), which was already low at

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

diagnosis, remained low. During the entire study period about 17% of the patients were undernourished on the basis of low FFM. Tube feeding and diminished activity level were related to increases in BMI and %FM respectively. No relationship was found between energy intake or corticosteroids and increase in BMI or %FM. Conclusion: BMI and FM increased during and after the period of intensive treatment, while FFM remained low. Improvement of nutritional status might be accomplished by increasing physical activity from the early phase of treatment. CORRESPONDING AUTHOR: MSc A Brinksma, University Medical Center Groningen, Nederland, [email protected]

P87 DEPRESSION, ANXIETY, QUALITY OF LIFE AND ITS RELATIONSHIP TO BINGE EATING DISORDER IN OBESE PATIENTS M. Rios Erazo Center for Treatment of Obesity, Pontifical Catholic University of Chile, SANTIAGO, Chile Introduction: Binge eating disorder (BED) is a disease of recent categorization according to the DSM-V, so it is important to identify psychosocial factors that distinguish the obese people that have binge eating disorder from those without. Methods: Descriptive - comparative, not experimental study, criteria intentioned. Sample of 961 chilean obese patients (336 men and 625 women) with an average age of 37 (sd: 11.7) and an average BMI of 37.33 (sd: 5.58). According to BMI categories, 4% were overweight, 34,9% were type I obese, 32,5% were type II obese and 28,5% were type III obeses. For measuring of binge eating disorder, Binge Eating Scale was used (Gormally, 1982), while for the assessment of quality of life scale was used EuroQoL (Badia, 1999). HAD scale (Zigmond, 1983) was used for the measurement of depression and anxiety. For data analysis, we used central tendency statistics, and t-student. Results: The 71.5% of the sample had no binge eating disorder, while the remaining 28.5% had BED in its moderate and severe degrees. There are significants differences in quality of life (t = 3.83, p = 0.00) anxiety (t = 10.12, P = 0.00) and depression (t = -9.90 p = 0.00). Conclusion: Obese patients with binge eating disorder have a lower quality of life and a higher rate of depression and anxiety that obese patients without binge eating disorder. CORRESPONDING AUTHOR: Ps M. Rios Erazo, Center for Treatment of Obesity, Pontifical Catholic University of Chile, Chile, [email protected]

P88 INDIVIDUAL VARIATION IN TEMPORAL RELATIONSHIPS BETWEEN STRESS AND FUNCTIONAL SOMATIC SYMPTOMS A. van Gils1, C. Burton2, E.H. Bos1, K.A.M. Janssens1, R.A. Schoevers1, J.G.M. Rosmalen1 1 University Medical Center Groningen, GRONINGEN, Nederland 2 Centre of Academic Primary Care, ABERDEEN, United Kingdom Introduction: Medically unexplained or functional somatic symptoms (FSS) constitute a major health problem because of their high prevalence and the suffering and disability they cause. Psychosocial stress is widely believed to be a precipitating or perpetuating factor, yet there is little empirical evidence to support this notion. Prior studies mainly focused on comparing groups, which has resulted in the obscuring of temporal complexity and individual differences. The aim of this study is to elucidate the relationship between stress and FSS over time within individual patients. Method: Data from 20 patients (17 females, ages 29-59) with multiple, persistent FSS were analyzed. Electronic diaries were used to assess stress

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

and FSS twice daily over the course of 12 weeks. Vector autoregressive (VAR) modeling was used to investigate possible associations between daily average stress and FSS scores within each individual. Results: In 6 subjects (30%), an increase in stress was followed by an increase in one or more FSS. In 3 subjects (15%), an increase in FSS was followed by an increase in stress. Additionally, negative and mixed associations were found. Only 2 subjects (10%) showed no crosslagged association between stress and FSS. We did not find specific types of symptoms to be more stress-related than others. Conclusions: Although stress does not seem to be a universal predictor of FSS, an increase in stress precedes an increase in symptoms in some individuals. Identifying these individuals using time-series analysis might contribute to a more patient-tailored treatment. CORRESPONDING AUTHOR: A. van Gils, University Medical Center Groningen, Nederland, [email protected]

P89 DIALECTICAL BEHAVIOR THERAPY FOR SUICIDAL REPEATERS WITH BORDERLINE PERSONALITY DISORDER IN TAIWAN S. Liu, S. Chen, C.J. Lin, C.F. Lin, M.Y. Chan, H.M. Yeh, J.C. Hu, S.I. W. Mackay Memorial Hospital, NEW TAIPEI CITY, Taiwan Background: Dialectical behavior therapy (DBT) has been recommended as a treatment for BPD. However, no compatible research was done in Asian countries. This paper aims to describe the outcome of a pilot trial of DBT in Taiwan, and to ascertain its clinical utility and acceptability. Methods: This study was an open trial with the 1-year standard DBT. Participants were suicidal repeaters who had at least three symptoms of the BPD of DSM-IV, had at least two suicidal attempts within the last year, > 18 years old, and scored > 40 on the Borderline Symptom Checklist. Participants participated in the standard one-year DBT and were evaluated using standardized self-reported questionnaires to measure various aspects of symptoms at the beginning and every 4 months after the program. Results: 18 participants were enrolled and 3 dropped out (16.7%). The average age was 31.3 (SD 7.3, ranging: 21-46) and female was predominant (n=14). Most of them were single (72.2%) and jobless (61.1%). Based on the ITT analyses, patients exhibited significant improvement from pre- to post-treatment in suicidal ideation, nonsuicidal self-injurious behavior, depressive symptoms, helplessness, and quality of life. Conclusions: This study shows that DBT is acceptable by Chinese BPD and can reduce their self-harm behavior and improve clinical outcome and quality of life. DBT may offer promise as an approach to the psychosocial treatment of Chinese borderline patients with suicide repetition. Currently, a randomized controlled trial is doing for evaluating the efficacy of DBT in Chinese people. CORRESPONDING AUTHOR: Prof. S Liu Mackay, Memorial Hospital, Taiwan, [email protected]

P90 FUNCTIONAL CONNECTIVITY IN DEPRESSED PATIENTS WITH DIFFERENT SYNDROMES IN RESTING-STATE FMRI Z.Y.F. zhang, H.Y. Han, W.Y.Z. Wang, Z.Y.F. Zhang, J.E.H. Jin, D.L.G. Deng, L.L. Li Beijing Friendship Hospital Affiliated to the Capital Medical University, BEIJING, China Background: Considerable evidence has described changes of cerebral activity in patients with depression from functional neuroimaging studies. Furthermore, depressed patients often not only have emotional and mental problems but also have somatic complaints. Based on the Traditional

S31

Chinese Medical (TCM) theory, according to the the clinical symptoms, tongues and pulse, depression involve two major syndromes: the deficiency syndrome (DS) and the excess syndrome (ES). To date, no study has directly linked the brain functional abnormalities to depression while focusing on symptoms. Methods: 21 ES patients and 20 NC underwent a functional MRI scan during resting state. The whole brain functional connectivity of a seed in the posterior cingulated cortex(PCC) was mapped. the differences between functional connectivity between each pair of the groups were statistically analyzed. Results: Compared to NC, the DS patients showed significantly decreased functional connectivity between PCC and right middle temporal gyrus and bilateral precuneus, and increased functional connectivity between PCC and bilateral middle frontal gyrus; the ES patients showed significantly decreased functional connectivity between PCC and bilateral middle temporal gyrus and bilateral precuneus, and increased functional connectivity between PCC and bilateral superior frontal gyrus. In comparison between the DS and ES patients: significantly decreased functional connectivity between PCC and bilateral cerebellum and left superior frontal gyrus were found in the ES group. Conclusion: The disruption patterns of functional connectivity differed in depression patients with different types of syndromes, suggesting that the different depression syndromes may interact with different functional connectivity circuits. CORRESPONDING AUTHOR: postgraduate Z.Y.F. zhang, Beijing Friendship Hospital Affiliated to the Capital Medical University, China, [email protected]

P91 FUNCTIONAL STATUS OF CHD PATIENTS: TRAIT/STATE ANXITY, AND CARDIAC SELF-EFFICACY H. Allahverdipour, Mohammad Asgharijafarabadi Tabriz university of Medical Sciences, TABRIZ, Iran Background: Beliefs and emotions could affect functional status, quality of life, and mortality amongst patients who are suffering coronary heart disease (CHD). Current study examined the role of anxiety: trait/ state, self-efficacy, health beliefs, and functional status among patient with history of CHD. Method: In this correlational study, 105 hospitalized patients and outpatients who were suffering CHD in the Tehran Heart Center Hospital participated by using convenience sampling method. Cardiac selfefficacy, Seattle Angina, and research-designed health beliefs questionnaires were used to gather data. Results: The functional status in CHD patients showed significant relationships with gender, job and having medical insurance of the participants based on univariate analysis. Also, Perceived vulnerability to face again cardiac attack in the future, perceived severity of next cardiac attack, anxiety, state anxiety and trait anxiety had significant and negative relationships with functional status. Conversely, the cardiac self-efficacy had a positive and significant relationship with functional status. In addition, the results of the two-steps hierarchical modeling lead in selecting education and insurance among the participant's characteristics and all psychological factors except for anxiety. Conclusion: These results suggest that psychological factors have important role in functional status and quality of life of patients who suffering CHD. Therefore, it is necessary to emphasize on supportive and complementary programs to promote Cardiac Rehabilitation Programs. CORRESPONDING AUTHOR: Prof. Allahverdipour, Tabriz university of Medical Sciences, Iran, [email protected]

S32

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

P92 WHAT MAKES INTERNET-DELIVERED TREATMENTS WORK? OUTCOME EXPECTATIONS PREDICT IMPROVEMENT IN AN RCT FOR TINNITUS PATIENTS J.R. Rheker1, G. Andersson2, C. Weise1 1 Philipps-Universität Marburg, MARBURG, Germany 2 Linköping University, LINKÖPING, Sweden

role of observer’s characteristics, i.e., dispostional empathy and hypervigilance for pain, proved to be less robust.Conclusion: Observing pain facilitates the detection of tactile stimuli, both in pain and non-pain responders. Interestingly, the pain responder group reported more vicarious somatosensory errors during the experimental paradigm compared to the non-pain responder group and this difference remained stable over time.

Objective. Despite the success of internet-based cognitive behavioral selfhelp treatments (iCBT), only little is known about what makes them work. Availability of therapeutic support and compliance are assumed to positively influence treatment outcome. In face-to-face therapy, patients’ positive outcome expectations were shown to have an advantageous effect on improvement. Thus, our aim was to investigate whether outcome expectations as well as therapeutic support and treatment compliance predict improvement in an iCBT for tinnitus sufferers. Methods. A total of 112 tinnitus patients were randomly assigned to one of two groups. Both groups received the same, well-established iCBT treatment. However, participants of the support group (n=56) could ask a therapist for additional support, whereas participants of the other group (n=56) did not receive therapeutic guidance. Outcome expectations were assessed pre-treatment with the subscales ‘hope of improvement’ and ‘suitability’ of the Patients’ Therapy Expectation and Evaluation Questionnaire (PATHEV). The number of processed modules and logins were used as indicators for compliance. Results. Regardless of group allocation, ‘hope of improvement’ (β=0.30, p=0.01) and compliance (β=0.31, p=0.001) significantly predicted reduction in tinnitus distress. Contrary to our hypothesis, neither therapeutic support, nor the perceived suitability of the intervention predicted outcome. Conclusion. Our results suggest the importance of outcome expectations for the efficacy of iCBT in tinnitus patients. In contrast to our hypothesis and previous findings, the availability of support did not predict treatment outcome. Future studies should thus examine further possible predictors for the efficacy of iCBT.

CORRESPONDING AUTHOR: S.V. Vandenbroucke, University Ghent, Belgium, [email protected]

CORRESPONDING AUTHOR: J.R. Rheker, Philipps-Universität Marburg, Germany, [email protected]

P93 VICARIOUS EXPERIENCES WHILE OBSERVING SOMEONE ELSE IN PAIN S.V. Vandenbroucke, Geert Crombez, Tom Loeys, Liesbet Goubert University Ghent, GHENT, Belgium Introduction: This study investigated whether pain responders (who report vicarious pain in daily life) and non-pain responders differ in the experience of vicarious somatosensory experiences, the detection of somatosensory stimuli while observing another in pain and its stability during an experimental paradigm. The putative moderating role of dispositional empathy and hypervigilance to pain was examined.Method: Pain responders (N=16) and non-pain responders (N=19) viewed videos depicting pain-related (hands being pricked) and non-pain related scenes, whilst occasionally experiencing vibrotactile stimuli themselves on the left, right or both hands. Participants reported the location at which they felt a somatosensory stimulus. Tactile and visual scenes were applied to the same spatial location (congruent trials, e.g., left-left) or opposite location (incongruent trials, e.g. left-right). We calculated the number of vicarious somatosensory experiences while observing pain-related scenes. Thirthy-three of all participants took part of the experiment several months later to investigate the temporal stability.Results: The pain responder group reported more vicarious somatosensory errors compared with non-pain responders and this effect was stable over time. Observing pain also facilitated the detection of tactile stimuli, especially during spatially congruent trials for the pain responder group. The moderating

P94 THE PREDICTIVE VALIDITY OF 'SOMATIC SYMPTOM DISORDER' IN COMPARISON TO COMPETING PROPOSALS FOR SOMATOFORM DISORDERS R.M. Mewes, W. Rief, K. Klaus Philipps-University of Marburg, MARBURG, Germany Introduction: With the DSM-5, a new diagnosis replacing former somatoform disorders was introduced, the Somatic Symptoms Disorder (SSD). However, its predictive validity in comparison to competing proposals for somatoform disorders (i.e., Somatic Symptom Index-4/6 (SSI-4/6), multisomatoform disorder (MSD), bodily distress disorder (BDD), and polysymptomatic distress disorder (PDD)) is unclear. The aim of this longitudinal study was to investigate if SSD shows better predictive validity (defined by mid- and longterm health care utilization (HCU) and disability) than those other proposals. Method: Persons from the general population (N=321) were interviewed at baseline, and one and four years later. Disability, HCU, criteria for different proposals for somatoform disorders, and clinical and psychological variables were assessed. Linear regression analyses with HCU and disability at the two follow-ups as dependent variables were conducted. Results: SSD neither predicted HCU nor disability at one year follow-up. BDD at baseline was the best predictor of HCU both one and four years later (adj. R2 for the final model=.19 and .26). PDD and SSI-4/6 best predicted higher disability at one year follow-up (adj. R2=.15). Disability at four years follow-up was highest in persons who had fullfilled the criteria for MSD, SSD, and PDD concomitantly at baseline (adj. R2=.45) (all p<.001). Conclusions: The predictive validity of SSD was not better than that of the competing proposals for somatoform disorders. It may be suggested, based on our results, that it could be improved by increasing the required number of somatic symptoms and by adding other relevant psychological features. CORRESPONDING AUTHOR: R.M. Mewes, Philipps-University of Marburg, Germany, [email protected] P95 ENRICHING COGNITIVE-BEHAVIORAL THERAPY FOR SOMATOFORM DISORDERS WITH EMOTION REGULATION TRAINING- A RANDOMIZED CONTROLLED TRIAL F. D. W. Weiβ, M. Kleinstaeuber, J.-M. Gottschalk, W. Rief Philipps-University Marburg, MARBURG, Germany Introduction: Somatoform disorders characterized by medically unexplained somatic symptoms are one of the most common mental disorders and have a high economic relevance due to the increased health care use of this patient group. Cognitive-behavioral therapy (CBT) is currently the best evidenced treatment for somatoform disorders, yet only with moderate effect sizes. Hence, more effective interventions are needed. Because there is evidence that patients with somatoform disorders have emotion regulation deficits, the primary goal of the present multi-center study is to compare CBT enriched with emotion regulation training (ENCERT) with standard CBT (SOMA-CBT) in a randomized design.

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

We expect that enriched CBT will be more effective in terms of improving symptom intensity and emotion regulation techniques. Method: N=244 patients with multiple (>3), chronic, medically unexplained physical symptoms causing significant functional impairment will be included. Both treatment arms encompass 20 weekly and regularly supervised individual sessions. SOMA-CBT contains traditional CBT techniques such as psychoeducation, attention defocusing, restructuring of dysfunctional symptom-related cognitions, increase of physical activity and stress management. The additional element of ENCERT is providing patients with techniques of non-judgmental awareness and acceptance of unpleasant body perceptions. Results: The primary outcome (SOMS) assesses somatic symptom severity. Secondary outcome measures include general psychopathology (SCL), depression (BDI), somatic symptom coping (FESV), health anxiety (mSHAI), symptom-caused functional impairment (PDI) and emotion-regulation skills (ERSQ). Conclusions: Enhancing the efficacy of CBT for somatoform disorders will significantly reduce health care costs and improve the quality of life of patients. CORRESPONDING AUTHOR: F. D. W. Weiβ, Philipps-University Marburg, Germany, [email protected]

P96 SCIENTIFIC YOGIC MEDITATION AN ANCIENT TECHNIQUE FOR DE-STRESSING DISTRESS IN FUNCTIONAL BOWEL DISORDERS M.R Kotwal1, C.Z. Rinchen1, S. Kotwal2 1 Government of Sikkim, GANGTOK, India 2 Shunyata, GANGTOK, India Introduction: We live in an age that is gushing with information and dizzying possibilities. To keep up with the choices- trivial and profoundthat confront us at every turn, we multitask. Functional and stress related disorders are on the rise. Etiology includes altered gut motor function, autonomic nervous system abnormality, and psychological stress. Aim. Evaluation of a self learning de-stressing technique Swasthia Sukh Satyam Shivam Sundram. Five step relaxation meditation technique is based on ancient wisdom of over 4 thousand years with modern scientific concept. Daily meditation physically transforms the cerebral cortex. Recent research confirms that the human brain retains an astonishing degree of plasticity and capacity for learning throughout life for gaining deep relaxation, converting distress to eustress. To stimulate interest so that GI centers could evaluate the technique especially in functional bowel disorders. Method. Five-step relaxation meditation practice for 22 minutes 5 days weekly was taught to five IBS patients for 8 weeks. None of the patients had any medication. Results: All patients had relief of symptoms. Conclusion. A preliminary evaluation of Relaxation Meditation in IBS patients. Repeated practice of the technique is required. Could be used in any other medical situation which generates undue psychological stress and anxiety. DVD presentation of the technique will be made. Any one including the gastroenterologists themselves could practice the technique. CORRESPONDING AUTHOR: Prof. M.R Kotwal, Government of Sikkim, India, [email protected]

P97 HEALTH DISPARITIES AMONG MALE COLLEGE STUDENTS BY SEXUAL ORIENTATION D. Brittain, M.K. Dinger Colorado School of Public Health at the University of Northern Colorado, GREELEY, United States of America INTRODUCTION.Gay and bisexual male college students may have an increased risk of many health disparities when compared to heterosexuals. The purpose of this study was to examine multiple health-related

S33

factors by sexual orientation identity among a national sample of male college students. METHOD. Participants were 12,803 self-identified gay, bisexual and heterosexual male college students aged 18-24 years who completed the NationalCollege Health Assessment Fall 2008/Spring 2009 survey. Hierarchical binary logistic regression analyses were used to examine the odds of health-related factors occurring by sexual orientation identity. Adjusted odds ratios, 95%confidence intervals and a Bonferroni-adjusted alpha of .001were used to determine statistical significance. RESULTS. Gay and bisexual males compared to heterosexual males h ad higher odds o f: (1) feelin g too depressed to function(AORgay=1.47; AORbisexual=1.74); (2) no strength training (AORgay=1.78; AORbisexual=2.05); (3) having 2 or more sexual partners in the last 12 months (AORgay=2.91; AORbisexual=2.75); (4) not using a condom or other protective barrier (AORgay=1.56); (5) physical inactivity (AORgay=1.56); (6) experiencing sexual penetration without consent (AORgay = 6.44); (7) being a victim of stalking (AORgay=1.79); and (8) experiencing greater stress (AORgay= 1.59). When compared to gay males, heterosexual males had significantly higher odds of: (1) binge drinking (AOR=0.56); (2) having an unhealthy body mass index (AOR=0.61); (3) not having an HIV test (AOR=0.26); and (4) being verbally threatened (AOR=0.54). CONCLUSION. In conclusion, gay and bisexual male students were more likely to experience a greater number of disparities for risky health-related factors when compared to their heterosexual counterparts. CORRESPONDING AUTHOR: Prof. D. Brittain, Colorado School of Public Health at the University of Northern Colorado United States of America, [email protected]

P98 MALE INVOLVEMENT IN UNINTENDED PREGNANCY, CONTRACEPTION, ABORTION: 'MEN WHO TOUCH AND GO'/ 'ARE TOTALLY RESPONSIBLE'? Y.L.W. Wong University of Malaya, KUALA LUMPUR, Malaysia Introduction: Rising unintended pregnancies among young unmarried women, low contraceptive prevalence, and 25% unmet need for contraception are urgent challenges to women’s sexual reproductive health. Often viewed solely as women’s problem, male involvement and responsibility, though vital, has been neglected. This paper thus aims to assess male involvement in these issues. Methods: On-line Focus Group Discussion (OFGD) and face-to-face indepth interviews (IDI) were conducted with 30 urban multiethnic adolescent boys and young men. Research ethics approval was granted (Ref No: 956.45). Informed consent was duly obtained from respondents. All IDI were digitally recorded, transcribed verbatim and checked. IDI and OFGD data were managed and analyzed using NVIVO Version 8. Results: Average age= 19.7 years and age range= 15-24 years. One third were sexually active but none reported they ever had impregnated their partners. Yet, all feared unintended pregnancy might occur. Sexually active men agreed they should be responsible in preventing unintended pregnancy. Some would marry i.e. “totally responsible”due to traditional male/macho role of being a man. However, majority neither practised contraception readily nor effectively. Instead, “men who touch and go”i.e. run away as they are not bearing the pregnancy. Majority perceived abortion as sinful in any society or religion and yet they did not necessarily view marriage as a solution should unintended pregnancy occur. Conclusion: Men’s ambivalence and lack of involvement impact women’s sexual reproductive health and rights. Approaches that consciously engage men are needed in resolving unintended pregnancy, low contraception, and women’s access to safe abortion. CORRESPONDING AUTHOR: Assoc.Prof.Dr. Y.L.W. Wong, University of Malaya, Malaysia, [email protected]

S34

P99 GENDER DIFFERENCE IN PHYSICAL ACTIVITY PREDICTING PSYCHOSOCIAL OUTCOMES AND ASTHMA CONTROL IN ADOLESCENTS WITH ASTHMA D. Tiggelman, M. O. M. Van de Ven Radboud University Nijmegen, NIJMEGEN, Nederland Introduction Adolescents with asthma experience more psychosocial and physiological problems than their healthy peers. Physical activity may decrease these problems. The current study examined if physical activity could predict changes in psychosocial outcomes (i.e., anxiety/depression, quality of life (QOL), stress) and asthma control over time in adolescents with asthma, and if these relations were moderated by gender. Methods Adolescents with asthma (aged 10-14 at baseline) were visited at home three times, in the spring/summer of 2011, 2012, and 2013 (N = 253). They completed questionnaires about symptoms of anxiety/depression, QOL, perceived stress, and asthma control. Path analyses using Mplus was used to examine longitudinal relations between physical activity, psychosocial variables and asthma control (controlled for BMI, age, and asthma duration) in males and females. Multi-group analyses were used to examine moderating effects of gender on these relations. Results In males, physical activity predicted increased asthma control over time, but did not predict changes in anxiety/depression, QOL, or stress. In females, physical activity predicted decreases in anxiety/depression, but did not predict changes in QOL, stress, or asthma control over time. The relation between physical activity and experiencing less symptoms of anxiety/depression over time was moderated by gender: this relation was significantly stronger for female adolescents. Conclusion In sum, interventions aimed at increasing physical activity could be beneficial for psychosocial and health outcomes in adolescents with asthma. Male adolescents would particularly benefit with regard to asthma control, while in female adolescents, increased physical activity would especially decrease symptoms of anxiety/depression. CORRESPONDING AUTHOR: M.Sc. D. Tiggelman, Radboud University, Nijmegen, Nederland, [email protected]

P100 CAN DISORDER-RELATED KNOWLEDGE IMPROVE THE RATING OF WOMEN SUFFERING FROM PREMENSTRUAL SYMPTOMS? - AN ONLINE-EXPERIMENT C.J. Janda, M. Herget, J. Kues, M. Kleinstäuber, F. Asbrock, C. Weise Philipps University Marburg, MARBURG, Germany Background: In the general population, prejudices about women suffering from premenstrual syndrome (PMS) are common. These prejudices often result from a lack of knowledge about PMS and can cause stigmatization of affected women. Studies in depression have shown that imparting disorder-related knowledge in relatives can reduce symptom burden in patients. The aim of our study was to investigate, (1) if experimentally induced prejudices about PMS result in stigmatization of affected women; and (2) if the provision of information can improve empathy towards PMS-patients. Method: A total of 216 students (50% female), were randomly assigned either to one of two experimental groups (EG1, EG2) or to a control group (CG). First, the groups read three different texts. EG1 read a text informing about PMS. EG2 read a text that presented prejudice-inducing false PMS-information. CG received a text with neutral information. Second, all participants watched a video of a woman reporting PMS. Finally, all participants appraised the woman shown in the video regarding the dimensions warmth and competence (according to the Stereotype Content Model). Results: Mean differences

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

have shown that (1) participants of EG2 rated the woman significantly more incompetent than CG (d=0.46). (2) Participants of EG1 rated the woman significantly warmer (d=0.36) and more competent (d=0.64) than CG. Conclusion: Induced prejudices can provoke stigmatization of women suffering from PMS. Furthermore, information about PMS can facilitate the understanding in significant others. Future research should investigate the benefit of integrating careful counseling of affected persons and their relatives into PMS-treatments. CORRESPONDING AUTHOR: Dipl. Psych. C.J. Janda Philipps, University Marburg, Germany, [email protected]

P101 WOMEN'S ATTACHMENT ANXIETY MEDIATES EFFECT OF NEED FOR PARENTHOOD ON PARTNER PSYCHOLOGICAL WELL BEING M. M. R. Moura-Ramos , M. C. Canavarro University of Coimbra, COIMBRA, Portugal Introduction: Infertility is a stressful event that threatens couples wellbeing. As infertility is a dyadic event, an important question is whether a person’s need for parenthood contributes not only to one own psychological wellbeing but also to the partner’s wellbeing. In addition, due to its interpersonal nature, infertility is likely to activate attachment patterns. Research has shown that attachment style is associated with infertile couples’ adjustment, but it remains unclear whether this effect occurs across partners. This study aimed to explore the mediating role of actor and partner attachment anxiety in the effect of the need for parenthood on psychological wellbeing. Methods: In this cross-sectional study, 45 couples (90 participants) undergoing in vitro fertilization treatment completed self-report questionnaires assessing representations on the importance of parenthood (Fertility Problem Inventory), attachment (Adult Attachment Scale) and psychological wellbeing (WHOQoL - WHO Quality of Life Questionnaire). Based on the Actor-partner model, Path analysis was used to examine direct and indirect effects. Results: Results indicated that women’s attachment anxiety mediated actor and partner effect of need for parenthood on psychological wellbeing. When focusing on the partners’ effects, no direct or indirect effects were found. Conclusions: This study demonstrated that the importance of parenthood in women’s’ life is associated with lower psychological wellbeing in both couple’s members, and this association is explained by the activation of attachment anxiety. This study highlights the importance of considering attachment patterns in clinical settings and of addressing the meaning of parenthood in one’s life to promote infertile couples wellbeing. CORRESPONDING AUTHOR: M. M. R. Moura-Ramos, University of Coimbra, Portugal, [email protected]

P102 EXCESS WEIGHT GAIN DURING PREGNANCY: INFLUENCE ON CURRENT WEIGHT IN PORTUGUESE WOMAN M.G.A. Aparício1, P. Nelas1, J. Duarte1, M.M. Ferreira1, P. Silva2 1 Polytechnic Institute of Viseu, VISEU, Portugal 2 Primary care/Health Centre, VISEU, Portugal Background: Excessive gestational weight gain has been related to the posterior control of body weight and overweight development in women. Objectives: Analyse the correlation between gestational weight gain and actual nutritional status of women of fertile age. Methods: Retrospective study in a sample of 1424 women between 19-51 years old (average=34,5; SD=5,26), who have given birth between 2004 and 2006, and are resident in different areas of Portugal. The gestational

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

weight gain was self-reported and the actual weight obtained by anthropometric evaluation and BMI classification (WHO, 2001). The gestational weight gain was classified according to the Institute of Medicine recommendations. Results: During pregnancy women were between 15-45 y (average=29,74; SD=5,26) and the gestational weight gain ranged between 15 and 40kg (average=11,8; SD=5,8). In 36,95% this fell within the recommended, whilst in 24,20% it classified above those values. Educational level, residency, mother age and gestational diabetes influenced the gestational weight gain. 56,9% had normal weight, 29,8% pre-obesity and 12,1% obesity. The relationship between woman’s nutritional status and gestational weight gain showed that 55,5% and 48,2% of the underweight and normal weight ones had gestational weight gain below the recommended, whereas 33,0% of the ones presently pre-obese and 46,3% obese, gained above the recommended during pregnancy (chi-square=93,32; p= .000). Conclusion: Results show a significant effect of gestational weight gain over posterior retention of weight, highlighting the importance of approaching the body weight subject in pre-conceptional and pre-natal consultation programs to effectively reduce obesity in women in fertile age. CORRESPONDING AUTHOR: Prof. M.G.A. Aparício, Polytechnic Institute of Viseu, Portugal, [email protected]

P103 ADDRESSING GAMBLING MISUSE FOR MAORI WOMEN IN NEW ZEALAND L. Morrison Auckland University of Technology, AUCKLAND, New Zealand

S35

P104 EFFECTS OF SOCIAL SUPPORTS AND HELP-SEEKING PREFERENCE ON INFANT REARING STRESS OF MOTHERS N.K. Kusakabe Fukuyama University, FUKUYAMA, Japan Introduction In this study, the relationship between the stress and social support was examined, and also the relationship between the social support and help seeking preference was examined. It has been said that social support from husbands would be one of the factors reducing stress of mothers who are parenting infants. Method: The participants were 193 mothers (M=34 years old, SD=4.92), whose infants were aged from one year to six years old. Results: The result of t-test showed that scores on the Social Support Scale (by husband and childcare workers) was higher in mothers who felt more stress in “difficulty of parenting”(husband: t=-2.78, p<.01; childcare works: t=-2.74, p<.01), “isolation”(childcare workers: t=-2.96, p<.001), “busyness”(childcare workers: t=-2.97, p<.001), and “infant problem behaviors”(husband: t=-3.13, p<.001; childcare workers: t=-3.16, p<.001). Also scores of Social Support Scale (by doctors) were higher in mothers who felt more stress in “busyness”and “infant problem behaviors”(t=-2.35, p<.05; t=-3.57, p<.010). Conclusions: It became clear that social supports not only from husband but from specialists reduced the stress of mothers. The results of correlation analysis showed that “positive attitude toward helpseeking”increased support seeking behavior to husbands and c h i ld c a re wo r k e rs . H o w ev e r, “ s u s p i ci o n ag a i n s t h e l p seeking”decreased support seeking behavior to doctors. It was suggested that help-seeking preference affected the support seeking behavior of mothers.

Background: The proliferation of gaming machines and the increase of Maori women’s gambling behaviour has been a concern following increased gambling opportunities in lower socio-economic environments in New Zealand, which makes relapse highly likely. However, research indicates that cultural and gender specific interventions and support groups minimizes gambling misuse and improves wellness. Evidence explaining how such interventions work is lacking. Objectives: To describe the behavioural experiences and insights of an intervention programme for gambling misuse with Maori and Pacific women. Method: In-depth interviews were used with Maori and Pacific women. The qualitative data used an inductive and Kaupapa Maori approach. Results: Women described early influences for gambling were fun and enjoyment, and it being a family affair. Gambling continued from childhood to adulthood. Benefits included financial, social and emotional incentives. However, financial, health and personal losses invariably outweighed the positives aspects of gambling. Recognising the need to change included addressing triggers to relapse, self esteem and confidence building, self efficacy and strengthening cultural identity. Family support, cultural reconnection and a structured programme were effective for those women who changed their gambling behaviour, and increased trust impacted positively on family wellness. Conclusions: The intervention programme demonstrated that support groups and cultural identity processes seem to be important in mediating changes in gambling behaviour. An effective strategy for the continuity of the programme is to ensure that Maori providers are trained proficiently in gambling misuse, use a collaborative delivery style, such as, a gambling counsellor and a cultural facilitator.

CORRESPONDING AUTHOR: PhD N.K. Kusakabe, Fukuyama University, Japan, [email protected]

CORRESPONDING AUTHOR: DL Morrison, Auckland University of Technology, New Zealand, [email protected]

CORRESPONDING AUTHOR: Y. Cui, Henan University of Traditional Chinese Medicine, [email protected]

P105 INVOLVEMENT OF MONOAMINERGIC SYSTEM IN THE ANTIDEPRESSANT-LIKE EFFECT OF ESSENTIAL OIL FROM MYRISTICA FRAGRANS HOUTT Y. Cui Henan University of Traditional Chinese Medicine, ZHENGZHOU, China Objective: To investigate the antidepressant effect of essential oil from Myristica fragrans Houtt. (MF-EO). Methods: Healthy male Kunming male mice were divided into several groups including model group, positive control group fluoxetine hydrochloride (FH), and MF-EO (7.5, 8, 8.5 mg/kg). A mouse tail suspension test (TST) and forced swim test (FST) were used for antidepressant observation. Enzyme-linked immunosorbent assay (ELISA) was used for measuring the levels of brain monoamine neurotransmitters including serotonin (5-hydroxytryptamine, 5-HT), dopamine (DA) and norepinephrine (NE) in mice. Results: Compared with the model group (82.60 ± 24.70), MF-EO (8 mg/ kg) significantly shortened the immobility time in TST (54.40±15.87). Moreover, compared with 5-HT (19.35 ± 2.79), DA (12.16 ± 0.71), NE (0.27 ± 0.12) in the model group, MF-EO enhanced mice brain 5-HT, DA and NE levels with the value of (14.95 ± 4.83), (11.32 ± 0.95), (0.20 ± 0.11). Conclusion: MF-EO has antidepressant-like effect, and the mechanism may be involved in brain monoaminergic system.

S36

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

P106 USING EXPERIMENTAL METHODS TO COMPARE IMPLICIT ATTITUDES TOWARDS GENETICS-RELATED WORDS C. Phelps, P. B. Hutchings University of Wales Trinity Saint David Swansea, SWANSEA, United Kingdom

was a significant difference in pain ratings, the same trends were observed for both island and mainland cohorts. Conclusion: This study adds to the understanding of the complex interplay between cognitive abilities and pain, by untangling and better understanding these mechanisms in healthy adults, unburdened by the effects associated with chronic pain.

Introduction It is important that health professionals understand the ways in which people make decisions based on genetic information and how they may influence these processes, as the consequences of poorly informed decision making can have serious consequences on health and lifestyle choices in the population at large. Whilst some elements of attitudes to genetic testing can be captured through the use of explicit attitude measures, subconscious (implicit) attitudes may also play a role in our judgments and behavioural choices. Methods 92 participants completed a word-association task capturing implicit and explicit attitudes towards twenty genetics-related words through valence and reaction times across four key probes. Study Two is currently underway and is exploring attitudes of health professionals (genetic specialists and non-specialists) towards the same target words. Results: Participants who responded to genetic testing as 'positive' demonstrated a significantly faster reaction time to judgements about other genetics-related words (e.g. whether DNA testing was “ethical”or “unethical”) than those reporting genetic testing as 'negative”(RT Diff pvalue ranging from .01-.03). Whilst having similar meanings, use of language impacted upon participants' attitudes e.g., whilst 25% of participants believed that genetic screening was unethical, only 7% believed that genetic diagnosis was unethical. Conclusions: This study suggests that individuals with more positive attitudes towards genetic testing may process genetics-related information more automatically than those with negative attitudes. Whether or not these implicit attitudes are based upon accurate knowledge and understanding, and how health professionals may influence these processes will continue to be explored.

CORRESPONDING AUTHOR: Dr A.B. Banozic, University of Split, Croatia, [email protected]

CORRESPONDING AUTHOR: Dr. C. Phelps, University of Wales Trinity Saint David Swansea, United Kingdom [email protected]

P107 C O G NI T IV E A B IL I TI E S AR E A S S O C IAT E D TO PA IN PERCEPTION A.B. Banozic1, A.Miljkovic1, I. Kolcic2, L. Puljak1, O. Polasek1 1 University of Split, SPLIT, Croatia 2 Department for Public Health, SPLIT, Croatia Background and Aims: The relationship between pain and cognitive abilities is rather complex. Considerable research so far has been done either to estimate cognitive abilities of patients in regards to health treatment (mostly memory and attention) or, more recently, indicating childhood intelligence as a predictor of future widespread chronic pain. Therefore this study aimed to explore the relationship between pain pressure ratings and performance on a battery of cognitive tests in healthy volunteers. Methods: A total of 2836 participants were enrolled from the Island of Korcula (n=1824) and the city of Split (n=1012) in Croatia. A battery of validated mental state tests were used to explore several cognitive domains: attention, executive functions and memory (digit symbol substitution subtest of the Wechsler adult intelligence scale (DST); Verbal fluency test; Rey Auditory Verbal Learning Test (AVLT) and Raven Standardised Progressive Matrices). The assessment lasted approximately 60 min, and was later followed by the mechanical pain pressure threshold and tolerance measurements. Results indicated that cognitive tests (DST and AVLT) were strongly associated to pain measurements in both cohorts (p<0.01). Furtherermore, when the sample was median split on the results of the SPM test, there

P108 SOCIAL AND PSYCHOLOGICAL FACTORS UNDERLYING INFLUENZA VACCINATION BEHAVIOUR IN THE US AND THE UK A. Wheelock, A. Parand, M. Miraldo, C. Vincent, N. Sevdalis Imperial College London, LONDON, United Kingdom Introduction: Influenza epidemics can cause between 500,000 and 1,000,000 yearly deaths worldwide. If left unaddressed, this health disparity is expected to widen due to a rapidly increasing elderly population. Social and psychological factors can significantly influence vaccination behaviour. This review has three aims: 1) to synthesise the available evidence and provide an up-to-date overview of the factors associated with seasonal influenza vaccination uptake among adults in the US and the UK, as the vast majority of studies in this area have focused on the aforementioned vaccine and geographies; 2) to identify the most salient social and psychological factors underpinning vaccination behaviour; 3) to identify areas that require attention and suggest directions for future research. We ultimately seek to inform vaccination policies and communication strategies. Methods and results: We conducted searches of pertinent databases and reviewed publications and references from relevant articles reporting qualitative and quantitative data on social and psychological factors of influenza vaccination in adults. We categorised these factors into eight themes, from most recurrent: social influence, disease-related factors, vaccine-related factors, habit, general attitudes toward health and vaccines, awareness and knowledge, practical barriers and motivators and altruism. Conclusions: Social and psychological factors significantly affect influenza vaccination uptake among adults in the US and the UK. Routinely collected data on social and psychological predictors of vaccination, which can be translated into effective policy and communication approaches, is critical to increase influenza vaccination coverage among high-risk populations. Social and psychological dimensions of vaccination should be progressively incorporated in national health surveys. CORRESPONDING AUTHOR: Ms A. Wheelock, Imperial College London, United Kingdom, [email protected]

P109 EXERCISE PRACTICE MAY INCREASE STRESS COPING ABILITY AND DECREASE DEPRESSION AMONG JAPANESE OFFICE WORKERS K.U. Urakawa1, K.Y. Prof. Yokoyama2, H.I. Itou2, T.M. Matsukawa2, F.K. Kitamura2 1 Juntendo University, MISHIMA, Japan 2 Department of Epidemiology and Environmental Health,Juntendo University, TOKYO, Japan Introduction: Examination of the relationships of exercise to stress coping ability and depression as measured by Sense of Coherence (SOC) and Center for Epidemiologic Studies Depression Scale (CES-D), respectively. Methods: Self-administered questionnaires were distributed among 458 office workers of a company in Tokyo and recovered by mail. The questionnaires included questions on demographic variables of subjects, SOC, CES-D and International Physical Activity Questionnaire (IPAQ).

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

Complete responses were returned from 115 workers (25.1%) and subjected to analysis. Results: Average age of 115 respondents was 46.5 years with a standard deviation (SD) of 11.4. They consisted of 90 males (78.3%) and 25 females (21.7%), with average ages of 49.4 (SD 10.1) and 35.8 (SD 9.5) years, respectively (p<0.001). SOC, CES-D and IPAQ scores were not significantly different between males and females (p<0.05). SOC scores (average 59.4, SD 11.4) were significantly correlated with CESD scores (average 10.2, SD 8.0) in 115 subjects (r =-0.679, p<0.001). IPAQ scores were significantly higher and CES-D scores were significantly lower in 61 subjects who exercie compared to 54 subjects who do not exercie (p<0.05). SOC scores were higher in the former group, but not statistically significant (p=0.055). There were no significant differences in age or gender distribution between the two groups (p>0.05). Conclusions: The practice of exercise appears to be related to better mental health status. CES-D scores were correlated with SOC scores as previously reported (Obidoa C et all , JOEM, 2011), suggesting that higher SOC is a key for good mental health status. CORRESPONDING AUTHOR: Prof. K.U. Urakawa, Juntendo University, Japan, [email protected]

P110 EXERCISE PRACTICE IS MOLDED BY PAST EXPERIENCE OF ATHLETICS IN JAPANESE WORKERS K.Y. Yokoyama1, H. Itoh1, F. Kitamura1, N. Hagi2, T. Mashiko3, T. Matsukawa1 1 Juntendo University Faculty of Medicine, BUNKYO-KU, Japan 2 Yokkaichi Nursing and Medical Care University, YOKKAICHI, Japan 3 The University of Tokyo, TOKYO, Japan Introduction: This study aimed to clarify the relationships of past experience of athletics to exercise practice in Japanese workers. Methods: Self-administered questionnaires were distributed among 1489 workers engaged in 3 companies and 6 hospitals in the Tokai area of Japan (central area of the main island). Completed questionnaires were returned from 968 workers (65%, 333 males and 635 females). Questionnaires included questions regarding exercise practice, past experience of athletics, and other demographic characteristics. In the present study, exercise practice was defined as “engaging in exercise for more than 30 minutes, more than twice per week, for duration of one year or more.” Results: Respondents were aged 19-67 years with an average of 40 and 37 years for males and females, respectively. Thirty-three males and 42 females reported that they engaged in exercise practice. The logistic regression analysis adjusting for age, body mass index, and other possible confounding variables indicated that, for both males and females, moderate levels of exercise at age of 20 years was significantly related to exercise practice at the time of study (OR =3.83 with 95%CI of 1.23-11.99, and 3.00 with 95%CI of 1.01-8.93 for males and females, respectively). For males, intense levels of athletics at the age of 12 years was also related to exercise practice in later life (OR = 4.13, 95%CI 1.21-14.1). Conclusion: Exercise practice in adults appears to be molded by the experience of athletics in childhood and adolescence. CORRESPONDING AUTHOR: K.Y. Yokoyama, Juntendo University Faculty of Medicine, Japan, [email protected]

P111 RISK FACTORS FOR SMOKING BEHAVIOR AMONG HIGH SCHOOL STUDENTS IN SOUGHT OF IRAN A.H. Madani1, T. Aghamolaei1, A.R. Sotoodeh Jahromi2 1 Research Center for Social DeterminantsHormozgan University of Medical Sciences, BANDAR ABBAS, Iran 2 Atherosclerosis Research Center, Department of Immunology, Faculty of Medicine, JAHROM, Iran

S37

Introduction: Smoking is one of the primary public health problems all over the world, and high school students are particularly vulnerable, and susceptible to smoking. The aim of this study was to determine factors associated with increased smoking risks among high school students in southern Iran. Methods: A cross-sectional study was conducted in Bandar Abbas, southern Iran, in 2012. The data was collected by structured questionnaires. The data was analyzed by SPSS software package used descriptive statistics, the chi-square test and bivariate logistic regressions. Results: A total of 410 respondents who participated in the survey, 23.2% stated they had smoked at least once in their lives. 8.3% of respondents reported having smoked during the past month (13.3% of males, and 2.9% of females). The risk of smoking was 5 times higher in males compared to females (Odds Ratio = 5.2; 95% Confidence Interval = 2.1-12.8). Having a close friend smoker increased the risk of smoking to 25 times (OR=25.6; 95%CI=7.7-85.6). Having a smoker at home increased the risk of smoking more than 3 times, (OR=3.6; 95% CI=1.5-8.4). Hearing about smoking hazards at home was a dissuasive factor to smoking. Hearing about smoking hazards in class was not a significant dissuasive factor. Conclusion: The presence of smokers in the family, close friend smokers and gender all play a significant role in smoking behavior among high school students. Parents' education level had no significant effect on smoking in the study area. CORRESPONDING AUTHOR: Prof. A.H. Madani, Research Center for Social Determinants, Hormozgan University of Medical Sciences, Iran [email protected]

P112 QUESTION-BEHAVIOR-EFFECT: DOES A SINGLE ITEM MAKE A DIFFERENCE? E. N. Neter1, N. Stein2, G. Renner3, L. Hagoel3 1 Ruppin Academic Center, HAIFA, Israel 2 Carmel Medical Center, HAIFA, Israel 3 Carmel Medical Center and Technion, HAIFA, Israel Introduction: This research examined the impact of being surveyed on subsequent test uptake. Surveys were conducted by telephone (experiment 1) and by a text-message (experiment 2). Methods: Following mailing of FOBT test kit, participants in the first experiment (N =2196) were randomly assigned to an experimental (telephone interview about cognitions related to FOBT performance) or a control condition (no interview). Participants in the second experiment (N=28,851) were randomly assigned to either a control group (no message), or in intervention of either a declarative or an interrogative sentence sent to their mobile phones regarding intentions to perform the FOBT test. Test performance was retrieved from HMO’s computerized database. Results: In the first experiment, the mean frequency of test performance among participants in the experimental group was 66.6% compared to 61.1% in the control group at 2 months (p=.0.020), and 75.8% as compared to 72.5% at 6 months ( p = .095), among experimental and control groups, respectively. In the second experiment, the mean frequency of test performance at six months among participants in the interrogative message group was 9.1%, as compared to 8.3% in the declarative message group and 7.2% among the control participants ( p = .05) at 6 months. Effect size were d=0.08 and 0.07, for the first and second experiments, respectively. Conclusion: These findings provide evidence that the questions-behavioreffect increases subsequent health behavior even in a lean modality and over time. Effect size was small and similar in the long and short modalities. CORRESPONDING AUTHOR: Dr. E.N. Neter, Ruppin Academic Center, Israel, [email protected]

S38

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

P113 UNDERSTANDING PSYCHOSOCIAL INFLUENCES ON COLLEGE STUDENTS' SLEEP PATTERN IN MACAO A. M. S. Wu, H. C. F. Lao University of Macau, MACAO, China

P115 RELATION BETWEEN HEALTH LOCUS OF CONTROL WITH HEALTH-PROMOTING BEHAVIORS AMONG PEOPLE OVER 15 YEARS OLD T.E. Aghamolaei, A. Ghanbarnejad

Introduction: Unhealthy sleep pattern is prevalent among young people in both the West and the East and hampers their physical and mental wellbeing. Guided by Fishbein and Ajzen’s reasoned action approach (2010), this study aims to investigate the psychosocial influences on healthy sleep pattern among college students in Macao, China. Method: A questionnaire survey was conducted with a convenience sampling at the campus of a local university. In total, 362 college students aged 18-25 years,without serious sleep or emotional disturbance, were successfully recruited with their informed consent and completed the study measures. Results: The results of correlation and regression analyses showed that intention toward healthy sleep and perceived behavioural control significantly explained the variance on college students’ healthy sleep pattern. Moreover, attitudes, subjective norms, and parental nurturance exerted indirect, positive effects on sleep pattern, via promoting intention and/or perceived control. Conclusion: The present findings suggest that positive attitudes toward healthy sleep, favourable social norms, high perceived behavioural control, as well as high parental nurturance are protective factors against unhealthy sleep among Chinese college students. In particular, they highlight the importance of the influences of significant others like parents on young people's healthy sleep, and provide practical insights on how to promote healthy sleep pattern by school-based interventions.

Hormozgan University of Medical Sciences, BANDAR ABBAS, Iran Introduction: Health locus of control is defined as the extent of one’s belief that his/her health is controlled by internal or external factors. The aim of this study was determining the health locus of control and its relation with the health-promoting behaviors among people over 15 years old in Bandar Abbas, Iran. Method: In this cross-sectional study, to determine the health locus of control, form A of Wallston questionnaire was used which consists of three dimensions including: internal health locus of control, chance, and powerful others. To determine the health-promoting behaviors, healthpromoting lifestyle profile II (HPLP II) was used. Out of 830 questionnaires, 687 were completed. Data were analyzed by SPSS 19 software. Results: The mean age of participants was 34.3 (SD=12.1) and 49.8% of them were male. There was a significant difference between internal health locus of control, chance and powerful others (p<0.001). The The lowest mean score belonged to chance (21.35); while the highest mean score belonged to powerful others (30.31), and internal health locus of control (30.32). Based on regression analysis, internal locus of control (β=0.23, p<0.001) and powerful others (β=0.29, p<0.001) significantly predict health-promoting behaviors. Conclusion: Health locus of control of participants was dominantly controlled internally or it was affected by powerful others and chance had less effect on it. Moreover, health-promoting behaviors were higher in participants with a health locus of control that was primarily internal.

CORRESPONDING AUTHOR: Prof. A.M.S. Wu, University of Macau, China, [email protected]

P114 RELATIONSHIP BETWEEN PERCEIVED BENEFITS/BARRIERS AND DECISIONAL BALANCE WITH STAGES OF CHANGE FOR PHYSICAL ACTIVITY T.E. Aghamolaei, A. Ghanbarnejad Hormozgan University of Medical Sciences, BANDAR ABBAS, Iran

CORRESPONDING AUTHOR: Prof. T.E. Aghamolaei, Hormozgan University of Medical Sciences, Iran, [email protected] P116 HYGIENE BEHAVIOUR AND RISK FACTORS AMONG IN-SCHOOL ADOLESCENTS IN FOUR SOUTHEAST ASIAN COUNTRIES S. Pengpid1, K.F. Peltzer2 1 Mahidol University, SALAYA, Thailand 2 Peltzer, SALAYA, Thailand

Introduction: This study aimed to assess relationship between perceived benefits/barriers and decisional balance with stages of change for physical activity among high school students in Bandar Abbas, Iran. Method: This cross-sectional study was conducted in high schools of Bandar Abbas, in the south of Iran. A total of 422 high school students were selected. Twenty two participants did not fill out questionnaires completely. A total of 400 questionnaires were analyzed (response rate=94.8%). The perceived benefits/ barriers to exercise and decisional balance were assessed using the Exercise Benefits/Barriers Scale; and stages of change were assessed using the Physical Activity staging. Data were analyzed in the SPSS v.16. Results: The mean age of the participants was 15.9 (SD=1.1) ranging from 14 to 18 years. More than half of the participants (51.8%) were male and 48.2% were female. About half of the participants (49.2%) were engaged in regular physical activities. Fifteen per cent of the participants were in the precontemplation stage, 15.8% in the contemplation stage, 20% in preparation stage, 15.8% in action stage, and 33.4% in maintenance stage for physical activity. From precontemplation stage to maintenance stage, the perceived benefits and decisional balance were increased and the perceived barriers was decreased. Conclusion: The results of this study are in line with the Transtheoretical model. This model can be used as a framework to design interventions for improving physical activity in this population.

Introduction. The aim of this study was to investigate oral and hand hygiene behaviour, and risk factors, among 13 to 15 year-old in-school adolescents in four Southeast Asian countries. Method. Data was collected by self-reported questionnaire from nationally representative samples (total 13824) of school children aged 13 to 15 years in India, Indonesia, Myanmar and Thailand. Results. Overall 22.4% of school children reported sub-optimal oral hygiene (less than twice a day tooth brushing), 45.2% did not always wash their hands before meals, 26.5% after going to the toilet, and 59.8% using soap to wash their hands. In multivariate analysis female gender, health risk behaviours and lack of protective factors were associated with sub-optimal tooth brushing. Lower socioeconomic status, health risk behaviours, psychological distress and lack of protective factors were found to be associated with sub-optimal hand washing and hygiene behaviour. Conclusion. The cross-national data on oral and hand hygiene behaviour from four Southeast Asian countries found sub-optimal hygiene behaviour. Several determinants of sub-optimal hygiene behaviour were identified, that can be used to inform relevant health programmes, in order to improve oral and hand hygiene behaviour of this adolescent population.

CORRESPONDING AUTHOR: Prof. T.E. Aghamolaei, Hormozgan University of Medical Sciences, Iran [email protected]

CORRESPONDING AUTHOR: S. Pengpid, Mahidol University, Thailand, [email protected]

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

P117 DIFFICULTIES DESCRIBING FEELINGS, STRESS COPING, AND ACCEPTING INJURIES BY ELITE COLLEGE ATHLETES T.T. Tatsumi, Takahiko Fukumoto Kio university, KITAKATSURAGI-GUN, NARA, Japan Purpose: The process model, which hypothesizes that Difficulties in Describing Feelings have an effect on the coping orientation and psychological acceptance of athletic injury, was investigated. Participants: Highly competitive college athletes in Japan that had been forced to stop playing for more than one week, because of an injury they suffered after entering university (N = 133, mean age=20.21 years; mean number of weeks unable to participate in sports = 7.97 weeks, SD=11.26 weeks). Method: Participants were asked to complete three indices: Difficulty Describing Feelings (DDF-S; TAS-20’ Subscale), Psychological Stress Coping Scale (PSC-S) and Athletic Injury Psychological Acceptance Scale (AIPA-S). Results: Covariance structure analysis indicated an indirect effect in which DDF negatively influenced Positive Thought Coping (PTC; β=0.23) and PTC positively influenced Self-Motivation (SM), as well as focus on the Present (FP) of AIPA (β=0.49, 0.52). Conversely, an indirect effect in which DDF positively influenced Avoidance Coping (AVC; β=0.29) and Catharsis Coping (CTC; β=0.17), whereas AVC negatively influenced SM and FP of AIPA (β=-0.28, -0.24) and CTC negatively influenced FP of AIPA (β=-0.21) was also observed. Moreover, the process model was adequately appropriate (GFI=0.954, AGFI=0.894, CFI=0.990, RMSEA=0.035). Conclusion: The process model is a valid model. It is suggested that intervention to prevent DDF enabled a shift, or differentiation from emotion-focused coping to problem-focused coping, and improved the level of AIPA. CORRESPONDING AUTHOR: T.T. Tatsumi, Kio University, Japan, [email protected]

P118 EVALUATING THE EFFICACY OF AN INTEGRATED SMOKING CESSATION INTERVENTION FOR MENTAL HEALTH PATIENTS: STUDY PROTOCOL. A. Metse1, J. Bowman1, P. Wye2, E. Stockings1, R. Clancy3, M. Terry3, L. Wolfenden2, M. Freund2, J. Allan4, J. Prochaska5, J. Wiggers2 1 University of Newcastle, CALLAGHAN, Australia 2 Hunter New England Population Health, WALLSEND, Australia 3 Mental Health and Substance Use Service, WARATAH, Australia 4 Mental Health and Drug and Alcohol Office NSW Department of Health, NORTH SYDNEY, Australia 5 Stanford Prevention Research Center, STANFORD, United States of America INTRODUCTION:As compared to the general population, smoking rates among people with mental illness (MI) are disproportionately high. As a result, up to 70% of people with MI may die from tobacco related diseases. Smoke free policies within mental health hospitals can positively impact on patients’ motivation and self-efficacy to address their smoking. However without post discharge support, preadmission smoking behaviours typically resume. This presentation describes a RCT aimed at assessing the efficacy of integrating mental health inpatient and community based smoking cessation supports, in reducing smoking prevalence. METHODS:Eight hundredparticipants with severe MI will be recruited into the trial. Participants will be recruited whilst patients at one of four psychiatric inpatient facilities in the state of New South Wales, Australia. After completing a baseline interview, participants will be randomly allocated to receive ‘Supported Care’, a multimodal smoking cessation

S39

intervention; or ‘Normal Care’, existing hospital care only. The ‘Supported Care’ intervention will consist of: a brief motivational interview and a package of self-help material for abstaining from smoking whilst in hospital; and following discharge, 16 weeks of motivational telephonebased counselling, 12 weeks of NRT, and a Quitline referral. Data will be collected at one, six and twelve months post discharge. The primary outcomes are abstinence from smoking, and secondary outcomes comprise daily cigarette consumption, nicotine dependence, quit attempts, and readiness to change smoking behaviour. DISCUSSION:If shown to be effective, the study will provide evidence for systemic changes in the provision of smoking cessation care to patients following discharge from psychiatric inpatient facilities. CORRESPONDING AUTHOR: A. Metse, University of Newcastle, Australia, [email protected]

P119 THE MEDIATING ROLE OF PLANNING BETWEEN SUBJECTIVE AGE AND PHYSICAL ACTIVITY J.W. Wienert, S.L. Lippke Jacobs University Bremen gGmbH, BREMEN, Germany Introduction: Past research has shown the interrelation between subjective age and social cognitive predictors of health behavior as well as health behavior themselves. As the relationship between domain specific subjective age and its’ interrelation with physical activity has not been investigated so far, the current study investigated this in more depth. Furthermore we tested if this relationship is mediated by self-regulatory abilities such as planning. Method: An online sample with 2201 participants with two measurements points, 4 weeks apart, was analyzed. Structural equation modeling using Mplus was applied to investigate the relationship between subjective age, intentions, health status, and past behavior with planning and behavior four weeks later, applying Maximum-Likelihood-Estimation. Planning was conceptualized as a latent factor consisting of action planning and coping planning assuming that the latent factor would represent the sole planning component of both. Further, a bias-corrected bootstrap with 5.000 was applied due to non-normal distribution of indirect effects. Results: The model showed a satisfactory fit (CFI=0.98; TLI=0.96; RMSEA=0.036) and supported the mediating role of planning between subjective physical age and physical activity via an indirect path (β=0.02, S.E.=0.008, p<.01). Conclusion: The results suggest that those who feel physically younger make more use of planning as a self-regulatory skill to engage into higher levels of physical activity. CORRESPONDING AUTHOR: J.W. Wienert, Jacobs University Bremen gGmbH, Germany, [email protected]

P120 OBESITY AND PHYSICAL ACTIVITY IN LESBIAN WOMEN IN NORWAY K.M. Malterud, N. Anderssen Uni Research, BERGEN, Norway Introduction: We conducted a review of existing research and found an increased risk of obesity among lesbians compared to heterosexual women in several studies from the US. These findings, however, were not confirmed in recent population based European studies. We therefore wanted to explore the situation for Norwegian lesbians, including their level of physical activity. Methods: Data were drawn from a large cross-sectional self-report survey about living conditions 2013. Participants came from a national population based web panel sample developed to represent the whole

S40

Norwegian population, including 2 902 individuals aged 16-82 years, among these 1 768 self-defined heterosexual participants and 1 134 selfdefined lesbian, gay or bisexual persons. Questions about BMI and physical activity (frequency of sport or exercise sufficient for breathlessness or sweat) were among the issues of a comprehensive questionnaire. For the substudy presented here, we compared responses from 197 lesbians and 979 heterosexual women, adjusting for age, income and education. Results: We found no significant differences in BMI for lesbians compared to heterosexual women. The majority of participants reported physical activity weekly or more frequently, and there were no significant differences across these groups of women. Conclusion: Previous hypotheses about lesbians having higher BMI and being more sedentary than heterosexual women could not be confirmed in this Norwegian study. CORRESPONDING AUTHOR: K.M. Malterud, Uni Research, Norway, [email protected]

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

Results Chi-squared tests for independence indicated that overall exercise and pacing were the two most preferred strategies. There was a significant association between age and preferred strategy. The 33-48 years group found exercise most useful, the over 48 years preferred group activity pacing, and 15-33 years diverse psychological strategies. No significant association was found between gender and preferred strategy. When asked to exclude exercise and pacing, participants identified destressing techniques as the most useful strategy but there was no significant difference for age or gender between psychological strategies. Conclusion The finding in relation to age suggests differences arising out of a stageof-life context for participants, but no significant differences in the selection of strategies were found between males and females. Future research would usefully consider more fully integrating psychological strategies into exercise and activity pacing with different emphases for age. CORRESPONDING AUTHOR: Mr. P. D. Knight, Fremantle Hospital, Australia, [email protected]

P121 EEG COHERENCE ILLUSTRATE CEREBRAL PROCESSING OF OCTAVE ILLUSION IN NORMAL RIGHT-HANDERS W.W. Wang, L.L. Zheng, W.Z. Chen Zhejiang University College of Medicine, HANGZHOU, China Introduction: Understanding how cerebral regions are synchronized when hearing illusion might help to explain some mechanisms of hallucination in psychiatric disorders. EEG theta coherence in response to octave illusion in healthy volunteers might be used for this purpose. Methods: Right-handed healthy volunteers were invited to hear silence, normal and reversed sequences of octave illusion. Results: After both normal and reversed sequences of octave illusion, 11 people reported hearing the high tone in the left ear (RL) and 18 in the right (RR). The task-related power change in the right frontal-temporal areas, and the task-related coherence difference in the right mediofronal-temporal areas were significantly increased in the RR group rather than in the RL group. Conclusion: When processing octave illusion, the right ear predominance was linked with a higher reactivity in the right frontal and right temporal areas in healthy right-handers. CORRESPONDING AUTHOR: Prof. W.W. Wang, Zhejiang University College of Medicine, China, [email protected]

P122 AGE, BUT NOT GENDER, INFLUENCES SELECTION OF BEHAVIORAL STRATEGIES FOLLOWING A MULTI-DISCIPLINARY PAIN MANAGEMENT PROGRAM. P D. Knight Fremantle Hospital, INGLEWOOD, Australia Introduction There is a paucity of studies relating to the effect of age or gender on selection of chronic pain management strategies. This study considered the influence of these two factors with the potential aim of enabling targeted evidence-based adjustments to the intervention according to the requirements of participants. Method Participants were hospital outpatients (N=73) attending a four-week cognitive behavioural multi-disciplinary pain management group. On completion, participants were asked to complete a feedback form identifying the pain management strategies they found most useful (including exercise, cognitive restructuring, activity pacing, acceptance, goal-setting, sleep hygiene, communication, and de-stressing techniques).

P123 THE AVOIDANCE MODEL IN KNEE AND HIP OSTEOARTHRITIS: A SYSTEMATIC REVIEW OF THE EVIDENCE J.F.M. Holla1, D.C. Sanchez-Ramirez2, M. van der Leeden1, J.C.F. Ket2, L.D. Roorda1, W.F. Lems3, M.P.M. Steultjens4, J. Dekker3 1 Reade, AMSTERDAM, Nederland 2 VU University, AMSTERDAM, Nederland 3 VU University Medical Centre, AMSTERDAM, Nederland 4 Glasgow Caledonian University, GLASGOW, United Kingdom Objective To review the scientific evidence for the validity of the avoidance model in patients with knee and hip osteoarthritis (OA). This model hypothesizes that pain and psychological distress lead to avoidance of activities, and thereby to muscle weakness and activity limitations. Methods A literature search was conducted in four electronic databases. Selection criteria included: knee or hip OA or pain; examination of consecutive components of the avoidance model; observational study; original research report. The methodological quality of the selected articles was assessed, and qualitative data synthesis identified levels of evidence. Results Sixty studies were included.In knee OA, strong evidence was found that avoidance of activities is associated with activity limitations via muscle weakness (mediation by muscle weakness); strong evidence was found for an association between muscle weakness and activity limitations; and weak evidence was found that pain and psychological distress are associated with muscle weakness via avoidance (mediation by avoidance). In hip OA, weak evidence was found for mediation by muscle weakness; and strong evidence was found for an association between muscle weakness and activity limitations. Conclusions In knee OA, the association between avoidance of activities and activity limitations is for a substantial part explained by muscle weakness. In both knee OA and hip OA, muscle weakness is associated with activity limitations. These results emphasize the importance of muscle strength in maintenance of activities. More research is needed on the consecutive associations between pain or psychological distress, avoidance of activities and muscle weakness, and to confirm causal relationships. CORRESPONDING AUTHOR: J.F.M. Holla Reade Nederland [email protected]

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

P124 WHAT IS IMPORTANT IN PAIN EDUCATION? THE EXPERIENCE OF PATIENTS WITH CHRONIC PAIN A.J. Wijma1, A. Crom-Ottens2, C.M. Knulst-Verlaan3, D. Keizer4, J. Nijs1, C.M. Speksnijder5, C.P Van Wilgen1 1 Vrije Universiteit Brussel, BRUSSEL, Belgium 2 Leveste, EMMEN, Nederland 3 Zorggroep Rijnmond, ROTTERDAM, Nederland 4 Transcare, HARKEMA, Nederland 5 Radboud University Nijmegen Medical Centre, NIJMEGEN, Nederland Introduction: Pain education (PE), using the sensitization model, is increasingly used and studied in cognitive behavioral treatments of patients with chronic pain. The purpose of this study was to understand how patients experience PE. Methods: 15 patients with chronic pain receiving PE at a transdisciplinary treatment center were interviewed via an interview guide. Two member checks were held. Interviews were transcribed verbatim. Analysis was done according to Grounded Theory and the QUAGOL and a focus group was conducted to improve analysis. Results: Four interacting concepts emerged from the data. Fundamentals; the primary needs to provide PE. Comprises of a biopsychosocial intake before the education and interpersonal aspects of the healthcare professionals. The Basics enhances the bond between patient and caregiver. Comprehensibility; the explanation is understandable, and sufficient interaction between the physiotherapist and psychologist during the education improves the comprehensibility. These two concepts influence the Outcomes of PE; there is an increased awareness, some found peace of mind and experienced fewer complaints. Scepticism; the fourth concept shows apparent doubt towards the diagnosis and sensitization. Some respondents rejected sensitization completely. Some experienced problems with how it was presented. This can negatively impact outcomes. Conclusion: This study provides insight into the factors contributing to the patients’ experience with PE. The intake and good interpersonal factors of the healthcare professionals enhances the bond with the patient. Together with the clear explanation this improves the outcomes of PE. The study findings can be used to facilitate healthcare professionals inproviding PE in a patient centered manner. CORRESPONDING AUTHOR: MSc A.J. Wijma, Vrije Universiteit Brussel, Belgium, [email protected]

P125 USE OF EMDR THERAPY IN PAIN MANAGEMENT M.J.B. Brennstuhl, C.T. pr Tarquinio, F.B. Bassan Universtiy of Lorraine, METZ, France Introduction: Chronic pain is a major healthcare issue. Additional costs involved in pain management each year in France amount to more than one billion euros, despite various government plans put in place between 1998 and 2010. As a multifactorial issue, chronic pain includes four key components, namely sensory, cognitive, behavioural and emotional aspects. While painkillers have been proven effective in treating acute pain, their long-term use has negative side-effects reducing their effectiveness. In fact, efficacy on those treatments on chronic pain is only moderate in more than half of patients. Psychotherapeutic schemes like hypnosis and cognitive-behavioural therapies also seem to show mixed results. Method: Considering the use of EMDR (Eye Movement Desensitization and Reprocessing) therapy is an innovative approach in chronic pain management, this presentation was divided into two steps. The first part focuses on studying the introduction of the traumatic component as a new aspect of chronic pain. As a result, the main focus of this research was to compare the use of EMDR therapy versus eclectic healthcare in a hospital unit specialized in the management of chronic pain. Forty-five patients divided into three

S41

groups were treated by standard protocol of EMDR therapy, pain protocol of EMDR therapy and eclectic therapy. Results and Conclusion: Results show the effectiveness of EMDR therapy on sensory, cognitive, behavioural and emotional but also traumatic components of pain, EMDR standard protocol being most efficient after five sessions, and a reduction of pain extending one month after therapy has been discontinued. CORRESPONDING AUTHOR: PhD M.J.B. Brennstuhl, Universtiy of Lorraine, France, [email protected]

P126 INFLAMMATION AND INNATE IMMUNE RESPONSE IN CHRONIC WIDESPREAD PAIN E. Generaal1, N. Vogelzangs1, G.J. Macfarlane2, R. Geenen3, J.H. Smit1, J. Dekker1, B.W.J.H. Penninx1 1 GGZingeest/VUmc, AMSTERDAM, Nederland 2 Aberdeen Pain Research Collaboration, University of Aberdeen, ABERDEEN, United Kingdom 3 Department of Clinical and Health Psychology, Utrecht University, UTRECHT, Nederland Introduction Dysregulation of the immune system may play a role in chronic widespread pain (CWP) although study findings so far are inconsistent. This cross-sectional study examined whether basal inflammatory markers and the innate immune response are associated with the presence and severity of CWP. Methods Data are from the Netherlands Study of Depression and Anxiety including 1632 subjects. The Chronic Pain Grade questionnaire was used to determine the presence and severity of CWP. Subjects were categorized in a CWP group (n=754) and a control group (n=878). Blood levels of the basal inflammatory markers C-reactive protein (CRP), interleukin (IL)-6 and tumor necrosis factor (TNF-) alpha were determined. To obtain a measure of the innate immune response, 13 inflammatory markers were assessed after lipopolysaccharide (LPS) stimulation in a random subsample (n=707). Results Compared to controls, subjects with CWP showed elevated levels of basal inflammatory markers. However, this association was no longer significant after adjustment for lifestyle and disease factors. For the LPSstimulated inflammatory markers, we did find elevated levels in CWP subjects both before and after adjustment for covariates. Within CWP subjects, pain severity was not associated with inflammation. Conclusion This study demonstrates an exaggerated innate immune response in CWP. Elevated basal inflammatory levels in CWP were driven by lifestyle and disease factors. We suggest that the innate immune response could be a potential biomarker for the onset or perpetuation of CWP. CORRESPONDING AUTHOR: E. Generaal, GGZingeest/VUmc, Nederland, [email protected]

P127 REDUCED HYPOTHALAMIC-PITUITARY-ADRENAL AXIS ACTIVITY IN CHRONIC WIDESPREAD PAIN: PARTLY MASKED BY DEPRESSIVE AND ANXIETY DISORDERS E. Generaal GGZingeest/VUmc, AMSTERDAM, Nederland Introduction Results of studies on hypothalamic-pituitary-adrenal axis (HPA-axis) function amongst patients with chronic widespread pain (CWP) are equivocal and well-controlled cohort studies are rare in this field. The goal of our study was to examine whether HPA-axis function is associated with the presence and the severity of CWP. Methods Data are from the Netherlands Study of Depression and Anxiety including 1125 subjects with and without lifetime depressive and anxiety disorders. The Chronic Pain Grade questionnaire was used to determine the presence and

S42

severity of CWP. Subjects were categorized into a CWP group (n=471) and a control group (n=654). Salivary cortisol samples were collected to assess HPA-axis function (1-h awakening response, evening level, diurnal slope and post-dexamethasone level). Results In comparison with the control group, subjects with CWP showed significantly lower awakening levels and a blunted diurnal slope, in particular among those without depressive/anxiety disorders. CWP subjects with depressive and/or anxiety disorders did not show altered cortisol levels. Conclusion Our results suggest hypocortisolemia in CWP. However, if CWP is accompanied by depressive or anxiety disorder, typically related to hypercortisolemia, the association between cortisol levels and CWP appears to be masked. Future studies should take psychopathology into account when examining HPA-axis function in CWP. CORRESPONDING AUTHOR: E. Generaal, GGZingeest/VUmc Nederland [email protected]

P128 LINE BISECTION PERFORMANCE IN CHRONIC TENSIONTYPE HEADACHE PATIENTS WITH AND WITHOUT MEDICATION-OVERUSE W.W. Wang, J.H. Zhang, Y. Xu, J. Hu Zhejiang University College of Medicine, HANGZHOU, China Introduction: The exact psychopathology of medication-overuse headache (MOH) is still not very clear up to date, a central sensitization due to the defective endogenous pain control and the related attention deficit have been suspected. Line bisection performance is a simpler technique for detecting pseudoneglect, which might help to describe the problem of spatial attention in MOH. Methods: We therefore administered the line bisection test in 26 patients with chronic tension-type headache (CTTH), 21 MOH (which evolved from CTTH), and 25 healthy volunteers. Their depression levels were measured by the Zung Self-rating Depression Scale (SDS), and their anxiety levels by the Zung Self-rating Anxiety Scale (SAS). Results: Compared to the healthy controls, CTTH scored significantly higher on SAS, while CTTH and MOH scored significantly higher on SDS. Compared to the healthy volunteers, the MOH patients bisected significantly leftward, while CTTH patients bisected slightly rightward. The leftward bisection errors in MOH were not correlated with either SAS or SDS scores. Conclusion: The significant leftward errors of the line bisection task might indicate a relative hyperactive right, or hypoactive left hemisphere, or both in MOH, similar to the hemispheric functions found in drug addiction patients. CORRESPONDING AUTHOR: Prof. W.W. Wang, Zhejiang University College of Medicine, China [email protected]

P129 ASSESSING THE ROLE OF COGNITIVE BEHAVIORAL THERAPY IN THE MANAGEMENT OF CHRONIC NONSPECIFIC BACK PAIN V. Sveinsdottir, H.R.E. Eriksen, S.E.R. Reme Uni Research, BERGEN, Norway Introduction: The aim of the study is to provide a narrative review of the current state of knowledge of the role of cognitive behavioral therapy (CBT) in the management of chronic nonspecific back pain. Method: A literature search on all studies published up until July 2012 (PubMed and PsycINFO) was performed. The search string consisted of 4 steps: cognitive behavioral therapy/treatment/management/modification/ intervention, chronic, back pain or low back pain, and randomized controlled trial. The conclusions are based on the results from randomized controlled trials (RCTs) and reviews of RCTs. Interventions were not required to be pure CBT interventions, but were required to include both cognitive and behavioral components.

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

Results: The search yielded 108 studies, with 46 included in the analysis. Eligible intervention studies were categorized as CBT compared to wait-list controls/treatment as usual, physical treatments/exercise, information/ education, biofeedback, operant behavioral treatment, lumbar spinal fusion surgery, and relaxation training. The results showed that CBT is a beneficial treat ment for chronic back pain on a wide range of relevant variables, especially when compared to wait-list controls/treatment as usual. With regards to the other comparison treatments, results were mixed and inconclusive. Conclusion: The results of this review suggest that CBT is a beneficial treatment for chronic nonspecific back pain, leading to improvements in a wide range of relevant cognitive, behav ioral and physical variables. This is especially evident when CBT is compared to treatment as usual or wait-list controls, but mixed and inconclusive when compared with various other treatments. CORRESPONDING AUTHOR: MSc V. Sveinsdottir, Uni Research, Norway, [email protected]

P130 WORK ABILITY AND WORK ROLE FUNCTIONING OUTCOME IN PATIENTS AFTER TOTAL KNEE ARTHROPLASTY M. Stevens1, T.J. Hylkema1, F.I. Abma1, M.R. de Boer2, S. Brouwer1 1 University Medical Center Groningen, GRONINGEN, Nederland 2 VU University, Department of Health Sciences., AMSTERDAM, Nederland Introduction. Not much research has been done with respect to Total Knee Arthroplasty (TKA) in relation to work participation. Aim was to assess work ability and work functioning in patients who underwent TKA and to explore which demographic, disease- and work-related variables were associated. Material and Methods. A retrospective cohort study. A questionnaire was sent to 296 patients who underwent TKA between 2006 and 2010. Work ability was assessed with a single item of the Work Ability Index (WAI). Work role functioning was measured with the Work Role Functioning Questionnaire 2.0 (WRFQ). To determine whether aspects of patient characteristics (age, sex, BMI, education, comorbidity, WOMAC score, having a partner and children) were associated with work ability and work functioning, logistic regression was used. Results. 226 patients (70%) returned the questionnaire. Hunderd-twenty did not have a job and twenty-one did not fill in the questionnaire completely. Leaving 85 patients for analyses. Forty-seven participants (55.3%) scored low (≤90) on work functioning and thirty-eight (44.7%) high (≥90). Forty-nine (57.6%) participants scored low (≤7) on work ability and 36 (42.4%) scored high (≥8). Logistic regression showed that having a partner, higher education and less comorbidity significantly predict a better work ability. For work functioning, a younger age, higher education, less comorbidity and not having children are significantly associated with work functioning. Conclusion. More than 50% of those who returned-to-work scored low levels of work ability and work functioning, which implies higher risk for sickness absence and long-term disabilities. In addition patient characteristics were associated. CORRESPONDING AUTHOR: Dr. M. Stevens, University Medical Center Groningen, Nederland [email protected]

P131 PSYCHOSOCIAL FACTORS ASSOCIATED WITH QUALITY OF LIFE IN COLOMBIAN PATIENTS WITH RHEUMATOID ARTHRITIS H. Brotherton1, H. Rogers1, S. Olivera-Plaza2, M. Segura Durán2, M. Peña-Altamar3 1 Universidad de Deusto, BILBAO, Spain 2 Universidad Surcolombiana, NEIVA, Colombia 3 Clinica Saludcup, NEIVA, Colombia

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

Introduction: Rheumatoid arthritis (RA) is a debilitating autoimmune disease resulting in joint stiffness and decreased mobility. RA impacts Quality of Life (QOL), but little is known about the relationship between disease-specific QOL and psychosocial factors in Latin American RA patients. Methods: One hundred and three RA patients recruited from ambulatory centers in Neiva, Colombia were administered the Disease Activity Scale (DAS), Quality of Life-RA (QOL-RA), Zung Depression Scale, StateTrait Anxiety Inventory (STAI), Interpersonal Support Evaluation List-12 (ISEL-12), and Symptom Checklist-90 Revised (SCL-90R). Results: In bivariate analyses, lower QOL-RA was associated with lower socio-economic status (SES; r=0.26, p<0.01), higher likelihood of using opioids (t=-2.51, p<0.05), and higher likelihood of comorbid pulmonary disease (t=-2.22, p<0.05). Lower QOL-RA was associated with higher Visual Analog Scale (VAS; r=-0.35, p<0.001), Zung depression (r=-0.72, p <0.001), STAI-state (r=-0.66, p<0.001), STAI-trait (r=-0.70, p<0.001), SCL-90R Global Severity Index (r=-0.50, p<0.001), SCL-90R Positive Symptom Total (r=-0.57, p<0.001), and all SCL-90R sub-scales (r’s=-0.54–0.21, p’s<0.01). Lower QOL-RA was associated with lower ISEL-12 sub-scales (r’s=0.41-0.31, p’s<0.001). The multivariate linear regression model indicated that SES (B=2.77, p<0.05), Zung depression (B=-0.53, p<0.001), STAI-State (B=-0.26, p<0.05), and ISEL-12 Belonging (B=1.15, p<0.01) were independently associated with QOL-RA. Conclusions: More depressive and anxiety symptoms were independently associated with lower disease-specific QOL, while more positive perceptions of having people to do things with (belonging social support) and higher SES were independently associated with higher disease-specific QOL. Psychosocial factors have a greater impact on QOL in RA than pain rating, disease activity, and medical factors. CORRESPONDING AUTHOR: H. Brotherton, Universidad de Deusto, Spain, [email protected]

P132 PSYCHOSOCIAL FACTORS ASSOCIATED WITH RHEUMATOID ARTHRITIS DISEASE SEVERITY IN A COLOMBIAN SAMPLE H. Rogers1, P.B. Perrin2, S.K. Trapp2, S. Olivera Plaza3, V.F. Gordillo Forero3, H. Brotherton1, M.L. Peña Altamar4 1 University of Deusto, BILBAO, Spain 2 Virginia Commonwealth University, RICHMOND, VA, United States of America 3 Surcolombian University, NEIVA, Colombia 4 Clínica Saludcoop Neiva, NEIVA, Colombia Introduction: Rheumatoid Arthritis (RA) is a chronic autoimmune disease with symptom exacerbations. Disease activity (including counts of swollen joints, measure of inflammation in the blood, and global pain assessment) in patients is clinically monitored and associated with disease outcomes. Methods: One hundred and three RA patients were recruited from ambulatory centers in Neiva, Colombia. The protocol included: The Disease Activity Scale (DAS), Zung Depression Scale, State-Trait Anxiety Inventory (STAI), Interpersonal Support Evaluation List (ISEL-12), Symptom Checklist - 90 Revised (SCL-90R), Brief COPE, Quality of Life-RA (QOL-RA), and Short Form-36 (SF-36). DAS score was dichotomized into severe (DAS>5.1) vs. moderate or less disease activity. Results: Forty-seven patients (45.6%) comprised the severe DAS group. In bivariate analyses, socio-demographic characteristics were not associated with DAS group. The severe group had higher Visual Analog Scale (VAS) pain ratings (p<0.001), more use of disease modifying antirheumatic drugs (p<0.01), higher Zung scores, and lower scores on: Brief COPE Venting sub-scale, QOL-RA, and 6 SF-36 sub-scales (physical functioning, physical role limitation, emotional role limitation, mental health, social functioning, and pain). Multivariate logistic regression results indicated that the odds of being the severe DAS group was

S43

independently associated with VAS score [OR=2.23(1.52-3.52), p<0.001], venting [OR=0.40(0.17-0.95), p<0.05], and QOL-RA [OR=0.91(0.83-0.99), p<0.05]. Conclusions: Pain rating has an important independent association with disease activity in RA, and QOL is also influenced. Interestingly, venting is a coping strategy that appears to be protective and associated with less disease activity in this Colombian sample, independent of sociodemographic, medical, and other psychosocial factors. CORRESPONDING AUTHOR: H. Rogers, University of Deusto, Spain, [email protected]

P133 PSYCHOSOCIAL FACTORS ASSOCIATED WITH FUNCTIONING IN COLOMBIAN PATIENTS WITH FIBROMYALGIA H. Rogers1, H. Brotherton1, S. Olivera Plaza2, A.F. Córdoba Patiño2, A. De Luis1, M.L. Peña Altamar3 1 University of Deusto, BILBAO, Spain 2 Surcolombian University, NEIVA, Colombia 3 Clínica Saludcoop Neiva, NEIVA, Colombia Introduction: Fibromyalgia (FM) is a central nervous system disorder characterized by widespread pain and allodynia at specific tender points. Other symptoms of FM include fatigue, sleep disturbances, joint stiffness, and cognitive impairments. FM symptoms influence physical functioning and mental health. The importance of psychosocial factors on health status of Latin Americans with FM has not been widely studied. Methods: In an on-going study, twenty-seven FM patients were recruited from ambulatory centers in Neiva, Colombia. The protocol included: Fibromyalgia Impact Questionnaire (FIQ), Zung Depression Scale, State-Trait Anxiety Inventory (STAI), Interpersonal Support Evaluation List (ISEL-12), and Symptom Checklist-90 Revised (SCL-90R). Results: FIQ was not associated with any socio-demographic factors, medications, or comorbidities. Higher FIQ was associated with higher scores on: Zung depression (rho=0.60, p<0.01), STAI-state (rho=0.50, p<0.01), STAI-trait (rho=0.60, p<0.01), and the SCL-90R Global Symptom Index (GSI; rho=0.57, p<0.01), Positive Symptom Total (PST; rho=0.63, p<0.001), Somatization (rho=0.54, p<0.01), Interpersonal Sensitivity (rho=0.46, p<0.05), Depression (rho=0.45, p<0.05), Anxiety (rho=0.45, p<0.05), and Psychoticism (rho=0.49, p<0.01). Two separate multivariate linear regression models showed that FIQ was independently associated with GSI (B=37.01, p<0.05) and PST (B=0.76, p<0.05), controlling for Zung depression, STAI-state, and STAI-trait. Conclusions: Total symptoms of psychopathology (PST) and their severity (GSI) are both uniquely associated with the impact of FM on physical functioning and mental health, taking into account depression and anxiety. Comorbid psychiatric conditions are frequent in patients with FM and psychopathological symptoms appear to have important influence on the extent to which the disorder influences functioning in this Colombian sample. CORRESPONDING AUTHOR: H. Rogers, University of Deusto, Spain, [email protected]

P134 SLEEP QUALITY IN CHRONIC PAIN: STRUCTURAL RELATIONS WITH PAIN, DEPRESSIVE SYMPTOMS, AND SLEEP HYGIENE S.M.G. Sin, S.C. Cho, Sungkun Chungnam National University, DAEJEON, South-Korea Introduction Patients with chronic pain often report poor sleep quality and it has been known to primarily correlate with pain, depressive symptoms, and sleep hygiene. As pain becomes prolonged and persistent, patients with chronic

S44

pain is more likely to develop depressive symptoms and poor sleep hygiene, thereby resulting in poor sleep quality. Thus, this study aimed to investigate the structural relations among pain, depressive symptoms, sleep hygiene, and sleep quality in patients with chronic pain. Methods A total of 155 patients (45.8% males; Mean age = 48.2 years, SD = 17.1) seeking treatment in a tertiary pain center located in Seoul, Korea participated in this study. Results The results indicated that depressive symptoms (b = .95, CI = .82-1.09, p = .002) and sleep hygiene (b = .52, CI = .23-.82, p = .007) had significant direct effects on sleep quality. Also, pain showed significant indirect effects on sleep quality through depressive symptoms and sleep hygiene (b = .51, CI = .32-.75, p = .001); depressive symptoms showed significant indirect effects on sleep quality through sleep hygiene (b = .07, CI = .03-.14, p = .004). Conclusion These findings suggest that pain itself may not elicit poor sleep quality among patients with chronic pain. Instead, grater pain may lead to greater depressive symptoms, thereby contributing to poor sleep hygiene, which may result in poor sleep quality. Thus, attenuating the level of depressive symptoms and encouraging sleep hygiene may be beneficial for sleep quality in patients with chronic pain. CORRESPONDING AUTHOR: S.M.G. Sin, Chungnam National University, South-Korea, [email protected] INVITED SYMPOSIUM: Socioeconomic inequalities in cancer screening participation: challenges to the equitable delivery of cancer control S135 SOCIOECONOMIC INEQUALITIES IN CANCER SCREENING PARTICIPATION: CHALLENGES TO THE EQUITABLE DELIVERY OF CANCER CONTROL J. Wardle1, C. Von Wagner1, M.L. Essink-Bot2, M. Wolf3 1 University College London, LONDON, United Kingdom 2 University of Amsterdam, AMSTERDAM, Nederland 3 Northwestern University, CHICAGO, United States of America Detecting cancer at an early stage of the disease, or finding and treating pre-cancerous changes, holds the promise of reducing cancer morbidity and mortality, and forms a major plank of cancer control policy. However within most screening systems, whether opportunistic or organised, and whether charged for or free, there is a gradient in participation from the more deprived to the more advantaged groups in society. If screening is effective, inequalities in cancer outcomes will be an inevitable result. Understanding and tackling differential uptake of cancer screening is therefore an important issue for equitable delivery of cancer control. In this symposium, the first presentation, by Dr Christian von Wagner (London, UK) will review the literature on inequalities in cancer screening participation and present a framework to help understand the determinants of inequalities in uptake. Professor Marie-Louise Essink-Bot (Amsterdam, Netherlands) will address the challenges inherent in achieving informed decision-making and consider strategies to ensure that provision of the necessary risk-benefit information doesn’t disadvantage lower socioeconomic status groups. The third presentation from Professor Michael Wolf (Chicago, US) will consider the impact of interventions designed specifically for lower income communities. Professor Jane Wardle (London, UK) will facilitate a discussion about priorities in research and implementation to help ensure that all population groups share the benefits of population based cancer screening.

CORRESPONDING AUTHOR: Prof. J. Wardle University College London, United Kingdom, [email protected]

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

S136 A FRAMEWORK FOR INVESTIGATING SOCIAL INEQUALITIES IN CANCER SCREENING C. Von Wagner, K. Whitaker, J. Wardle UCL, LONDON, United Kingdom Introduction: Socioeconomic status (SES) inequalities in cancer screening uptake are ubiquitous. Even in established, organised programmes, such as the UK National Health Service Breast and Cervical screening programmes, strong social gradients in uptake persist. The colorectal cancer screening programme uses a home-based test which removes barriers related to time, transport, and contact with health professionals, but uptake of the first 2.6 million invitations was still strongly graded by SES; varying from 35% in the most deprived quintile of areas to 61% in the least deprived. Method: The aim was to construct a framework to help explain and explore inequalities in cancer screening. Material was drawn from a review of 195 quantitative and qualitative studies, across all screening modalities, which explored associations between SES corollaries (e.g. life stress) and psychosocial correlates of screening uptake. Results: The review highlighted links between corollaries of SES and psychosocial influences on uptake (perceived value of screening, cancer fatalism, self-efficacy). The resulting framework proposed that lower SES has characteristics and consequences that make screening seem more threatening, more difficult to accomplish, and less beneficial. However, existing evidence was largely cross-sectional, and focused on attitudinal factors, leaving questions about causation, and a gap in research into potential non-cognitive influences. Conclusion: The literature review provides a framework of processes linking aspects of life experience associated with SES to screening participation. The next steps are to broaden the range of potential mediators, and to develop interventions that manipulate the hypothesised mediators and assess effects on decision-making and uptake. CORRESPONDING AUTHOR: Dr. C. Von Wagner, UCL, United Kingdom, [email protected]

S137 INFORMED DECISION-MAKING IN COLORECTAL CANCER SCREENING: SPECIAL CHALLENGES AMONG LOWEDUCATED GROUPS M.L. Essink-Bot, E. K. Laan, M. P. Fransen AMC/UvA, AMSTERDAM, Nederland Introduction Colorectal cancer (CRC) screening aims at early detection of CRC to improve health outcomes, but screening has inevitable disadvantageous side effects. Therefore, invitations aim at informed decision-making (IDM). IDM involves individual deliberation on risks and benefits of participation (or non-participation) and reaching a decision based on adequate knowledge and consistent with the individual’s attitude. However, IDM poses specific challenges in less educated groups. Methods We analysed the accessibility and comprehension of the information materials of the Dutch CRC screening program through qualitative interviews among 30 low educated individuals (55-75 years) and conducted a quantitative digital survey to test the effects of the information materials on IDM among individuals with low health literacy (HL; n=150) and adequate HL (n=150). Results The qualitative interviews showed that the information materials were cognitively demanding for less educated individuals. The risk of falsepositive initial test results and the potential risks of colonoscopy appeared especially difficult to grasp. Some respondents questioned the need for deliberation of the pros and cons of participation because they perceived

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

CRC screening as beneficial only, and suggested obligatory participation. The quantitative survey showed lower levels of CRC-screening relevant knowledge among low HL individuals. Conclusion Specific challenges of IDM among less educated individuals include the feasibility of IDM in this group, the content of ‘essential’ knowledge about CRC screening, the situations and conditions for shared decisionmaking and/or delegating a screening decision to an expert, and the feasibility and potential benefits of decision aids to enhance IDM among less educated groups. CORRESPONDING AUTHOR: Prof. M.L. Essink-Bot, AMC/UvA, Nederland, [email protected]

S138 INTERVENTIONS TO PROMOTE CANCER SCREENING IN SOCIALLY DEPRIVED AREAS: CASE OF COLORECTAL CANCER SCREEENING M. Wolf Northwestern University, CHICAGO, United States of America Background: The need to address both initial and repeat colorectal cancer screening completion within public health programmes has become of increasing interest. Objective: To provide an overview of existing U.S. programmes targeting repeat Faecal testing within general practices caring for screening-eligible patients from socioeconomically deprived areas. Methods: Two specific clinical trials are reviewed. One was an individually randomized clinical trial performed among an urban, predominantly Hispanic population, while the other was a clinic-randomized study conducted among rural patients. In both studies, primary care settings specifically situated in areas of high social deprivation served as performance sites. For the ‘urban’ trial, electronic health records and mobile technologies were leveraged to initiate contact with patients pertaining to repeat screening. In the ‘rural’ trial, available clinic resources were used (nurse educator, automated phone follow-up) to prompt both initial and repeat screening completion. Results: Both urban and rural trials demonstrated a nearly two-fold improvement in repeat screening rates compared to a usual care arm. While the urban study utilized available technologies, the rural study was dependent on staff involvement to engage in patients in returning screening tests, making the intervention effective, but not cost-effective. Disparities by literacy level or age were not attenuated. Conclusion: Primary care strategies can be used in diverse settings within the US, leveraging available resources, to significantly improve repeat colorectal cancer screening. CORRESPONDING AUTHOR: Prof. M. Wolf, Northwestern University, United States of America, [email protected] The biology of self-regulation: behaviors, thoughts and emotions

S139 THE BIOLOGY OF SELF-REGULATION: BEHAVIORS, THOUGHTS AND EMOTIONS D. Zahn1, K. Takano2, A. R. Schwerdtfeger3, D. Zahn1, C. Vögele4 1 Johannes Gutenberg University Mainz, MAINZ, Germany 2 KU Leuven, LEUVEN, Belgium 3 Karl-Franzens-University Graz, GRAZ, Austria 4 Université du Luxembourg, UNIVERSITÉ DU LUXEMBOURG, Luxembourg The effective regulation of one’s thoughts, emotions, and behaviors is a core ingredient for maintaining mental and physical health. This

S45

symposium will focus on the relationship between psychological aspects of regulation (ruminative thinking, self-control capacity, and its concomitants), and biological aspects (cardiac arousal, sleep and glucose metabolism) with methodologies ranging from ambulatory monitoring to experimental research. In the first contribution, Keisuke Takano presents results on the effects of ruminative thinking and worry on physiological arousal and sleep quality. In the second presentation, Andreas Schwerdtfeger will offers novel insights into the relationship of positive affect and heart rate variability in daily life and provides evidence of interand intra-individual variability in terms of the affect-heart rate variability link. The third presentation by Daniela Zahn et al. revolves around the notion of glucose metabolism as a biological correlate of self-control, which was examined in a series of experiments that combines selfcontrols task with gold standard metabolic markers (insulin resistance). The symposium will conclude with an integrative discussion by Claus Voegele on further directions of psychobiological self-control research.

CORRESPONDING AUTHOR: Dr. D. Zahn, Johannes Gutenberg University Mainz, Germany, [email protected]

S140 PERSISTENT COGNITION AND SLEEP: PSYCHOLOGICAL AND PHYISIOLOGICAL PATHWAY K. Takano University of Leuven, LEUVEN, Belgium Negative persistent cognition, such as rumination and worry, is known to be a cross-diagnostic risk factor for various psychological and physiological problems. Although most studies on persistent cognition have focused on its maladaptive functions in depression and anxiety, recent theories on sleep disturbances suggest that persistent cognition also plays a central role in the onset and maintenance of insomnia. I will talk about our recent studies that investigated the influences of persistent cognition on subjective and objective sleep quality with longitudinal survey, cognitive experiment, and ambulatory monitoring techniques. These studies suggested that trait rumination and related attentional dysfunction contribute to perception of impaired sleep quality. Furthermore, rumination that occurs in the evening is associated with inhibition of nighttime physiological de-arousal as indexed by reduced heart rate variability, which disturbs normal sleep initiation and maintenance. These findings support recent insights regarding the interconnectedness of psychological and biological processes underlying sleep disturbances, and highlight the importance of emotion regulation and self-control ability in promoting good quality sleep. CORRESPONDING AUTHOR: K. Takano, University of Leuven, Belgium, [email protected]

S141 THE MANIFULD EFFECTS OF POSITIVE AFFECT ON HEART RATE VARIABILITY IN EVERYDAY LIFE A. R. Schwerdtfeger, A. K. S. Gerteis University of Graz, GRAZ, Austria Objective: Positive affect (PA) has been related with better health. However, the biological pathways underlying this link are not well understood. Different facets of PA (e.g., low activated vs. high activated) and both state and trait PA might have different effects on physiology. Methods: This study aimed to relate low and high activated facets of momentary assessed (i.e., state) and aggregated (i.e., trait-like) PA to cardiac vagal tone (heart rate variability; HRV) in everyday life (N = 122 with 3 successive recording days). Results: Aggregated activated PA was associated with higher ambulatory HRV, whereas activated momentary PA

S46

was accompanied by lower HRV. Moreover, momentary deactivated PA was accompanied by higher HRV but aggregated deactivated PA was not. Findings were independent of negative affect and various other demographic and behavioral confounds. Conclusions: Together the findings suggest a more complex picture of the health-related effects of PA. Whereas state-related fluctuations in PA seem to exert both activating and deactivating effects on the heart, activated PA on an aggregated level seems to dampen cardiac arousal, thus suggesting beneficial effects for health. CORRESPONDING AUTHOR: Prof. A. R. Schwerdtfeger, University of Graz, Austria, [email protected]

S142 GLUCOSE METABOLISM, SELF-CONTROL, AND AFFECT INSULIN RESISTANCE IS ASSOCIATED WITH LOWER TRAIT SELF-CONTROL D. Zahn, P. Simon, S. Tug, T. Kubiak. Johannes Gutenberg University Mainz, MAINZ, Germany Introduction. Glucose metabolism has been hypothesized as a biological correlate of self-control and ego depletion, but current studies focused on changes in glucose levels rather than interindividual differences in glucose regulation. Our study aimed at examining the association between glucose regulation and indicators of trait and state self-control in individuals without metabolic disorders. Methods. Insulin resistance (IR), oral glucose tolerance (OGT), trait selfcontrol (Brief Self-control Scale, SCS), trait disinhibition and eating behavior were measured in 71 participants (aged 35.8 ± 13.8 years, 58 % female). Self-control performance was assessed in a dual-task paradigm (control task and Stroop task versus thought suppression task and Stroop task). Associations were determined using Pearson correlations and hierarchical regression analyses. Results. We observed significant correlations between IR and SCS (r =-.35, p < .01) and disinhibited eating (r= .28, p < .05). After adjustment for age, sex, lipids, physical activity and body mass index, there was a significant association for IR and SCS. Stroop performance was not related to IR or OGT. Conclusions. Our study indicates that higher trait self-control is independently related with better insulin resistance - albeit the direction of this association has still to be determined. CORRESPONDING AUTHOR: Dr. D. Zahn, Johannes Gutenberg University Mainz, Germany, [email protected] Back to basics, forward to application: a continuum of knowledge generation in behavioural medicine

S143 BACK TO BASICS, FORWARD TO APPLICATION: A CONTINUUM OF KNOWLEDGE GENERATION IN BEHAVIOURAL MEDICINE M.S. Hagger1, 2 1 Curtin University, PERTH, Australia 2 Health Psychology & Behavioural Medicine Research Group, Curtin University Behavioural medicine is informed by multiple research approaches that can be conceptualized on a continuum ranging from the development of fundamental mechanisms to the application and translation of evidence. Fundamental or basic research represents one ‘pole’ of the continuum and tests proposed mechanisms, usually from theory, in controlled environments. At the opposing ‘pole’ lies applied research representing the translation of interventions and health messages into recommendations

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

in applied settings. Experiments conducted in field settings or randomized controlled trials evaluating the effectiveness of interventions lie at intermediate points on the continuum adjacent to the fundamental and applied poles. The papers in this symposium provide different perspectives on knowledge generation approaches based on the continuum. Each will provide examples of their approach based on their empirical work. Mullan’s paper examines the virtues of testing process and mechanism using laboratory experimentation. Chatzisarantis’ paper outlines how experiments can be used to test theories outside the lab in field settings with real-world implications. Hagger’s paper illustrates the importance of basing behaviour-change interventions on basic research and testing them using appropriate designs. Weed’s paper demonstrates how an established body of evidence on health interventions can inform public health recommendations. The continuum provides a framework to illustrate how knowledge in behavioural medicine develops over time starting with initial small-scale experimental work focusing on efficacy and mechanism, progressing to larger field work and intervention testing, and eventually to large-scale public health recommendations. CORRESPONDING AUTHOR: Prof. M.S. Hagger, Curtin University, Australia, [email protected]

S144 LABORATORY BASED EXPERIMENTAL RESEARCH: TOWARDS AN INTEGRATED RESEARCH METHODOLOGY P. Mullan Curtin University, PERTH, Australia Introduction In behavioural medicine there is a preponderance of predictive studies that make an important contribution to theory development and testing; however such research only provides one part of the solution needed for converging evidence for psychological effects and processes in health behaviour. Thus, research that provides a detailed exploration of the causal mechanisms of behaviour change is needed. Unfortunately, however, it is not uncommon for interventions aimed at promoting health-related behaviour to be designed and conducted with insufficient understanding of these mechanisms. As fundamental or basic research represents one ‘pole’ of the continuum explored in this symposium the purpose of this paper is to explore its importance within behavioural medicine. Methods Research methods used in behavioural medicine are critically evaluated regarding their usefulness in extending theoretical understanding of behaviour, and in promoting behaviour change. Recent examples of research are used to illustrate an approach that targets these goals. Results Incorporating laboratory-based experimental methods can lead to more effective and resource-efficient interventions, through the identification of key behaviour change mechanisms and causal relationships. As such, more innovative and rigorous pilot testing that is relevant to the research question, prior to conducting large-scale interventions, is called for. Conclusion Two main recommendations are made. First, the inclusion of laboratorybased experimental research as standard part of a research protocol in the lead up to intervention development is proposed. Second, a systematic multi-method approach is advocated which incorporates laboratory-based experimental methods to overcome limitations of single method approaches within applied fields of research. CORRESPONDING AUTHOR: Assoc. Prof. P. Mullan Curtin University, Australia, [email protected]

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

S145 THE VIRTUES OF FIELD EXPERIMENTS IN TESTING MECHANISMS AND BUILDING EVIDENCE IN BEHAVIORAL MEDICINE M S. Hagger1, N. L. Chatzisarantis2 1 PERTH, Australia 2 School of Psychology, Curtin University Introduction Experimental evidence is regarded highly as providing quality data on mechanisms such as effects of manipulating psychological variables on health-related outcomes. It assists in isolating processes in the absence of extraneous variables. A drawback of laboratory experiments is lack of ‘real world’ applicability. An alternative is to conduct experiments in ‘field’ or ‘real world’ settings with similar control over extraneous variables but countering applicability problems. This paper will showcase the effectiveness of field experiments in providing evidence for mechanisms in behavioral medicine and leans toward the fundamental end of the research continuum. Methods Two field experiments testing different processes are introduced to illustrate the importance of field experiments in contributing to knowledge in behavioral medicine. The first was conducted in a school setting and manipulated children’s beliefs toward physical activity through leaflet messages based on the theory of planned behavior. The second experiment demonstrated the efficacy of a planning intervention to reduce students’ alcohol purchasing. Results The experiments successfully changed in the target variables with substantial effect sizes. The leaflet message changed children’s physical activity intentions outside school (experiment 1) and the manipulation of plans predicted decision to drink (experiment 2). Conclusion The research demonstrates that with careful control and attention to detail, field experiments can provide high quality data on mechanisms with minimal interference from extraneous variables. Field experiments can therefore be a useful tool for researchers wanting to build evidence for specific processes based on theory. CORRESPONDING AUTHOR: Prof. N. Chatzisarantis, Curtin U, Australia, [email protected] S146 THEORY-BASED INTERVENTIONS, BEHAVOUR-CHANGE TECHNIQUES, MEDIATORS, AND MECHANISMS: CONTRIBUTING TO KNOWLEDGE Martin S. Hagger Health Psychology & Behavioural Medicine Research Group, Curtin University Introduction Recent research on health-behaviour change techniques illustrates the importance of developing interventions that not only work, but can be replicated and whose mechanisms can be pinpointed. To this end, researchers in behavioural medicine are being more explicit in (1)identifying the appropriate intervention components targeting relevant psychological constructs from basic research; (2)developing research to test intervention efficacy based on randomised controlled trial (RCT) designs; (3)testing mechanisms involved inclusion of appropriate mediators; and (4)developing protocols for replication. This paper outlines how this research process is important for knowledge generation on intervention effectiveness in behavioural medicine. The work is located adjacent to the application ‘pole’ of the proposed continuum. Methods Research adopting the methods that incorporate the four components above will be reviewed and essential components highlighted. Recent

S47

examples of theory-based interventions conducted by our research group to reduce alcohol consumption will provide illustrative examples. The studies adopt RCT designs, and include appropriate protocol and mediators. Results The research illustrates the virtues of a systematic approach to developing interventions based on theory and using appropriate designs to generate knowledge regarding intervention effectiveness. The examples demonstrate how isolating specific intervention components can help pinpoint the active ingredients and potential mechanisms for change. Conclusion Knowledge of research interventions needs to be built up through interventions that are based on basic research but are evaluated using gold standard intervention designs with attention paid to precision in protocol and mediators for mechanism. CORRESPONDING AUTHOR: Prof M.S. Hagger, Curtin University, Australia, [email protected] S147 TRANSLATING EVIDENCE FOR PUBLIC HEALTH: INSIGHTS FROM BEHAVIOURAL MEDICINE WILL IMPROVE NATIONAL PHYSICAL ACTIVITY GUIDELINES M. Weed Canterbury Christ Church University, CANTERBURY, United Kingdom Introduction In the USA, Australia and the UK, less than half the populations achieve guideline levels of physical activity, and so questions remain about the usefulness and appropriateness of national physical activity guidelines. The biomedical evidence is clear, so such questions focus on intended audiences, communication and reception. This suggests the need for an approach that moves beyond the realm of biomedical sciences to consider insights from behavioural medicine. Such an approach is clearly translation in nature, but draws on all types of evidence across the proposed continuum. Methods Physical activity guidelines between 1999 and 2014 in the USA, Australia and the UK, and their underpinning evidence, are reviewed to examine: evolution over time, range of evidence bases used, intended audience, communication and reception. Results The results demonstrate that guidelines: (a) have become more bespoke over time; (b) make little use of insights from behavioural science; (c) are confused about intended audiences; (d) are poorly communicated in summary and intent to health professionals, the media and the public; and, (e) have driven research to focus on guideline physical activity levels. Conclusions Physical activity guidelines appear to have had little effect on increasing physical activity levels. This appears to be because biomedical science has driven such guidelines with little consideration of insights from a more integrated behavioural medicine approach. To shift population levels of physical activity, guidelines need to be clearer, better communicated to all stakeholders, and consider issues of reception, take-up and adherence in the least active. CORRESPONDING AUTHOR: Prof. M. Weed, Canterbury Christ Church University, United Kingdom, [email protected] Dyadic family processes and health S148 DYADIC FAMILY PROCESSES AND HEALTH A. DeLongis1, H. Badr2, E. Stephenson1, M. Hagedoorn3, M. Chan4 1 University of British Columbia, VANCOUVER, Canada 2 Mount Sinai, NEW YORK, United States of America 3 University of Groningen, GRONINGEN, Nederland 4 Northwestern University, EVANSTON, United States of America

S48

Chronic health conditions not only restrict physical functioning but also limit the performance of social roles, necessitating a growing body of literature examining interpersonal characteristics during multiple stages of disease development. We present data utilizing dyadic approaches to examine specific psychosocial characteristics within close, intimate relationships and its implications on adjustment and physiological functioning throughout the course of multiple chronic conditions. First, Chan and colleagues show that within parent-child dyads, discrepant appraisals of a standardized social scenario predict parents’ cardiovascular reactivity to family conflict. Then, Badr and colleagues show that the frequency of spousal positive social control was significantly associated with head and neck cancer patients’ self-efficacy and mood, and that engaging in social control does not adversely affect spouse mood. Stephenson and colleagues showed that spouse responses impact the pain experienced by patients with rheumatoid arthritis, and that forms of support that patients report as uniformly satisfying vary in whether or not they are helpful in reducing pain. . Finally, in patients who received a kidney transplantation, Hagedoorn and colleagues demonstrated that those whose partners showed relatively high levels of overprotection had more trouble adjusting after renal transplantation. Together, these findings demonstrate a need for specific assessments of interpersonal characteristics within closely tied individuals across multiple disease models. Utilizing a dyadic approach can help unravel the specific interpersonal characteristics that moderate and have implications for chronic health conditions. CORRESPONDING AUTHOR: A. DeLongis, University of British Columbia, Canada, [email protected]

S149 DISCREPANT APPRAISALS IN PARENT-CHILD DYADS: IMPLICATIONS FOR CARDIOVASCULAR REACTIVITY TO FAMILY CONFLICT M. Chan1, A. DeLongis2, E. Chen1 1 Northwestern University, CHICAGO, United States of America 2 University of British Columbia, VANCOUVER, Canada Introduction: There is extant evidence documenting the implications of discrepancy between closely connected individuals for the quality of relationships. We extend this research to explore how discrepancy between family members relates to physiological responses to conflict and examined 1) how parents and children differed in their views toward the same standardized social scenario, 2) how this parent-child discrepancy predicted the cardiovascular reactivity of each family member during a conflict discussion. Method: One hundred and twenty-five children aged 10 to 20 and their parents were studied. Children and parents viewed videos depicting an ambiguous social scenario and were interviewed separately about their appraisal of the situation. Responses were coded and parent-child discrepancy scores were computed. Families then engaged in a discussion task about a highly rated area of disagreement in their relationship (conflict task). Blood pressure (BP) and heart rate (HR) readings were taken at baseline and during the conflict task. Results: Greater absolute parent-child discrepancy (greater difference between parent and child appraisals, regardless of directionality) predicted greater SBP reactivity in parents during the conflict task. Greater directional parent-child discrepancy predicted greater DBP and HR reactivity in parents during the conflict task, such that the more negatively children viewed an ambiguous social situation relative to their parents, the greater the parents' reactivity during the conflict task. Conclusions: These findings suggest that discrepancy between children and parents in terms of their appraisal of the same social scenario may play a role in shaping how parents in particular respond physiologically to family conflict. CORRESPONDING AUTHOR: M. Chan, Northwestern University, United States of America, [email protected]

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

S150 CONTROL AND SYMPTOM MANAGEMENT IN COUPLES COPING WITH HEAD AND NECK CANCER H. Badr1, C.W. Yeung1, M.A.L. Lewis2, K. Milbury3, W.H. Redd1 1 Mount Sinai School of Medicine, NEW YORK, United States of America 2 RTI International, RESEARCH TRIANGLE PARK, NC, United States of America 3 MD Anderson, HOUSTON, TX, United States of America Head and neck cancer (HNC) patients experience debilitating side effects, such as abnormally reduced salivation and difficulty in swallowing. Intensive self-management protocols are often prescribed to control side effects and minimize discomfort, but patient non-adherence rates are high. Although spouses and significant others are in a prime position to encourage adherence, few studies in cancer have examined how spouse social control tactics (e.g., nagging, hinting) affect patient mood and selfefficacy for symptom management or how engaging in social control affects the spouse. In this study, 125 HNC patients (86% male) and their spouses were recorded in the lab as they discussed a cancer-related concern of their choosing. Evidence of the use of spouse social control tactics was noted in 90% of the 68 discussions about side effects. Oral complications were among the most commonly discussed side-effects but were the least likely to be targeted by spouses (7%), who focused on maintaining the patient’s weight and hydration (58%). Spouses engaged in both positive (51%) and negative (49%) tactics, but only the frequency of positive tactics was significantly associated with patient self-efficacy and mood. There was no association between engaging in social control and spouse mood. Taken together, our data suggest that HNC patients understand and appreciate their spouses’ control attempts and that engaging in social control does not adversely affect spouse mood. Couplebased programs that teach spouses how to use positive social control tactics may help boost patients’ mood during treatment and empower them to engage in recommended self-care behaviors. CORRESPONDING AUTHOR: Prof. H. Badr, Mount Sinai School of Medicine, United States of America, [email protected]

S151 DYADIC COPING WITH RHEUMATOID ARTHRITIS: THE ROLE OF SPOUSE RESPONSES E. Stephenson 1 , D. B. King 1 , S. Holtzman 1 , M. Hagedoorn 2 , A. DeLongis1 1 University of British Columbia, VANCOUVER, Canada 2 University of Groningen, GRONINGEN, Nederland Introduction. Within the marital context of chronic illness, the impact of both partners’ coping responses on disease activity is not well understood. We employed an intensive longitudinal design to examine the influence of spouse support responses on patient rheumatoid arthritis (RA). Method. Data were collected twice daily for a period of one week from each member of married couples in which one spouse was diagnosed with RA. Multilevel modeling was used to examine the within day effects of morning patient pain coping responses and morning spouse support on evening patient pain, controlling for morning pain. Results. Love and acceptance from the spouse was found to reduce patient pain. However spouse’s offers of comfort and solicitous responses, even though considered positive and supportive by the patient, did not have the same effect. Such spouse responses were found to predict increased pain. Conclusions. Findings underscore the importance of examining spouse reports of support within a dyadic coping framework. They further suggest that: 1) not all forms of support offered by the spouse are beneficial, and 2) even forms of support patients report as satisfying (such as solicitousness) are not necessarily helpful in reducing pain. These findings suggest the need for the finer assessment of specific support transactions between spouses.

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

S49

CORRESPONDING AUTHOR: E. Stephenson, University of British Columbia, Canada, [email protected]

with many cancer worries and emotional distress prior to genetic testing, with a high risk perception and those with a weak social network were at increased risk of psychological problems after an unfavorable test result.

S152 DYADIC ADJUSTMENT BEFORE AND AFTER KIDNEY TRAN SPLANTATION M. Hagedoorn, F. Hein, A.V. Ranchor UMCG, GRONINGEN, Nederland

CORRESPONDING AUTHOR: drs. J.S. Voorwinden, UMCG, Nederland, [email protected]

Background: Usually, the physical and emotional wellbeing of patients improves considerably when they receive a kidney transplant after being treated with renal dialysis for quite some time. However, little is known about the role of the spouse in this process. During dialysis the spouse often fulfills caregiving tasks that are no longer required after the transplant. This study aimed to examine changes in spousal behavior during the transition, and the predictive value of spousal support before transplant for distress in couples after transplant. Methods: 66 couples completed questionnaires, including measures of distress (GHQ), self-rated health (VAS-scale and symptoms) and spousal support before as well as 3, 6 and 12 months after transplantation. Results: Both physical and emotional wellbeing in patients improved over time, as did emotional wellbeing in spouses. Along with these improvements, spousal support as well as overprotection decreased over time. However, spousal overprotection before transplantation was associated with higher levels of distress in patients 12 months after the transplant. Conclusion: It appears that the patients whose partners showed relatively high levels of overprotection had more trouble adjusting after renal transplantation. This suggests the need for attention to the dyadic context and changes in partner roles even before transplantation. CORRESPONDING AUTHOR: Prof. M. Hagedoorn, UMCG, Nederland, [email protected]

O153 PROGNOSTIC FACTORS FOR PSYCHOLOGICAL DISTRESS AFTER PRESYMPTOMATIC GENETIC TESTING FOR HEREDITARY CANCER J.S. Voorwinden, J.P.C. Jaspers, J.G. Ter Beest, R.H. Sijmons, J.C. Oosterwijk UMCG, GRONINGEN, Nederland Background: The psychological impact of an unfavorable genetic test result for counselees at risk for hereditary cancer seems to be limited: only 10-20% of them have psychological problems. Objective: To find prognostic factors that can predict which counselees are most likely to develop psychological problems after presymptomatic genetic testing. Method: Counselees with a 50% risk of BRCA1/2 or Lynch syndrome were stratified into a group with a favorable DNA test result and a group with an unfavorable result. In these groups, the psychological impact of the genetic test result, i.e. cancer worries and emotional distress, was examined shortly after learning the test result and 4-6 weeks later using questionnaires. Subsequently, the influence of various potentially prognostic factors on psychological impact were examined in the whole group Results: Data from 165 counselees were analyzed. Counselees with an unfavorable outcome did not have more emotional distress, but showed significantly more cancer worries 4-6 weeks after learning their DNA test result. Prognostic factors for cancer worries after genetic testing were preexisting cancer worries, being single, a high risk perception and an unfavorable test result. Emotional distress was best predicted by preexisting cancer worries and pre-existing emotional distress. Conclusions: The psychological impact of an unfavorable genetic test result is considerable if it is measured as worries about cancer. Counselees

O154 QUALITY OF LIFE AND PSYCHOLOGICAL DISTRESS AMONG MELANOMA PATIENTS - PREDICTORS IN LONGITUDINAL RESEARCH C. Krajewski 1, S. Benson2, S. Elsenbruch2, D. Schadendorf3, E. Livingstone3 1 Inst. of Medical Psychology & Behavioral Immunobiology, ESSEN, Germany 2 Inst. of Medical Psychology & Behavioral Immunobiology, University Hospital, ESSEN, Germany 3 Department of Dermatology, University Hospital of Essen, ESSEN, Germany Background: Melanoma is the most lethal type of skin cancer. Incidence rates are increasing especially in Nordic and North-Western European countries whereas mortality rates are constant or even decreasing. Most patients are diagnosed with thin tumours with a low mortality rate but a latent risk of recurrence, thus characterising melanoma as a chronic disease which requires long-term medical surveillance. Although psychological variables and especially quality of life (QoL) are essential outcome factors in oncology, studies in melanoma patients are scarce. Methods: The aim of this study was to investigate QoL and psychological distress in melanoma patients and to identify predictors of QoL. Patient and tumour characteristics as well as treatment modalities were collected for N=1,264 patients who were diagnosed with melanoma in 2008 in this German nationwide, multicentre study and followed-up prospectively for four years. At four years, N=561 patients (response rate: 57%) answered a self-reported questionnaire addressing QoL (EORTC QLQ-C30), anxiety and depression (HADS), and coping mechanisms (Trier Scales on coping with illness). Results: Stepwise regression analysis revealed that better QoL was predicted by lower HADS depression (ß=-.296) and anxiety (ß=-.229), younger age (ß=-.155), decreased TSK “exchange of information”(ß=-.142) and increased “social integration”(ß=.140) scores. Interestingly, clinical parameters did not contribute and were therefore not included into the model, which explained 34% of total variance (corrected R2). Discussion: This study offers a first step to identify risk factors of impaired QoL in melanoma patients. It may help to elucidate patient characteristics indicating potential need of psycho-oncological care. CORRESPONDING AUTHOR: C. Krajewski, Inst. of Medical Psychology & Behavioral Immunobiology, Germany, [email protected]

O155 VAGAL NERVE ACTIVITY PREDICTS SURVIVAL IN METAST ATIC PANCREATIC CANCER, MEDIATED BY INFLAMMATION Y.G. Gidron1, M.D.C. De Couck1, R.M. Maréchal2, J.L.V.L. Van Laethem2 1 Free University of Brussels (VUB), BRUSSELS, Belgium 2 Eramse Hospital, ULB, BRUSSELS, Belgium Introduction: Recent research findings suggest neuro-modulation of tumors, with the vagal nerve possibly slowing tumor progression. This study examined the relationship between vagal nerve activity, indexed by heart rate variability (HRV), and overall survival (OS) in advanced pancreatic cancer patients, and the mediating role of inflammation. Methods: A “historical prospective”design was used, where HRV and other confounders were retroactively obtained from medical charts at diagnosis, and subsequent OS was examined. Data of N = 272 Belgian

S50

patients with metastatic pancreatic cancer were examined. HRV was obtained from ECGs near diagnosis. Levels of C-reactive protein (CRP) were measured as an inflammatory marker. OS and survival date were obtained from medical charts and the Belgian national registry. Results: Patients with high HRV (SDNN > 20 msec) survived on average more than double the days (133.5) than those with low HRV (64.0). In a cox regression, higher initial HRV significantly predicted lower risk of death, independent of confounders. This relationship was mediated by CRP levels. In patients who lived only up to one month from diagnosis, HRV was unrelated to CRP, while in patients surviving longer, HRV was significantly inversely related to CRP (r = -0.20, p < 0.05). Conclusions: These results are in line with vagal nerve protection in a fatal cancer, and propose that the mechanism may involve neuroimmunomodulation. Future studies must test whether vagal nerve activation (behaviorally or electrically) may help patients with advanced cancers. CORRESPONDING AUTHOR: Prof. Y.G. Gidron, Free University of Brussels (VUB), Belgium, [email protected]

O156 IDENTIFICATION OF DISTINCT TRAJECTORIES OF BENEFIT FINDING AND PSYCHOLOGICAL SYMPTOMS IN CANCER PATIENTS L. Zhu1, A.V. Ranchor2, M. Van der Lee3, B. Garssen3, R.E. Stewart4, V.S. Helgeson5, R. Sanderman2, M.J. Schroevers2 1 Dept. of Health Psychology, University Medical Center Groningen, Netherlands, GRONINGEN, Nederland 2 Department of Health Psychology, University Medical Center Groningen, GRONINGEN, Nederland 3 Helen Dowling Institute, Centre for Psycho-Oncology, BILTHOVEN, Nederland 4 Department of Public Health, University Medical Center Groningen, GRONINGEN, Nederland 5 Department of Psychology, Carnegie Mellon University, PITTSBURGH, United States of America Introduction: This study aims to (1) identify distinct trajectories of benefit finding among cancer patients receiving psychological care; (2) examine whether socio-demographic, medical, and psychological care characteristics could differentiate these trajectories; (3) examine the associations of trajectories of benefit finding with psychological symptoms. Method: Participants were 241 cancer patients who were receiving psychological care at seven psycho-oncology institutions in the Netherlands. Data were collected before the start of psychological care, and 3 months and 9 months thereafter. A Latent Class Growth Modelling was applied to examine research questions in Mplus 7.1. Results: Five trajectories of benefit finding were identified: “high”(8%), “moderate”(29%), “low-large improvement”(8%), “low-small improvement”(39%), “very low”(16%). Socio-demographic, medical, and psychological care characteristics could not distinguish those trajectories. Distinct benefit finding trajectories had differential courses of depressive symptoms but not anxiety symptoms. Conclusions: Around half of cancer patients were able to maintain a high level of benefit finding or experience improvements on benefit finding during psychological care. Yet, there were still half of patients who remained low on benefit finding. Those who are able to improve or maintain a high level of benefit finding are more likely to improve depressive symptoms. This study suggests that there might be a vulnerable group of cancer patients, who were having difficulties in searching benefits from cancer and improving their psychological symptoms. Special attention should be given to this group in future psychological care. CORRESPONDING AUTHOR: L. Zhu, Dept. of Health Psychology, University Medical Center Groningen, Netherlands Nederland [email protected]

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

O157 POSTTRAUMATIC GROWTH AND BENEFIT-FINDING IN LUNG CANCER (LC) SURVIVORS: THE BENEFIT OF RURAL RESIDENCE? M. Andrykowski1, R. F. Steffens2 1 University of Kentucky, LEXINGTON, United States of America 2 Purdue University - North Central, WESTVILLE, United States of America Purpose: Prior research suggests cancer survivors residing in rural areas are at greater risk for experiencing distress compared to survivors residing in urban areas. Trauma theory suggests positive mental health outcomes (posttraumatic growth (PTG), benefit-finding (BF)) can follow traumatic or threatening events. No research, however, has examined whether rural residence is linked to reports of PTG and BF in cancer survivors. Methods: A population-based sample of 193 LC survivors (M=15 months post-diagnosis, range 11-22 months) was recruited using a statewide SEER cancer registry. Survivors completed a telephone interview and questionnaire packet consisting of measures of physical comorbidity and positive mental health outcomes (Posttraumatic Growth Inventory (PTGI), Benefit-Finding Questionnaire (BFQ)). Survivors were categorized into Rural (n-117) and Urban (n=76) groups based on county of residence using US Department of Agriculture Rural-Urban continuum codes. Results: Compared to Urban cancer survivors, Rural cancer survivors reported more growth on all 5 PTGI subscales with significant differences evident for the PTGI total score (p<.05) and PTGI-Spiritual Change subscale (p<.05). This pattern of differences in reports of PTG between the Rural and Urban survivor groups was retained after adjustment for covariates (education, physical comorbidity). In contrast, no rural-urban differences in BF were found in either unadjusted or adjusted analyses. Conclusion: Findings are consistent with contemporary theory and suggest for the first time that rural residence might be beneficial with regard to potential for PTG following cancer diagnosis and treatment. Furthermore, while PTG and BF are often considered interchangeable constructs, results suggest these may be distinct outcomes. CORRESPONDING AUTHOR: Prof. M. Andrykowski, University of Kentucky, United States of America, [email protected]

O158 DISTRESS, PROBLEMS, AND SUPPORTIVE CARE NEEDS AFTER AUTOLOGOUS OR ALLOGENEIC STEM CELL TRANSPLANTATION A.M.J. Braamse1, B.K.G. Van Meijel2, O.J. Visser.3, P.C. Huijgens.3, A.T.F. Beekman1, J. Dekker3 1 GGZ inGeest / VU University Medical Center, AMSTERDAM, Nederland 2 Inholland University of Applied Sciences, AMSTERDAM, Nederland 3 VU University Medical Center, AMSTERDAM, Nederland Introduction. Hematological malignancies and treatment with hematopoietic SCT are known to affect patients’ quality of life. The problem profile and care needs of this patient group need clarification, however. This study aimed to assess distress, problems and care needs after allo- or autoSCT, and to identify risk factors for distress, problems or care needs. Methods. In this cross-sectional study, patients treated with allo-SCT or auto-SCT for hematological malignancies completed the Distress Thermometer and Problem List. Three patient groups were created: 0-1, 1-2.5 and 2.5-5.5 years after transplantation. Results. After allo-SCT, distress and the number of problems tended to be lower with longer follow-up. After auto-SCT, distress was highest at 1-2.5 year(s). Patients mainly reported physical problems, followed by cognitive-emotional and practical problems. A minority reported care needs. Risk factors for distress as well as problems after allo-SCT

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

included younger age, shorter time after transplantation and GVHD. A risk factor for distress as well as problems after auto-SCT was the presence of comorbid diseases. Conclusion. Up to 5 years after auto-SCT or allo-SCT, patients continue to experience distress and problems. Judged by prevalence, physical problems are first priority in supportive care, followed by cognitiveemotional and practical problems. CORRESPONDING AUTHOR: A.M.J. Braamse, GGZ inGeest / VU University Medical Center, Nederland, [email protected]

O159 POSITIVE PSYCHOSOCIAL FACTORS AND A HEALTHY LIFESTYLE PROTECT AGAINST EARLY SIGNS OF CVD IN WOMEN I. Janssen1, L.H. Powell1, K.A. Matthews2, S.A. Everson-Rose3 1 Rush University Medical Center, CHICAGO, United States of America 2 University of Pittsburgh, PITTSBURGH, United States of America 3 University of Minnesota, MINNEAPOLIS, United States of America Introduction. Few studies have investigated protective factors for cardiovascular disease (CVD), the leading cause of death worldwide. Level of coronary calcium (CAC) as well as CAC progression predicts future CVD events. We investigated the effects of a healthy lifestyle as well as protective psychosocial factors on CAC and 2-year CAC progression in women. Methods. Subjects were 199 black and 345 white women (age 50.9±2.9 years). A healthy lifestyle score (HLS) was based on normal weight maintenance, not smoking, engaging in physical activity, and keeping a healthy diet. A psychosocial protective factor (PPF) was defined as the mean of standardized life engagement, optimism, and positive affect. We used partial proportional odds models to analyze baseline CAC (categorized as 0, 0-10, ≥10 Agatston units), and relative risk regression to predict CAC progression (ΔCAC≥10 Agatston units). All models adjusted for ethnicity and age. Results. Healthy lifestyle was inversely associated with both baseline CAC and CAC progression. For every 1 SD higher HLS, we observed significantly lower odds of higher CAC (OR=0.47; 95% CI, 0.39'0.57) and of CAC progression (OR=0.61; 95% CI, 0.45'0.81). The psychosocial factor was unrelated to baseline CAC but significantly inversely associated with CAC progression. For every 1 SD higher PPF, the OR of CAC progression was 0.65 (95% CI, 0.48'0.89). Interactions with ethnicity were not significant. Including both HLS and PPF and/or standard CVD risk factors attenuated estimates; both factors remained protective. Conclusion. Positive psychosocial factors and healthy lifestyle both offer protection in the early development of CVD for women. CORRESPONDING AUTHOR: I. Janssen, Rush University Medical Center, United States of America, [email protected]

O160 THE ASSOCIATION OF DEPRESSION AND ANGINA PECTORIS ACROSS 47 COUNTRIES: THE 2002 WORLD HEALTH SURVEY A. Loerbroks1, J.A. Bosch2, P.M.C. Mommersteeg3, R.M. Herr4, P. Angerer1, J. Li1 1 University of Düsseldorf, DÜSSELDORF, Germany 2 University of Amsterdam, AMSTERDAM, Nederland 3 Tilburg University, TILBURG, Nederland 4 Heidelberg University, MANNHEIM, Germany Introduction: Comorbid depression predicts poor health outcomes in patients with angina pectoris (AP). However, epidemiological data on the depression-AP comorbidity is limited and largely restricted to studies from Western countries, making generalizability to other regions

S51

uncertain. We aimed to provide additional epidemiological data for non-Western as well as Western countries. Methods: The present study used population-based data gathered in 47 countries from four continents (Africa, Asia, South America, and Europe) included in the cross-sectional 2002 WHO World Health Survey. Selfreported indicators of depression included: a) its diagnosis, b) its treatment, and c) seven symptom items to determine presence of a major depressive episode. Similarly, information on AP comprised a) a self-reported diagnosis, b) self-reported AP treatment, c) and a definition according to the WHO Rose questionnaire. In primary analyses, we operationalized depression or AP as positive if any of the respective indicators was present. Associations were estimated by multivariate logistic regression. Results: In the entire sample (n=213264), the odds of AP were more than doubled among those with depression (odds ratio [OR] =2.65, 95% confidence interval=2.41-2.92) versus those without depression. These positive associations were replicated across all continents. Likewise, meaningful associations (ORs ≥ 1.5) were observed in virtually all individual countries (46/47). Application of different operationalizations of depression and AP confirmed the above findings, both in the entire sample and in continent-specific analyses. Conclusion: Our study extends the current evidence accrued in Western populations to non-Western populations. The co-occurrence of AP and depression appears to represent a universal phenomenon. CORRESPONDING AUTHOR: Dr. A. Loerbroks, University of Düsseldorf, Germany, [email protected]

O161 DEPRESSION AND ITS ASSOCIATION WITH BLOOD PRESSURE DIPPING IN SOUTH-AMERICAN HISPANICS D.A. Chirinos1, M. Llabre1, B. Salinas2, H. Zea-Diaz2, J. Chirinos.3, J. Medina-Lezama2 1 University of Miami, MIAMI, United States of America 2 Santa Maria Catholic University, AREQUIPA, Peru 3 University of Pennsylvania, PHILADELPHIA, United States of America A blunted nocturnal blood pressure (BP) fall (nondipping pattern) has been shown to predict cardiovascular risk and prognosis. Little is known about the factors associated with BP dipping, particularly in minority populations. However, preliminary findings suggest psychosocial factors may play an important role. This study examined the relationship between depressive symptoms and nocturnal BP dipping in a community sample of SouthAmerican Hispanics. Participants included 122 adults (61.5% female) with available 24-hour ambulatory BP monitoring readings enrolled in the population-based PREVENCION study. Depressive symptoms were assessed with the Hospital Anxiety and Depression Scale. Systolic, diastolic BP and mean arterial pressure (MAP) dipping were examined as continuous outcome variables in separate multiple regression models. Participants had a mean age of 59 years (SD=14.81). Mean depression score was 6.33 (3.61). Mean daytime systolic BP, diastolic BP and MAP values were 129.74 (14.51), 76.16 (8.63) and 93.68 (9.85), respectively. Approximately 53.3% of individuals were classified as nondippers. Greater depressive symptoms were associated with reduced diastolic BP (β=-0.23, p=0.023) and MAP (β=-0.20, p=0.044), but not with systolic BP dipping after controlling for age, gender, waist circumference, low- and high-density lipoprotein cholesterol (HDL), smoking, alcohol consumption and use of hypertensive medications. Lower HDL was associated with reduced diastolic BP dipping (β=0.20, p=0.033). No other significant predictor of BP dipping was found. These findings suggest depressive symptoms may be important in understanding blunted nocturnal diastolic BP and MAP dipping patterns in South-American Hispanic adults. Future studies should further examine these associations with the use of prospective designs. CORRESPONDING AUTHOR: D.A. Chirinos, University of Miami, United States of America, [email protected]

S52

O162 SYMPTOM DIMENSIONS OF DEPRESSION FOLLOWING MYOCARDIAL INFARCTION AND CARDIOVASCULAR PROGNOSIS: AN INDIVIDUAL PATIENT DATA META-ANALYSIS R. de Miranda Azevedo, A.M. Roest, P. De Jonge UMCG, GRONINGEN, Nederland Introduction: Recent evidence suggests that somatic/affective but not cognitive/affective depressive symptoms predict adverse outcomes in patients with different types of heart disease. The aim of this study is to assess if there is a differential association between symptom dimensions of depression in predicting cardiovascular prognosis in myocardial infarction (MI) patients, by combining individual patient data (IPD) from studies available in the literature. Methods: We conducted an IPD meta-analysis of prognostic studies in MI patients. Prognostic endpoints were all-cause mortality and cardiac events. Symptom dimensions were generated from Beck’s Depression Inventory-1 using a multiple indicators multiple causes model. The association between symptom dimensions and prognosis was assessed using multilevel Cox regression. Results:Nine studies were included (n=7652). Two symptom dimensions were found: A somatic/affective and a cognitive/affective dimension. When predicting all-cause mortality, HRs were 1.30 (95%CI 1.22-1.38, P<.001) for somatic/affective symptoms and 1.07 (95%CI 0.99-1.15, P=.056) for cognitive/affective symptoms. For cardiovascular events, the HRs were 1.16 (95%CI 1.11-1.21, P<.001) and 1.01 (95%CI 0.931.02, P=.23), respectively. After adjustment for demographics, measures of heart disease severity, and health-related behaviors, the HRs decreased for the somatic/affective dimension in predicting all-cause mortality (HR: 1.19; 95%CI 1.07-1.32, P<.001) and cardiac events (HR: 1.18; 95%CI 1.08-1.28, P<.001). Cognitive/affective symptoms were not associated with any of the outcomes after adjustment. Conclusions: Only somatic/affective symptoms of depression were associated with cardiovascular prognosis following MI. The association decreased but remained significant after statistical adjustment was provided. Treatment strategies targeting somatic/affective symptoms in cardiac patients are warranted. CORRESPONDING AUTHOR: Mr. R. de Miranda Azevedo, UMCG, Nederland, [email protected]

O163 DEPRESSION AND SELF-CARE IN HEART FAILURE: A RANDOMIZED CONTROLLED TRIAL K.E. Freedland, R.M. Carney, B.C. Steinmeyer, E.H. Rubin, M.W. Rich Washington University, ST LOUIS, United States of America Introduction: Both depression and inadequate self-care are common in heart failure (HF). Previous trials have tested interventions for depression or for HF self-care, but none have treated both problems simultaneously. This RCT tested an integrated intervention for both problems. Method: Participants were randomly assigned to individual cognitive behavior therapy (CBT) or to usual care (UC); all received HF education. Randomization was stratified by antidepressant use at baseline. Treatment included up to 6 months of CBT plus maintenance contacts. Collaborative goal-setting was used to prioritize self-care and depression-related problems and to individualize treatment plans. CBT techniques including cognitive restructuring, behavioral activation, and problem-solving were used to address both sets of problems. Assessments were conducted at baseline, 3, 6, 9, and 12 months. The primary outcome was the BDI-2 score at 6 months. Secondary outcomes included HF self-care, anxiety, physical functioning, and quality of life. The treatment phase ended in September 2013; follow-ups will end in April 2014. Results: We enrolled 158 outpatients with HF and major depression (46% female, 37% minority, age 56+12 years, 58% NYHA Class I-II, 42% Class III, LVEF 39+

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

16%, 34% on an antidepressant at baseline). Most participants were moderately depressed (BDI-2, 30.2+8.5; HAM-D-17, 24.2+5.5). Conclusions: This presentation will report the trial’s primary and secondary outcomes, and an analysis of the relative contributions of CBT and antidepressant medications to the outcomes. CORRESPONDING AUTHOR: K.E. Freedland, Washington University, United States of America, [email protected]

O164 DRD2 SNP RS4586205, CHD METABOLIC RISK FACTORS IN NORMALS AND CLINICAL COURSE IN CHD PATIENTS R.B. Williams, A. Georgiades, B.H. Brummett, M.A. Babyak, R. Jiang, S.H. Shah, R.C. Becker, C. Haynes, M. Chryst-Ladd, D.M. Craig, E.R. Hauser, I.C. Siegler, R.S. Surwit, P.T. Costa Duke University Medical Center Durham, DURHAM, United States of America Introduction: The DRD2 gene is involved in addictive behaviors and possibly carbohydrate intake.. In this study we investigate the association of the DRD2 SNP RS4586205 to central adiposity, fasting glucose, insulin and HOMA-IR in normals as well as its effect on clinical course in CHD patients. Methods: SNPs were genotyped using Taqman platform in 330 healthy Caucasian men and women and 2198 Caucasian CHD patients from the CATHGEN biorepository. In normals, we assessed the association of RS4586205 to percent trunk fat measured through Dualenergy X-ray absorptiometry (DEXA)and to metabolic variables using a mixed model analysis controlling for relatedness, sex and age. In CHD patients we assessed RS4586205 association to death or MI using an additive model and adjusting for age, sex, hyperlipidemia, hypertension, diabetes, BMI, LVEF, # diseased vessels, smoking and subsequent CABG. Results: In the healthy sample minor allele homozygotes( G/G) had significantly (P=0.01) higher levels of trunk fat (28±10) compared to the G/T (22±8) and T/T allele (23±9) genotypes and significantly (P=0.02) higher levels of insulin resistance (1.2±.7) compared to the G/T (.82±.6) and T/T allele (.88±.6) genotypes. RS4586205 also showed a trend towards similar effects on time to MI/death in CATHGEN patients(p=.07) with G/G showing the shortest time (1000 days) vs G/T (1100 days) and T/T (1280 days). Conclusion: Normals with the G/G allele showed significantly higher levels of central adiposity and insulin resistance compared to those carrying the T allele. In addition, heart patients with the G/G genotype had the shortest time to death/MI. CORRESPONDING AUTHOR: Prof. R.B. Williams, Duke University Medical Center Durham, United States of America, [email protected]

O165 THE RELATIVE IMPORTANCE OF STRESSORS IN THE TREATMENT OF PATIENTS WIRTH BURN-OUT J.J. van Dixhoorn Centre for Breathing Therapy, AMERSFOORT, Nederland Introduction. The common view of burn-out is that it results from stressors which are coped with inadequately and that management of the stressors results in recovery. However, insufficient regulation of tension within the individual, preventing for instance effective sleep, may also be important. To what extent do stressors obstruct the effect of tension regulation? Goal. To assess the role of stressors in the effectiveness of a process-oriented and multimodal method of tension regulation in patients with burn-out. Design. In an observational study the presence of stressors was assessed by the therapist. Complaints were measured before and after treatment using a validated questionnaire for general distress state (GDS). Work status was assessed before and after treatment.

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

Results. Treatment of 145 subjects took on average 7 sessions. General Distress (GDS) improved significantly (p<0.001). Most (73%) patients did not exhibit stressors that prohibited tension regulation. In the absence of stressors, the percentage of subjects working full time rose van 20% to 57% and GDS normalized in two-thirds of them. When stressors prohibited tension regulation, the percentage working full time remained low (13%) and GDS normalized in only 20%. Interaction was significant (p<0.01). Conclusion. Assessment by the therapist of the role of stressors is a valid way to differentiate patients in whom stressors have little importance versus patients in whom they are of major importance for recovery. Tension regulation is a realistic treatment option, particularly in the first phase of treatment of patients with burn-out. CORRESPONDING AUTHOR: Dr. J.J. van Dixhoorn, Centre for Breathing Therapy, Nederland, [email protected]

O166 EFFECTS OF A BRIEF PSYCHOLOGICAL INTERVENTION ADDRESSING ILLNESS PERCEPTIONS IN SUBJECTS WITH NON-CARDIAC CHEST PAIN A.S. Martin1, S. Schroeder2 1 University of Wuppertal, WUPPERTAL, Germany 2 University of Erlangen-Nuernberg, ERLANGEN, Germany Introduction:Subjects with chest pain often continue to believe their symptoms being related to heart disease despite exclusion of cardiovascular diseases (non-cardiac chest pain NCCP), and these illness beliefs predicted symptom persistence (Martin & Schroeder 2012). To prevent the condition from becoming chronic, stepped care starting with an early, brief intervention addressing the subject’s illness beliefs may be useful. The aim of this randomized controlled trial was the development and evaluation of such a one session treatment.Methods: Subjects presenting chest pain at the clinic for cardiology were asked for health related worries prior and post angiography. 136 subjects with NCCP were then randomized to intervention or standard medical care SMC. Assessments of pain and of illness representations were at baseline, 1- and 6-months FU.Results: NCCP patients’ (60 % female, M = 53 years) health related worries and heart disease beliefs decreased with acknowledgement of medical test results, but both remained on a substantial level despite reassurance. Thus an intervention concerning illness representations seemed appropriate. Post ratings indicated the treatment to be credible and successful (e. g. > 85% agreement). FU-data of the RCT will be presented evaluating whether the brief intervention is also more effective than SMC in reducing the risk of pain persistence.Conclusion: Many NCCP subjects were not entirely reassured after standard provision of medical test results, and these may need enhanced care. The present psychological intervention seems to be acceptable to patients and feasible in specialist care. Strengths and limitations of the low-intensity treatment approach need to be discussed. CORRESPONDING AUTHOR: Prof. A.S. Martin, University of Wuppertal, Germany, [email protected]

O167 GUIDED SELF-INSTRUCTION FOR PATIENTS WITH IDIOPATHIC CHRONIC FATIGUE: A RANDOMISED CONTROLLED TRIAL A. L. Janse Expert Center for Chronic Fatigue, NIJMEGEN, Nederland Introduction We determined the efficacy of a cognitive behavioural minimal intervention consisting of a booklet with instructions combined with two weekly email contact for patients meeting the US Center for Disease Control (CDC) criteria for idiopathic chronic fatigue (ICF).

S53

Method A randomised controlled trial was performed in a specialist clinic center for the treatment of chronic fatigue in the Netherlands. All included patients suffered from severe and persistent fatigue but had less additional symptoms or less severe impairment than chronic fatigue syndrome (CFS) patients. In total, 100 patients with ICF were randomly allocated to the minimal intervention (N=50) or a waiting list control group (N=50). Treatment is based on the protocol of face to face CBT for CFS. Patients learn to change the fatigue-related perpetuating cognitions and behaviours. Primary outcome measures were fatigue severity (Checklist Individual Strength) and level of overall impairment (Sickness Impact Profile). Outcome measures were assessed prior to randomisation and following treatment or a waiting list control condition. Results Data of 84 patients was used in our main analyses. Following the minimal intervention, patients reported significantly less fatigue and impairment than patients in the waiting list. 53% of patients was clinically significant improved on fatigue compared to 16% percent in the control group. The number of additional symptoms and level of overall impairment did not moderate the effect of the intervention. Conclusions Patients meeting CDC criteria of idiopathic chronic fatigue can be treated effectively with a minimal intervention that was originally designed for patients with CFS. CORRESPONDING AUTHOR: MSc. A. L. Janse, Expert Center for Chronic Fatigue, Nederland, [email protected]

O168 PSYCHOGENIC GAIT DISORDER: A RANDOMIZED CONTROLLED TRIAL OF PHYSICAL REHABILITATION WITH ONE-YEAR FOLLOW-UP A. Jordbru1, L.M.S. Smedstad1, O.K. Klungsøyr2, E.W.M. Martinsen2 1 Vestfold Hospital Trust, TØNSBERG, Norway 2 Oslo University Hospital, OSLO, Norway INTRODUCTION: Psychogenic gait disorder, defined as loss of ability to walk without neurological aetiologies, has poor rehabilitation options that are well documented. Left untreated these patients have substantial and long-lasting dysfunction. The present study examined the effect of a 3-week inpatient rehabilitation programme compared with a waiting list control condition, and whether eventual gains were maintained at 1month and 1-year follow-up.METHODS: A cross-over design evaluated the effect of treatment, and a carry-over effect was considered as a longlasting treatment effect. Treatment consisted of adapted physical activity within a cognitive behavioural framework, and focused on offering an alternative explanation of symptoms, positively reinforcing normal gait and not reinforcing dysfunction. Patients: A total of 60 patients were recruited from neurological departments and were randomly assigned to immediate treatment (intervention) or treatment after 4 weeks (controls).RESULTS: Cross-over design revealed that the mean difference between treatment vs no treatment was 8.4 Functional Independence Measure units (p”<“0.001, 95% confidence interval 5.2-11.7), and 6.9 Functional Mobility Scale units (p”<“0.001, 95% confidence interval 5.58.3). Patients significantly improved their ability to walk and their quality of life after inpatient rehabilitation compared with the untreated control group. The improvements in gait were sustained at 1-month and 1-year follow-up.CONCLUSION: Substantial and lasting improvement can be achieved by inpatient rehabilitation with cognitive behavioural framework of patients with psychogenic gait, and the gains are maintained during follow-up. CORRESPONDING AUTHOR: Dr. A. Jordbru, Vestfold Hospital Trust, Norway, [email protected]

S54

08 - Functional I - Interventions in functional, somatic and somatoform disorders

O169 A C O G N I T I V E - B E H AV I O U R A L G R O U P T R E AT M E N T IMPROVED WORK ABILITY IN PATIENTS WITH SEVERE FUNCTIONAL SOMATIC SYNDROMES A. Schröder, E. Oernboel, J. Soendergaard, P. Fink Aarhus University Hospital, AARHUS C, Denmark Objective: Functional somatic syndromes (FSS) such as fibromyalgia, irritable bowel and chronic fatigue syndrome often disrupt employment and may lead to long-term dependence on social benefits and permanently reduced work ability. Cognitive-behavioural treatments (CBT) relief symptoms and improve functioning in FSS, but their effect on work ability is unclear. The aim of this study was to estimate the long-term effect of group CBT on work ability in patients with severe FSS. Methods: 120 Patients from a recently published randomised controlled trial comparing group CBT with enhanced usual care (EUC) were matched with 518 individuals from the general population. Complete data covering all public transfer payments were obtained from the Danish administrative DREAM database. We analysed the number of weeks per year without any transfer payments (i.e. weeks with complete selfsupport) from 10 years before to 3 years after treatment by means of random effects modelling allowing individual levels and slopes. Results: Compared with the general population, FSS patients showed a continuous decline in self-support, leading to markedly reduced work ability at trial entry. In the following years, EUC patients showed further decline, while the CBT group showed a trend for improvement (difference in mean change 4 (95 % CI 2-6) weeks per year), leading to an ultimate difference of 10 (95 % CI 3-18) weeks in the third year after treatment. Conclusion: A group CBT programme prevented further deterioration of work ability in severely impaired FSS patients. This may be associated with large savings in health-related and labour-market-related benefits. CORRESPONDING AUTHOR: A.S. Schröder, Aarhus University Hospital, Denmark, [email protected]

O170 INTERNET-BASED SELF-HELP FOR TINNITUS - MORE PROMISING WITH OR WITHOUT THERAPEUTIC SUPPORT? C. Weise1, J. Rheker1, G. Andersson2 1 Philipps-University Marburg, MARBURG, Germany 2 Linköping University, Dept. of Behavioural Sciences and Learning, LINKÖPING, Sweden Introduction: Internet-based cognitive-behavioural therapy (iCBT) has been shown to be effective in reducing distress in tinnitus sufferers. Most iCBT-approaches were designed as guided self-help, that is, patients receive therapeutic support while working on self-help material. To date, the role of support has rarely been studied, however, it is assumed to positively influence treatment outcome. Thus, the aim of our study was to investigate the impact of therapeutic support on treatment efficacy. Methods: We randomized 112 tinnitus sufferers to one of two groups which both received the same iCBT-programme. Participants assigned to the non-support group worked on the programme without additional support, whereas participants of the support group could request therapeutic support whenever needed. We assessed tinnitus distress, associated symptoms, treatment credibility, and satisfaction pre and/or post treatment. Results: Before the treatment, patients rated the support condition as more credible than the one without support. Post-treatment, both groups showed significant improvements in tinnitus distress (time effect:

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

F(1,110)=117.99; p<.001) and depressive symptoms (F(1,110)=15.42; p<.001). However, there was no significant group difference. Also in regard to treatment satisfaction both groups did not differ. Conclusion: The internet-based self-help was effective in reducing tinnitus distress. Interestingly, despite higher initial credibility ratings for the supported iCBT, tinnitus distress improved equally in the support and the non-support condition. Future research should thus further investigate the impact of therapeutic support on treatment efficacy, but also on treatment satisfaction, compliance, or dropout rates. In addition, possible mediators and moderators of the efficacy of internet-based treatments should be identified. CORRESPONDING AUTHOR: Dr. C. Weise, Philipps-University Marburg, Germany, [email protected]

O171 ASSESSING CHANGES IN THE ADOPTION OF OBESITY PREVENTION PRACTICES IN AUSTRALIAN CHILDCARE SERVICES S.L. Yoong1, L. Wolfenden1, M. Finch2, N. Nathan2, C. Lecathelinais2, P. Dodds1, J. Jones1, J. Wiggers2 1 University of Newcastle, CALLAGHAN, Australia 2 Hunter New England Population Health, WALLSEND, Australia Background Despite substantial government investment to improve healthy eating and physical activity environments of childcare services, it is unknown whether services have changed their practices across time. The study aims to describe changes in proportion of Australian childcare services that have adopted recommended healthy eating and physical activity promoting practices between 2006 and 2013, and assess whether adoption varied by socioeconomic (SES) and locality of services in 2013. Methods A randomly selected sample of authorised supervisors from childcare services located in New South Wales, Australia completed a telephone survey in 2006, 2009, 2010 and 2013. Supervisors reported on service adoption of six healthy eating and physical activity practices related to policy, staff training, scheduled time for activity, restricting screen time opportunities and serving non-sweetened beverages. To determine the change between 2006 and 2013 as well overall trend, repeated measures logistic regression in a GEE framework was undertaken. Results There was a significant increase in the prevalence of services adopting all but one practice between 2006 and 2013. 27% of services reported adopting all six practices in 2013, a significant increase from 3.7% (2006). There were no differences in the proportion that had adopted each practice by locality and SES in 2013. Conclusions These findings suggest that government investment in obesity prevention programs can equitably improve childcare service’s healthy eating and physical activity environments. Further investment to support adoption and monitoring of a more comprehensive set of practices is required to improve obesity prevention efforts in this setting CORRESPONDING AUTHOR: S.L. Yoong, University of Newcastle, Australia, serene.yoong hnehealth.nsw.gov.au

O172 DUAL PROCESS MODEL OF PHYSICIANS' AND NURSE'S TYPE 2 DIABETES-RELATED PRESCRIBING, ADVISING AND EXAMINING BEHAVIOURS J. Presseau 1 , M. Johnston 2 , J. Francis 3 , T. Heponiemi 4 , M. Elovainio4, M. Eccles1, N. Steen1, S. Hrisos1, E. Stamp1, J. Grimshaw5, G. Hawthorne1, F. Sniehotta1

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216 1

Newcastle University, NEWCASTLE UPON TYNE, United Kingdom University of Aberdeen, ABERDEEN, United Kingdom 3 City University, LONDON, United Kingdom 4 National Institute for Health and Welfare, HELSINKI, Finland 5 Ottawa Hospital Research Institute, OTTAWA, Canada 2

Introduction Providing high quality primary care for people with type 2 diabetes involves skilled, reflective decision-making, and regular performance in clinical environments replete with prompts/cues which may trigger clinicians’ behaviour. We tested a dual process model involving both impulsive and reflective processes applied to six different clinician behaviours. Methods Physicians and nurses in 99 primary care practices in the United Kingdom completed postal questionnaires focused on six behaviours: prescribing for blood pressure (N=335), prescribing for glycaemic control (N=288), providing advice for self-management (N=332), providing advice for weight management (N=417), providing diabetes-related education (N=346) and examining feet (N=218). Questionnaires included measures of intention, action planning and coping planning (sequential motivational and volitional reflective process) and automaticity (impulsive process). Self-reported behaviours at 12 months was the primary outcome. Results Both reflective and impulsive processes predicted prescribing behaviours, weight advice, and examining feet. Intention was a direct predictor of 5/6 behaviours and an indirect predictor of behaviour via action or coping planning for 4/6 behaviours. Automaticity predicted 4/6 clinician behaviours alongside the reflective process. Conclusion A dual process model involving a reflective process alongside a parallel impulsive process provides a better understanding of the clinician behaviours involved in managing type 2 diabetes. The results could inform clinician behaviour change interventions to improve healthcare quality. CORRESPONDING AUTHOR: Dr. J. Presseau, Newcastle University, United Kingdom, [email protected]

O173 EFFECTIVENESS OF A PRACTICE CHANGE INTERVENTION IN INCREASING MENTAL HEALTH CLINICIAN PROVISION OF PREVENTIVE CARE K. Bartlem1, J. Bowman1, M. Freund1, P. Wye1, K. McElwaine1, J. Knight2, P. McElduff1, K. Gillham2, J. Wiggers2 1 University of Newcastle, CALLAGHAN, Australia 2 Population Health, Hunter New England Local Health District, WALLSEND, Australia INTRODUCTION: Addressing the physical health needs of people with a mental illness is an increasingly recognised priority; however, mental health services provide sub-optimal levels of preventive care to address modifiable health risk behaviours. A study was undertaken to evaluate the effectiveness of a clinical practice change intervention to increase the provision of preventive care within community mental health (CMH) settings. METHODS: A 12 month multi-strategic intervention was implemented sequentially across two groups of CMH services to increase clinician provision of assessment, advice, and referral/follow-up for smoking, inadequate nutrition, harmful alcohol consumption and physical inactivity. Weekly cross-sectional telephone interviews with clients receiving care from CMH services measured the receipt of preventive care from 6 months prior to intervention implementation in the first group of services until 6 months post intervention in the second group of services (36 months total). RESULTS: Preliminary results (n=1777) demonstrate significant (p<.05) increases from baseline to follow-up in assessment for smoking (74%-

S55

81%), nutrition (24%-34%), alcohol (76%-81%), and all behaviours (18%-29%); and in advice for nutrition (22%-28%). Significant increases in referral/follow-up were identified for: talking about telephone support services for nutrition (7%-16%) and physical activity (6%-16%); arranging a referral to a telephone support service for nutrition (1%-6%) and physical activity (1%-5%), and providing any type of referral/follow-up option for all behaviours (11%-16%). CONCLUSIONS: The intervention was successful in small increases in preventive care provision for some behaviours and care elements. Further intervention is required to improve preventive care delivery within CMH services. CORRESPONDING AUTHOR: Ms K. Bartlem, University of Newcastle, Australia, [email protected]

O174 INNOVATIVE EVIDENCE-BASED INTERVENTION TO INCREASE NICOTINE DEPENDENCE TREATMENT IN MENTAL HEALTH HOSPITALS B. Wye1, J.H. Wiggers2, J.A. Bowman1, J. Constable2 1 University of Newcastle, CALLAGHAN, Australia 2 Hunter New England Health, NEWCASTLE, Australia INTRODUCTION: There remains a significant disparity in smoking prevalence between the general adult population (16%) and among inpatients in mental health hospitals (50%), with minimal treatment provided despite the existence of guidelines. This project aimed to translate policy into practice by increasing the provision of nicotine dependence treatment for mental health inpatients who smoke. METHODS: An evidence-based, multi-strategic intervention was implemented to increase clinician adoption of guidelines in 7 mental health inpatient units in NSW, Australia. Outcomes were measured from medical record audit of all patients discharged from mental health services, with pre-post and time series analysis used to determine results. RESULTS: From 4370 mental health hospital discharges over 20 months, 44% of patients were identified as smokers. Assessment of patient smoking increased from 32% to 48% (p=.000) and quit advice increased from 6% to 23% (p=.000). Offer of any nicotine replacement therapy increased from 59% to 74% (p=.001), and prescription of nicotine replacement therapy increased from 54% to 65% (p=.000). Nicotine dependence treatment on discharge record increased from 14% to 32% (p=.000). Time series analysis indicates a continued increase in the provision of treatment after completion of the intervention phase. CONCLUSIONS: Such evidence-based, multi-strategic interventions can significantly increase assessment and treatment of nicotine dependence within mental health inpatient settings, with change sustainable over time. CORRESPONDING AUTHOR: Dr. B. Wye, University of Newcastle, Australia, [email protected]

O175 ACCESS TO INTERVENE. ETHNOGRAPHIC EXPLORATIONS OF INEQUALITIES IN HEALTH AS AN OBJECT FOR PUBLIC INTERVENTION M.T.H. Høybye1, M.H. Rod2 1 Silkeborg Regional Hospital, SILKEBORG, Denmark 2 National Institute of Public Health, University of Southern Denmark, COPENHAGEN, Denmark Introduction This study explores how Danish municipality public health officers seek to gain access to intervene and promote health in the population, targeting social inequalities in health. Focusing on the ways in which health inequalities are constructed as an object for municipal intervention, this

S56

raises questions on how “target groups”are defined, identified and conceptualized, and which values are promoted by particular practices, policies and interventions. Method and material The study is based on ethnographic fieldwork in four Danish municipalities using methods of participant observation, interviews and document analysis. We mainly followed the work of public health officers investigating the forms of practice, norms and modes of reasoning that persist as ethical elements in the municipality health promotion efforts. Results Our ethnographic material indicates that, policies to reduce health inequalities are employed by different means across the municipalities, drawing on different modes of reasoning about marginalized population groups. Representing particular understandings of access to health, some seek to promote health through broad, uniform interventions for the general population, based on arguments of the risk for stigmatization, while others promote equality in health through targeted interventions. Conclusions This study found that health promotion intervention targeting vulnerable populations and inequality in health is to a great extent relational work dependent on the personal and political agility of the individual health promotion officers to produce access to intervene in target populations. CORRESPONDING AUTHOR: Dr. M.T.H. Høybye, Silkeborg Regional Hospital, Denmark, [email protected]

O176 BUILDING CAPACITY AMONG URBAN POOR: HOW DOES COMPUTER LITERACY TRAINING INFORM IMPLEMENTATION OF E-HEALTH INTERVENTIONS? C. Chan1, R. McCloud2, K. Viswanath2 1 Monash University Malaysia, KUALA LUMPUR, Malaysia 2 Dana-Farber Cancer Institute, Harvard School of Public Health, BOSTON, United States of America Introduction:People of low socioeconomic position (SEP) are often constrained by limited resources to seek health information. Improving computer and Internet literacy among populations of low socioeconomic position can potentially maximize exposure and enhance their access to health information on the Internet. This study evaluated an Internet literacy training of a community-based e-Health intervention among novice computer users and identified the challenges and benefits associated with the implementation of such training among urban poor. Methods:A computer literacy training program was designed with nine 2-hour modules featuring basic computer and Internet knowledge,(e.g.,email, Internet search and networking). Different health topics were also introduced. Adults (N=155) were recruited from adult literacy centres in New England, in three waves. Professional teachers for adult learners facilitated the training sessions. Participants were provided with free computers and internet access. Results:Overall attendance was between 78.4 to 87.5%. The percentage of lateness decreased as training progressed. Self-perceived competence in computer literacy was satisfactory. Content analyses identified the major reasons for absence in class were work (14%), health (34%), logistics (10%) and other competing committments (26%). Generally, participants were satisfied and highly positive with the classes. Many preferred longer sessions for practice and acknowledged the health topics being introduced to them. Time spent on internet has increased. Conclusion:The findings suggest that a computer literacy intervention is acceptable to people of low SES. Future scale-up of similar intervention can address the barriers identified for implementing e-Health training programs to improve competence of novice computer users and build capacity among urban poor. CORRESPONDING AUTHOR: Dr. C. Chan, Monash University Malaysia, Malaysia, [email protected]

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

S177 POSITIVE OUTCOME EFFECTS OF WEB-BASED INTERVENTIONS FOR PEOPLE WITH CHRONIC MEDICAL CONDITIONS: EVIDENCE AND CHALLENGES J.B. Prins1, J.L.M. Boselie2, H.R. Trompetter3, S.W. Van den Berg1, A.L. Stanton4 1 Radboud university medical center, NIJMEGEN, The Netherlands 2 Maastricht University, MAASTRICHT, The Netherlands 3 University of Twente, ENSCHEDE, The Netherlands 4 UCLA, LOS ANGELS, United States of America Traditionally, trials in behavioral medicine have solely focused on improving negative patient outcomes (such as symptoms of depression, distress, or anxiety). In the past decade, interventions have emerged to increase positive outcomes (such as acceptance, positive mood, optimism, benefit finding, psychological flexibility or empowerment). Bringing together experts in the innovative field of eHealth research, this symposium addresses the question how web-based interventions can increase positive outcome variables in cancer survivors and people with chronic pain. First scientific results from randomized controlled trials (RCTs) will be presented to provide a state-of-the-art overview of the current evidence on web-based interventions targeting positive outcome variables in addition to negative outcomes. So far, the results are mixed and inconclusive. Therefore, methodological challenges and recommendations will be discussed to further the research field of increasing positive outcomes through eHealth. First presented will be results of the internet-based intervention for chronic pain patients ‘Happy without pain’, which is based on positive psychology. Also targeting chronic pain patients, the second presentation will shed light on positive mediators and moderators of change in a large RCT of a web-based self-help intervention based on Acceptance & Commitment Therapy. The third presentation will gain insight whether it is possible to increase positive adjustment outcomes in early breast cancer survivors through an unguided self-management website (the BREATH trial). Last, ‘Project Connect Online’ presents the findings of an intervention trial to harness social support and promote sharing of the cancer experience. CORRESPONDING AUTHOR: Prof. J.B. Prins, Radboud university medical center, The Netherlands, [email protected]

S178 ‘'HAPPY DESPITE PAIN'': A POSITIVE PSYCHOLOGY INTERNETBASED INTERVENTION FOR CHRONIC PAIN PATIENTS J.J.L.M Boselie, L.M. Vancleef, E. Smeets, M.L. Peters Maastricht University, MAASTRICHT, Nederland Introduction: Chronic pain may not only lead to difficulties in physical and emotional functioning but also in cognitive functioning. Optimism, the general tendency to expect that good things will happen in the future, and positive affectivity may help chronic pain patients to cope with these challenges. In two successive studies we tested the effects of a 7-week internet-based positive psychology intervention on self-reported physical and emotional functioning (study 1 & 2) and objectively assessed cognitive functioning in patients with chronic musculoskeletal pain (study 2). Both studies also assessed potential mediators. Method: The positive psychology intervention consisted of four different positive psychology exercises: self-compassion training, gratitude journaling, savoring and best possible self-visualization. In Study 1, patients were randomized to the positive intervention (n=114), an internet-based cognitive behavioral intervention (n=112) or a wait-list condition (n=50). Study 2 randomized patients to the positive intervention (n=50) or a wait-list condition (n=30).

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

Both studies assessed pain interference, depression, wellbeing and disability pre- and post-intervention. Optimism and positive affect were measured as potential mediators. Study 2 additionally included a cognitive task battery pre- and post-intervention. Results: The positive psychology intervention increased wellbeing and decreased pain interference and depression with effects lasting up to 6 months. Changes in optimism and positive affect were identified as potential mediators. Results of the cognitive tasks will be presented at the conference. Conclusion: These results demonstrate that increasing optimism and positive affect by an online positive psychology intervention may lead to improved functioning in patients with chronic pain. CORRESPONDING AUTHOR: MSc J.J.L.M Boselie, Maastricht, University, Nederland, [email protected]

S179 POSITIVE MEDIATORS AND MODERATORS OF CHANGE IN WEB-BASED ACCEPTANCE & COMMITMENT THERAPY FOR CHRONIC PAIN H.R. Trompetter, E. T. Bohlmeijer, M. M. Veehof, J.-P. Fox, K. M. G. Schreurs University of Twente, ENSCHEDE, Nederland Introduction. Chronic pain affects a large number of people in society. As complete pain removal is not always a realistic treatment perspective, psychological strategies focus on effective pain management. One of these strategies is Acceptance & Commitment Therapy (ACT). We previously evaluated a web-based self-help intervention based on ACT compared to ‘Expressive Writing’ and a Waiting List Control condition. In addition to symptom-based outcomes, psychological flexibility (process), values-based living (process) and positive mental health (outcome) were assessed as positive factors. We now explored possible mediating and moderating mechanisms of change. Methods. 238 participants were recruited through advertisements in newspapers and websites. Measurements were taken at baseline, posttreatment (3 months) and follow-up (6 months). Mediating mechanisms were evaluated by method of Preacher & Hayes (2004). Post-hoc, moderating mechanisms were explored using linear regression models. Results. As previously reported, small to moderate effects were found for ACT in improving pain interference in daily life (primary outcome) and other symptom-based outcomes. Also, psychological flexibility improved, but not values-based living and positive mental health. Outcomes of present analyses revealed moderate effect sizes for psychological flexibility as a mediator of changes in pain interference. Additionally, positive mental health proved to be the only moderator of change. Conclusions. Web-based ACT may be effective in improving functioning of chronic pain patients, but not for those experiencing low positive mental health. Additionally, psychological flexibility is an important resilience process in learning to live with pain. Possible reasons for absence of effect in other positive outcomes are discussed. CORRESPONDING AUTHOR: MSc. H.R. Trompetter, University of Twente, Nederland, [email protected]

S180 POSITIVE ADJUSTMENT OUTCOMES OF A SELF-MANAGEMENT WEBSITE FOR EARLY BREAST CANCER SURVIVORS: THE BREATH RCT S.W. van den Berg, M.F.M. Gielissen, W.T.A. Van der Graaf, P.B. Ottevanger, J. Prins Radboud university medical center, NIJMEGEN, Nederland INTRODUCTION Easy-accessible and standardized care to support psychological post-treatment adjustment is lacking for early breast cancer

S57

(BCS). The BREATH [BREAst cancer e-healTH] is a generic unguided CBT-based self-management website that aims to reduce negative adjustment variables and increase positive adjustment variables. METHOD Aim of this multicenter randomized controlled trial was to evaluate the superiority of BREATH compared to care as usual (CAU). BREATH was a fully-automated website with 16 generic weekly modules covering universal post-treatment topics. BCS between 2 and 4 months post-treatment were randomized (ratio 1:1) and assessed before randomization at baseline (T1) and 4 months (T2) follow-up. Primary outcomes of distress (Symptom Checklist-90) and empowerment (Cancer Empowerment Questionnaire) were tested on statistical (ANCOVA) and clinical effect (Reliable Change Index). Secondary positive adjustment variables included self-efficacy, quality of life, remoralization, mastery, fulfillment, re-evaluation, new ways of living and valuing life. All outcomes were tested intention to treat (T1-T2). RESULTS 70 BCS were analyzed for BREATH and 80 for CAU. Greater distress reduction was found in BREATH compared to CAU, being statistically (p=0.020) and clinically (p=0.039) relevant with a small-tomedium effect size (d=0.33). No empowerment effect was found (p =0.336; d=0.01).On 3 out of 8 secondary positive adjustment outcomes, BREATH showed significant 4-months improvements (self-efficacy, remoralization, new ways of living; p<0.05). CONCLUSIONS The unguided self-management website BREATH was only effective in reducing negative adjustment outcomes after primary breast cancer treatment. Increasing positive adjustment through intervention is a new research area and still accommodates multiple methodological and theoretical challenges. CORRESPONDING AUTHOR: S.W. van den Berg, Radboud University Medical Center, Nederland, [email protected]

S181 PROJECT CONNECT ONLINE: RCT OF AN INTERVENTION TO HARNESS SOCIAL SUPPORT AND PROMOTE SHARING OF THE CANCER EXPERIENCE Annette L. Stanton PhD, Professor, Psychology & Psychiatry/Biobehavioral Sciences Member, Jonsson Comprehensive Cancer Center, UCLA, Los Angeles, USA PURPOSE: Evidence suggests that expressing emotions related to cancer and receiving support can promote psychological and physical health in women diagnosed with breast cancer. We conducted a randomized controlled trial of an intervention, Project Connect Online (PCO), for breast cancer patients to create personal websites to chronicle their experience and communicate with their social network. PATIENTS AND METHODS: Women (N = 88) diagnosed with breast cancer were randomly assigned to participate in a three-hour workshop for hands-on creation of personal websites with a follow-up call to facilitate website use, or to a waiting-list control. Assessed prior to randomization and six months following the intervention, dependent variables included depressive symptoms, positive and negative mood, cancer-related intrusive thoughts, and perceived cancer-related benefits in life appreciation and strengthened relationships. Putative mediators of PCO's efficacy also were assessed. RESULTS: Relative to control participants, women randomized to PCO evidenced significant benefit six months later on depressive symptoms, positive mood, and life appreciation, as well as on hypothesized mediators (i.e., coping self-confidence, loneliness, social support from friends). Women currently undergoing medical treatment for cancer benefitted significantly more from PCO on specific outcomes than did women not receiving treatment. Change over six months in coping self-confidence, loneliness, and social support from friends mediated the relationship between PCO and improvement in particular outcomes. Expression of positive and negative emotions and description of cancer-related benefits

S58

and treatment side effects on the websites were potential 'active ingredients' of PCO. Website visitors were most commonly female friends who valued the websites as a way to connect emotionally with participants and receive information about their health. CONCLUSIONS: Findings suggest the promise of an intervention to facilitate the ability of women diagnosed with breast cancer to chronicle their experience and communicate with their social network via the internet. Findings also highlight modifiable factors that contributed to the efficacy of PCO. Ways in which the intervention can be enhanced are addressed. CORRESPONDING AUTHOR: A.L. Stanton, UCLA, United States of America, [email protected]

S182 SEVERE FATIGUE: POSSIBLE DETERMINANTS AND ITS MANAGEMENT R. Moss-Morris1, H. Knoop2, J. Wiborg2, A. Wearden3 1 Institute of Psychiatry, LONDON, United Kingdom 2 Expert centre for chronic fatigue, Radboud University Medical Center,, NIJMEGEN, Nederland 3 Manchester Centre for Health Psychology, University of Manchester, MANCHESTER, United Kingdom Fatigue is an often occurring symptom that can substantially interfere with daily life. It is not only highly prevalent in chronic medical conditions but is also often reported after stressful life events. Research so far suggests that fatigue is multifactorially determined. Biological, cognitivebehavioural factors and the social environment can influence fatigue symptoms. In this symposium research will be presented that illustrates this and provides more insight into the relationship between the different factors and fatigue. Of the possible biological determinants, changes in cortisol levels as a marker of HPA functioning has been proposed to cause or maintain fatigue in a wide range of conditions. In this symposium the role of cortisol in persistent fatigue in patients with multiple sclerosis is discussed based on a longitudinal study using multiple measures of salivary cortisol over four days. The contribution of psychological factors in the development and maintenance of fatigue will be illustrated with a prospective longitudinal study that investigated the prevalence and predictors of fatigue in military personnel after deployment. The role of the social environment in the persistence of fatigue in patients suffering from chronic fatigue syndrome will be discussed, using two studies exploring different types of significant other responses and patient outcomes of fatigue and disability. Finally, it will be shown how models of the possible determinants of fatigue can aid its management. Data will presented of a large randomized controlled trial testing the efficacy of group therapy in chronic fatigue syndrome based on a cognitive behavioural model of fatigue. CORRESPONDING AUTHOR: Prof. R. Moss-Morris, Institute of Psychiatry, United Kingdom, [email protected]

S183 THE EFFICACY OF COGNITIVE BEHAVIOUR GROUP THERAPY FOR PATIENTS WITH CHRONIC FATIGUE SYNDROME H. Knoop, G. Bleijenberg, J. Wiborg Radboud University Medical Centre, NIJMEGEN, Nederland Introduction: Chronic fatigue syndrome (CFS) is characterised by severe and disabling fatigue which persists at least six months and which is not the result of a medical condition or ongoing exertion. According to the cognitivebehavioural model of CFS, fatigue related cognitions and behaviours perpetuate the symptoms. Aim of the present study was to investigate

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

the efficacy of cognitive behaviour group therapy (CBGT) for CFS in a randomised controlled trial. Methods: CBGT was delivered in a small (4 patients, 1 therapist) and large (8 patients, 2 therapists) group format. All patients met U.S. centre for diseases criteria for CFS and indicated that they were willing to participate in group therapy. Patients were assessed prior and subsequent to the intervention or control condition. Primary outcome measures were fatigue severity (Checklist Individual Strength) and the level of disabilities (Sickness Impact Profile and SF-36 physical functioning). Our analysis was based on 196 patients of which 122 received treatment in either the small (n=66) or the large group format (n=66) and 64 patients were allocated to the waiting list condition. Results: Preliminary analysis showed that the intervention lead to a significant reduction in fatigue and disabilities compared to the waiting list condition. No significant differences in efficacy were found between the small and the large group format on the primary outcome measures. Discussion CBT delivered in groups is effective in reducing fatigue severity and disabilities in patients with CFS. CORRESPONDING AUTHOR: Dr. H. Knoop, Radboud University Medical Centre, Nederland, [email protected]

S184 FATIGUE IN AFGHANISTAN VETERANS: COURSE AND MECHANISMS J.F. Wiborg1, A.R. Rademaker2, E. Geuze2, E. Vermetten2, H. Knoop1 1 Radboud University Medical Center, NIJMEGEN, Nederland 2 Research Centre Military Mental Healthcare, UTRECHT, Nederland Introduction: A number of veterans suffer from medically unexplained symptoms after deployment to combat areas. Some studies suggest that persistent fatigue is a major problem of these veterans. However, most of these studies are based on cross-sectional data and do not allow for inferences about the course of fatigue. In addition, pre-deployment data are often not available in these studies to determine which variables may predispose military personnel to develop persistent fatigue after deployment. Methods: The prospective research in stress during military operations (PRISMO) offers data from Dutch Afghanistan veterans from before to several years after deployment. In total, 1032 soldiers were assessed at baseline. Preliminary data of this group are presented based on linear mixed models analysis using SPSS. Results: A significant increase in fatigue severity from baseline to two years after deployment was found. The largest rate of increase in fatigue was observed between baseline assessment and directly after deployment. Significant differences were also found in fatigue trajectories between the individual soldiers. Baseline scores of childhood abuse and personality helped to predict differences between individual fatigue trajectories, while controlling for demographic variables and concurrent levels of psychopathology. Discussion: Based on this prospective study of Afghanistan veterans, fatigue seems to be a problem in deployed military personnel which persists over time. Childhood abuse and personality factors may predispose soldiers to develop persistent fatigue in the face of deployment. CORRESPONDING AUTHOR: Dr. J.F. Wiborg, Radboud University Medical Center, Nederland, [email protected]

S185 SIGNIFICANT OTHERS' RESPONSES TO SYMPTOMS IN CHRONIC FATIGUE SYNDROME (CFS) A. Wearden, R.J. Band, C. Barrowclough, R. Emsley, M. Machin University of Manchester, MANCHESTER, United Kingdom

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

Introduction: Emerging evidence suggests that significant other responses to patients’ symptoms may affect patient illness outcomes in CFS. Here we report on two studies exploring the relationships between different types of significant other responses and patient outcomes of depression, fatigue and disability. Methods: In study 1, 23 patients with CFS and their significant others (dyads) were recruited from specialist CFS services. Using experience sampling methods, all participants reported affect, contact with each other and significant other responses (negative and solicitous) to the patient. Patients reported symptom severity, disability and activity management strategies. In study 2, 55 significant others of patients undergoing treatment in specialist centres were interviewed at start of treatment, to determine significant other criticism and emotional over -involvement. Patient fatigue, disability and depression were assessed at start of treatment and 6 months later. Results: In study 1, negative significant other responses were associated with increased patient symptom severity at the same momentary assessment. There was evidence of a mediating role of concurrent distress on symptom severity. Solicitous responses were associated with reduced patient activity and disability reported at the same momentary assessment, but these effects were transitory. In study 2, patients with critical relatives had smaller improvements in fatigue over treatment, and this effect was mediated by increased depressive symptoms in these patients. Conclusions: Significant other responses are important influences on the day-to-day experience of CFS, and are predictive of patient response to treatment. These insights will inform the development of a family-based intervention for fatigue. CORRESPONDING AUTHOR: Prof. A. Wearden, University of Manchester, United Kingdom, [email protected]

S186 SALIVARY CORTISOL AND FATIGUE IN EVERYDAY LIFE IN RELAPSING-REMITTING MULTIPLE SCLEROSIS (RRMS) R. Moss-Morris1, D. Powell2, C. Liossi2, W. Schlotz3 1 Institute of Psychiatry, LONDON, United Kingdom 2 University of Southampton, SOUTHAMPTON, United Kingdom 3 University of Regensburg, REGENSBURG, Germany Introduction: Fatigue is a common distressing symptom in RRMS. Given the regulatory role of cortisol in energy metabolism and immune functioning, dysfunction of the hypothalamic-pituitary-adrenal (HPA) axis may be implicated in fatigue experience in MS. This study sought to examine the relationship between fatigue and cortisol secretory activity in everyday life in RRMS at baseline and at 6-months follow-up. Method: 38 individuals with RRMS and 38 healthy controls provided salivary samples over four consecutive weekdays whilst undertaking usual daily activities. Samples were prompted by handheld device upon awakening and 30 and 45 minutes afterwards to determine the cortisol awakening response and on six further quasi-randomised occasions between 10.00am and 8.00pm to compute the diurnal cortisol slope. Fatigue severity, stress and mood were measured at baseline and 6-month followup. Analysis was by multilevel modelling. Results: Greater fatigue severity in RRMS was associated with lower awakening cortisol levels but larger cortisol increases post-awakening. This association remained after accounting for employment status, sleep quality, chronic stress, depressive symptoms, and neurological disability. No cortisol marker was associated with change in fatigue after 6-months in either group. Conclusion: Cortisol changes were related to fatigue in RRMS and may represent a neuroinflammatory response. Causality could not be inferred possibly due to minimal change overtime and drop out which minimised power. The larger MS fatigue cortisol awakening response is in contrast to the blunted cortisol response reported in chronic fatigue syndrome.

S59

Whilst psychological correlates of fatigue are similar across these two conditions, the biological underpinnings may differ. CORRESPONDING AUTHOR: Prof. R. Moss-Morris, Institute of Psychiatry, United Kingdom, [email protected]

S187 TASKS, EMOTIONS, STRESS, AND CARDIOVASCULAR REGULATION M. Tuomisto1, A.T. Ginty2, R.A. Hackett3, D.W. Johnston4 1 University of Tampere, TAMPERE, Finland 2 University of Birmingham, BIRMINGHAM, United Kingdom 3 University College London, LONDON, United Kingdom 4 University of Aberdeen, ABERDEEN, United Kingdom The researchers in this symposium have aimed to study both the basis of cardiovascular regulation (awareness of cardiovascular reactions, longterm stability of emotional and other self-reported behaviour) and determinants of it (perceived and acute stress, stress reactivity, type 2 diabetes, nursing tasks, demand, tense arousal). Annie Ginty will discuss the extent to which people are aware of their cardiovascular reactions to exposure to psychological stress. She will present two studies investigating the relationship between perceived stress, perceptions of their cardiovascular stress reactions, and actual cardiovascular stress reactivity. Ruth Hackett will present a study comparing cardiovascular responses to acute stress in people with type 2 diabetes and healthy controls. She will discuss how attenuated responses to stress may reflect chronic allostatic load and contribute to increased risk of cardiovascular disease in people with type 2 diabetes. Derek Johnston uses ambulatory data from a study of real time stress in ward nurses to examine the sources of heart rate increases during the working day by comparing events (nursing tasks), perception of work environment (demand) and emotion (tense arousal). He shows that the nature of the nursing task and perceptions of work are the main determinants of heart rate increase, with emotion having variable effects. Martti Tuomisto will present the results from a cardiovascular study in which emotional and stress responses were monitored with an interval of 10 years using psychometric scales and diaries. He shows that these responses in general were stable. He then discusses the importance of this for cardiovascular regulation. CORRESPONDING AUTHOR: Prof. M. Tuomisto, University of Tampere, Finland, [email protected]

S188 DO YOU LISTEN TO YOUR HEAD OR YOUR HEART: DISASSOCIATION BETWEEN PSYCHOLOGICAL AND PHYSIOLOGICAL SYSTEMS A. Ginty, S.E. Williams, R.C. Brindle, A. Bibbey, A.C. Phillips, D. Carroll University of Birmingham, BIRMINGHAM, United Kingdom Introduction: Stress elicits psychological and physiological responses. Few studies have examined individuals’ awareness of their cardiovascular reactions to stress and how perceptions of stress and anxiety contribute to this awareness. Methods Study 1: Undergraduates (N=272) were exposed to a 10-minute mental arithmetic stress task. Heart rate (HR) was measured at baseline and during stress. Participants graphed their heart rate perceptions (HRP) and stress perceptions (SP) across time. Area under the curve (AUC) was calculated for HR during baseline and stress exposure, HRP and SP. Results Study 1: There was a positive association between HRP AUC and SP AUC (p < .001), but no relationship between HR AUC and PHR AUC or SP AUC. Method Study 2: High school students (N=180) completed the anxiety subscale of the Hospital Anxiety

S60

and Depression Scale (HADS) and undertook the stress task, and indicated somatic anxiety levels before and during the task. Cardiovascular activity was recorded continuously and stress reactivity for blood pressure, HR, and HR variability was the difference between stress and baseline averages. Results Study 2: There was a positive association between HADS anxiety and somatic anxiety experienced before (p = .004) and during the task (p = .014), but no association between cardiovascular reactivity and HADS anxiety or somatic anxiety before and during the task. Conclusions: Higher reported stress and anxiety, both trait and state, is associated with perceptions of greater cardiovascular reactivity, but not with actual cardiovascular changes indicating a disassociation between psychological and physiological systems during stress exposure. CORRESPONDING AUTHOR: A. Ginty, University of Birmingham, United Kingdom, [email protected]

S189 DYSREGUALTED CARDIOVASCULAR RESPONSES TO LABORATORY STRESS IN PEOPLE WITH TYPE 2 DIABETES R. Hackett1, M. Hamer1, R. La Marca2, A. Steptoe1 1 University College London, LONDON, United Kingdom 2 University of Zurich, ZURICH, Switzerland Introduction:Cardiovascular disease (CVD) is a major cause of mortality and morbidity in type 2 diabetes (T2DM). Epidemiological and clinical evidence suggests that psychosocial stress is involved in both conditions, but the biological pathways involved remain unclear. In healthy participants exaggerated cardiovascular responses to laboratory stress are associated with an increased risk of future CVD, but blunted responses may be indicative of chronic allostatic load. The aim of this study was to compare cardiovascular responses to acute stress in participants with T2DM and healthy controls. Methods: 140 T2DM subjects were age, sex and income-matched with 280 healthy middle age participants from the Whitehall II cohort. Systolic blood pressure (SBP), diastolic blood pressure (DBP), cardiac index (CI) and heart rate variability (HRV) were monitored using a Finometer at baseline, during two stressful behavioural tasks and at 45 and 75 minutes later. Results: Cardiovascular measures increased in both groups in response to stress. T2DM status was associated with blunted SBP (p= 0.001) responses to psychological stress independently of body mass index (BMI) and smoking. T2DM subjects had reduced CI (p = 0.026) responsivity additionally controlling for diabetic medication and beta-blockers. HRV responses were also attenuated in the T2DM sample (p = .011) after adjustment for covariates. No associations were observed for DBP. Conclusions:T2DM is associated with reduced cardiovascular stress reactivity. Attenuated responses to stress may reflect chronic allostatic load and contribute to increased risk of CVD in people with T2DM.

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

aroused by the event. These possibilities were investigated in a real time study of occupational stress in nurses. Method. 100 nurses completed PDA based assessments of nursing tasks (based on the WOMBAT classification of nursing tasks,) perception of work demand (5 items), and tense arousal (4 items), every 90 minutes over two 12 hour shifts. Heart rate (HR) and activity were measured continuously using the Actiheart system. Analysis was based on Multilevel modelling (MLwiN) testing for fixed and random effects of the 3 predictors on nurses’ heart rate. Results. After allowance for physical activity and HR changes over time, HR was higher when nurses were involved in direct patient care than all other activities combined, perceived the work as demanding and reported higher tense arousal (i.e. the fixed effects significant). These effects varied considerably between people (i.e. random effects significant). When all 3 predictors were analysed simultaneously the fixed and random effects of direct care and perceived demand remained significant. However only the random effect of tense arousal remained significant. Conclusion.In the real world HR elevations can be predicted by all 3 possible determinants; event, perception and emotion. In this study they had separate additive effects, although the effect of emotion was much reduced and confined to a subset of nurses when both event (nursing task) and perception of the event (demand) were allowed for CORRESPONDING AUTHOR: Prof. D.W. Johnston, University of Aberdeen, United Kingdom, [email protected]

S191 STABILITY OF EMOTIONAL AND OTHER SELF-REPORTED DAILY RESPONSES IN A CARDIOVASCULAR STUDY M. Tuomisto University of Tampere, TAMPERE, Finland

S190 REAL TIME DETERMINANTS OF HEART RATE INCREASES IN NURSES: TASK,PERCEIVED DEMAND OR EMOTION? D.W. Johnston1, C. Bell1, M. Johnston1, M. Jones2, I. Ricketts1, B. Farquharson3, P. Schofield4 1 University of Aberdeen, ABERDEEN, United Kingdom 2 University of Dundee, DUNDEE, United Kingdom 3 Universsity of Stirling, STIRLING, United Kingdom 4 Greenwich University, LONDON, United Kingdom

Introduction: Predictive models in behavioural medicine include as an axiom that behaviour, especially emotional behaviour, is stable or recurrent over long time intervals. However, it is not well-established how stable these behavioural characteristics are in the context of health studies. The purpose of this study was to investigate to what extent different psychometric scales often used in cardiovascular behavioural medicine or self-reported responses in diaries used in ecological momentary assessment show stability in a cardiovascular study with healthy participants. Methods: In the Tampere Ambulatory Hypertension Study, 124 middle-aged men completed several questionnaires measuring anxiety, anger and hostility, emotional expression and reactivity. In addition, 95 men monitored their emotions and other self-reported variables in preformatted diaries. Ten years later, the diary study was repeated during ambulatory BP monitoring for the same participants. Results: Emotional responses in general were stable: no significant differences in the means of the same emotions were found between the two time points. Moreover, the order of the particular participants’ means remained the same between the two time points. The correlations between the questionnaire scores of different emotions and the corresponding emotions in the diaries were quite high and significant in general. The emotions monitored in the diaries having the best correlations with questionnaires measuring the same emotions were irritation and tension. Conclusions: The general axiom in behavioural medicine that emotional behaviour is stable and recurrent seems to be valid. This makes it wellfounded to use emotional assessment as the basis for research in behavioural medicine.

Introduction. The cardiovascular response to challenging events may be determined by the event, the perception of the event or the emotion

CORRESPONDING AUTHOR: Prof. M. Tuomisto, University of Tampere, Finland, [email protected]

CORRESPONDING AUTHOR: Ms R. Hackett, University College London, United Kingdom, [email protected]

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

S192 FINANCIAL INCENTIVES FOR SMOKING CESSATION DURING PREGNANCY: A SYMPOSIUM L. Bauld1, T. Marteau2, P. Hoddinott3, D. Tappin4, I. Berlin5 1 Uk Centre for Tobacco and Alcohol Studies, STIRLING, United Kingdom 2 Behaviour and Health Research Unit, University of Cambridge, CAMBRIDGE, United Kingdom 3 School of Nursing and Midwifery, University of Stirling, STIRLING, United Kingdom 4 Paediatric Epidemiology and Child Health Unit, University of Glasgow, GLASGOW, United Kingdom 5 Hôpital Pitié-Salpêtrière, Faculté de médecine Université Paris, PARIS, France Introduction: Financial incentives to promote behaviour change are controversial and evidence of efficacy for complex behaviours remains limited.This symposium presents new findings on this topic for smoking cessation during pregnancy. Methods: a phase II RCT (n=600), a cohort study (n=615) and a crosssectional public opinion survey (n=1144). All 3 studies were conducted in different parts of the UK, and findings will be placed in an international context by an expert discussant from France. Results: In the RCT, vouchers added to standard care (behavioural therapy + NRT) increased the cotinine validated quit rate from 9% to 22% at the end of pregnancy. In the cohort study, 8% who received vouchers plus standard care were cotinine validated quitters at the end of pregnancy and 4% at 3 months post-partum. There was some evidence of cheating or 'gaming' in both studies. The survey found that UK public opinion on incentives (for either smoking cessation or breastfeeding in pregnancy) was mixed, with negative opinions more prevalent amongst women and those with lower educational attainment, and support higher amongst those of child-bearing age. Conclusion: These three studies provide important new evidence to suggest that financial incentives, when added to services to help women stop smoking in pregnancy, can increase quit rates but the public support for these types of interventions remains limited. These findings will be discussed in the context of previous studies in the USA and of ongoing cessation in pregnancy research in France and elsewhere in Europe. CORRESPONDING AUTHOR: Prof. L. Bauld, Uk Centre for Tobacco and Alcohol Studies, United Kingdom, [email protected]

S193 FINANCIAL INCENTIVES FOR SMOKING CESSATION IN PREGNANCY: SINGLE ARM INTERVENTION STUDY ASSESSING CESSATION AND 'GAMING' T. Marteau1, D. Ierfino2, J. Hirst3, T. Jones4, E. Mantzari1, P. Aveyard5 1 University of Cambridge, CAMBRIDGE, United Kingdom 2 Kings College London, LONDON, United Kingdom 3 Derbyshire County Council, MATLOCK, United Kingdom 4 Derbyshire Community Health Services, CHESTERFIELD, United Kingdom 5 University of Oxford, OXFORD, United Kingdom Introduction. Financial incentives were the single most effective intervention for smoking cessation in pregnancy in a recent Cochrane Review, but based on few, small trials in the USA using 7-day point prevalence cessation measures. This study estimates (a) prolonged cessation in unselected English pregnant smokers and (b) “gaming”
S61

Russell criteria for cessation; gaming was assessed using urinary and salivary cotinine (or anabasine) measures on five occasions. Results. 39% (239/615) of smokers enrolled on the scheme, of whom 60% (143/239) made a quit attempt. Of the whole cohort, 8% (48/615) were quit at delivery and 4% (25/615) at six months post-partum. There was no evidence that women gamed to enter the scheme. 4% (10/239) of those enrolled gamed on one or more occasions to gain vouchers. Conclusions. The offer of an incentive scheme in an unselected English cohort of pregnant smokers was associated with prolonged cessation rates comparable to those reported in USA trials. Rates of gaming were arguably insufficiently high to invalidate the use of such schemes. CORRESPONDING AUTHOR: Prof. T. Marteau, University of Cambridge, United Kingdom, [email protected]

S194 PUBLIC ACCEPTABILITY OF FINANCIAL INCENTIVES FOR SMOKING CESSATION IN PREGNANCY AND BREASTFEEDING P. Hoddinott1, H. Morgan2, G. MacLennan2, K. Sewel3, G. Thomson4, L. Bauld1, D. Yi5, A. Ludbrook2, M.K. Campbell2 1 University of Stirling, STIRLING, United Kingdom 2 University of Aberdeen, ABERDEEN, United Kingdom 3 Ipsos MORI, EDINBURGH, United Kingdom 4 University of Central Lancashire, PRESTON, United Kingdom 5 Kings College, LONDON, United Kingdom Introduction. Financial incentives are being promoted for lifestyle behaviour change, but are controversial. We surveyed public attitudes about incentives for smoking cessation and breastfeeding in the UK as part of a mixed methods study to inform incentive trial design. Methods. Seven promising incentive strategies were identified from evidence syntheses and qualitative interview data: (i) validated smoking cessation in pregnancy and (ii) after birth; (iii) for a smoke-free home; (iv) for proven breastfeeding; (v) a free breast pump; (vi) payments to health services for reaching smoking cessation in pregnancy targets and (vii) breastfeeding targets. We used area quota sampling, home-administered computer-assisted questionnaires, agreement measured on a 5 point scale and multivariable ordered logit models. Results. Amongst 1144 UK adults agreement with incentives was mixed (ranging from 34-46%). Mean agreement score was highest for a free breast pump; and lowest for incentives for smoking abstinence after birth. More women than men disagreed with shopping vouchers. Those with lower levels of education disagreed more with smoking cessation incentives and a breast pump. Those aged 44 or under agreed more with all incentives compared to those aged 65 and over, particularly provider targets for smoking cessation. Non-white ethnic groups agreed particularly with breastfeeding incentives. Current smokers and respondents with breastfed children agreed with providing vouchers for the respective behaviours. Up to £40 per month vouchers were acceptable (>85%). Conclusion. Women and the less educated are more likely to disagree, but those of child-bearing age to agree, with incentives which are designed for their benefit. CORRESPONDING AUTHOR: Prof. P. Hoddinott, University of Stirling, United Kingdom, [email protected]

S195 CESSATION IN PREGNANCY INCENTIVES TRIAL (CPIT): A PHASE II TRIAL IN SCOTLAND L. Bauld1, D. Tappin2, L. Sinclair1 1 Uk Centre for Tobacco and Alcohol Studies, STIRLING, United Kingdom 2 Paediatric Epidemiology and Child Health Unit, University of Glasgow, GLASGOW, United Kingdom

S62

Introduction. Small trials in the USA have shown that financial incentives for smoking cessation in pregnancy can be effective, but further evidence is required. In the UK, smoking cessation services for pregnant women exist, but uptake is low and outcomes can be poor. This study aimed to explore whether adding incentives to services is effective, acceptable and cost-effective. Methods. A phase II individually randomised controlled trial comparing standard cessation care (behavioural support plus NRT) with standard care plus financial voucher incentives. Participants (n=612) were pregnant smokers identified at maternity booking in Glasgow, Scotland. Vouchers up to the value of £400 were available if clients attended counselling, set a quit date and abstinent from smoking at 4 weeks, 12 weeks and the end of pregnancy. The primary outcome was self-reported abstinence at the end of pregnancy verified by saliva cotinine. Results. The offer of vouchers increased the quit rate from 9% to 22%. Lost to follow up was equal between groups (15%). ‘Gaming’ of cotinine tests was seen in 20% of control and 22% of intervention quitters. A process evaluation found that incentives were acceptable to women and professionals. Participants were representative of all pregnant smokers in Glasgow. Conclusion. Financial incentives may double the quit rate when added to smoking cessation in pregnancy services. Standard operating procedures for this exploratory study will support a future definitive trial in the UK. Discussant for the symposium will be Dr Ivan Berlin from the University of Paris. CORRESPONDING AUTHOR: Prof. L. Bauld, Uk Centre for Tobacco and Alcohol Studies, United Kingdom, [email protected]

O196 SIDE-EFFECTS OF FUNDRAISING MESSAGES CONCERNING SERIOUS ILLNESSES ON HEALTH BEHAVIORS A. Dijkstra University of Groningen, GRONINGEN, Nederland Introduction. To motivate the general audience to donate money to illness charity organizations, two main strategies are used. The first strategy is to portray the illness as serious (threatening information). The second strategy is to convince the audience that there is good progress in developing effective treatments (response efficacy information). When it comes to donation, these types of information may have desired effects. However, when it comes to health behaviors that are related to the illness the campaign is about, the response efficacy information may work detrimental: The more effective the available treatments, the less threat the illness represents. This will lower the motivation to engage in healthy behavior. Methods. This effect was studied in a field experiment (N=216) in which the threat and the response efficacy information each were manipulated (in a poster-format of the Heart Foundation) in a 2 x 2 design, with the intention to donate and the intention to engage in different health behaviors as dependent variables. Results. The combination of threat and response efficacy information led to the highest level of intention to donate. However, moderation analyses showed that in different subgroups of respondents the intention to engage in healthy behaviors was significantly lowered by the response efficacy information. For example, response efficacy information significantly lowered health behavior intentions in people with high self-esteem but also in men. Conclusions. These findings suggest that fundraising campaigns that use response efficacy information may have serious sideeffects: They may be responsible for the maintenance of unhealthy behaviors. CORRESPONDING AUTHOR: Prof. Dr. A. Dijkstra, University of Groningen, Nederland, [email protected]

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

O197 SHORT-TERM EFFECTIVENESS OF A COMPUTER TAILORED GAMIFICATION APPROACH TO DECREASE ALCOHOL USE IN DUTCH ADOLESCENTS A.F. Jander, R. Crutzen, L.A.G. Mercken, H. De Vries Maastricht University, MAASTRICHT, Nederland Introduction. Alcohol use in adolescents is associated with many negative consequences like getting into fights, sexual risk behavior, (illicit) drug use and brain damage. In the Netherlands alcohol use in adolescents is particularly high, compared to other European countries. A possibility to decrease alcohol use are computer tailoring (CT) strategies to provide people with relevant and personalized feedback and advice. The current study has combined CT strategies with a game aimed at decreasing binge drinking in Dutch adolescents aged 16 to 18 years. Methods. In the beginning of 2014 a RCT started in which 40 schools participated with approximately 3000 adolescents. In the experimental condition adolescents filled out an online baseline questionnaire assessing demographic variables, drinking behavior and motivational determinants (modeling, social norm, perceived pressure, self-efficacy). After that, they started playing a CT game about binge drinking. During the game, adolescents received text messages on a virtual cell phone asking them to answer questions about motivational determinants. Their answers were used to generate highly personalized computer tailored feedback to prevent binge drinking in identified difficult situations. The control condition only filled out the baseline and post-test questionnaire. Results. Changes in drinking behavior and motivational determinants from baseline to 3 months follow-up will be tested. Subgroup analyses describe the differences in the effectiveness and appreciation of the computer tailored game based on gender and SES status. Conclusions .The results will give an insight into the short term effectiveness of the game and potential differences between different groups of adolescents. CORRESPONDING AUTHOR: MSc A.F. Jander, Maastricht University, Nederland, [email protected]

O198 PRODUCING AVATAR IDENTIFICATION IN ANIMATION VIDEO EDUCATION ON SPINAL ANESTHESIA IN PATIENTS UNDERGOING ELECTIVE SURGERY M.T.H. Høybye1, M. Vesterby2, L.B. Jørgensen1 1 Silkeborg Regional Hospital, SILKEBORG, Denmark 2 INNO-X Healthcare, Institute of Clinical Medicine, University of Aarhus, AARHUS, Denmark Introduction. In relation to a current clinical implementation of animation video in patient education this study addressed the knowledge gap on the mechanisms and meanings of animated educational interactions with patients undergoing total hip arthroplasty (THA) surgery. Methods and material. By comparing two versions of a 2D animated video on spinal anesthesia, the study tested the patient - avatar identification within two different narrative models: a general narrator and a patient narrator. To explore the perspectives of THA patients on the difference we employed complementary qualitative methods of interviews and ethnographic observation. All 15 patients had undergone THA surgery within 5-15 months of the interview. Results. The animated presentation of the spinal anesthesia procedure in both narrative forms was immediatedly recognized by all participants as reflecting their experience of the procedure. The avatar gender did not affect this identification. We found no preference for either the general or patient narrative form. The majority of patients prefering the general narrator highlighted the fact based clarity and the professionality of the narrator as a preference. The majority of patients prefering the patient narrator highlighted the engaging presence of a co-patient as reassuring in information delivery.

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

A few patients had strong emotional reactions to the animated illustration of spinal anesthesia procedures. Conclusion. This study supports the potential of using animation video in patient education as a didactic model for qualifying patient behavior. Animation video creates a high degree of identification with the information provided that may work of reduce pre-surgical anxiety. CORRESPONDING AUTHOR: Dr. M.T.H. Høybye, Silkeborg Regional Hospital, Denmark, [email protected]

O199 EFFECTIVENESS OF INDIVIDUALIZED COUNSELLING LETTERS FOSTERING REDUCED SMOKING: A GENERAL POPULATION BASED RANDOMIZED CONTROLLED TRIAL C. Meyer1, S. Haug2, H.J. Rumpf3, U. John1, S. Ulbricht1 1 Institute of Social Medicine and Prevention, GREIFSWALD, Germany 2 Swiss Research Institute for Public Health and Addiction ISGF, ZÜRICH, Switzerland 3 University of Lübeck, LÜBECK, Germany Introduction: Previous research approved the efficacy of individually tailored counselling letters for smoking cessation in unselected populations of smokers. However, the degree of individualization of the intervention content and the intervention effects were smaller in individuals initially not intending to quit compared to those in advanced stages of change. The aim of the present study was to test a computer based expertsystem intervention targeting the reduction of cigarettes smoked per day. Methods: We drew a nationwide sample from the German general population (age 18 to 64) by using a random digit dialing procedure. Among the identified smokers (n=2776) 50% consented to take part in the intervention trial. Participants reporting daily smoking of at least 10 cigarettes per day and no intention to quit smoking within the next six months (n=1462) were randomized to one of two intervention groups receiving three tailored letters within six months focusing either on smoking reduction (RG) or cessation (CG) and a minimal assessment only control group (MG). Follow-up data on self-reported 7-day point abstinence were collected 12 and 24 months after study inclusion for 81% and 76% of the participants, respectively. Results: Abstinence rates were 5.9% (MG), 7.4% (CG) and 9.3% (RG) at month 12 and 8.1% (MG), 13.0% (CG), and 14.3% (RG) at month 24. GEE analyses revealed that both interventions were effective across time (CG: OR = 1.6, p=.045; RG: OR=1.8 p=.007) compared to MG. Conclusion: Smoking reduction should be considered as an alternative intervention goal for smokers unable or unwilling to quit. CORRESPONDING AUTHOR: Dr. C. Meyer, Institute of Social Medicine and Prevention, Germany, [email protected]

O200 CAN BOWEL CANCER SCREENING DECISION STAGE PREDICT AN INDIVIDUAL'S INFORMATION NEEDS? T. Zajac1, I. Flight1, C. Wilson2 1 CSIRO, ADELAIDE, Australia 2 Flinders University, ADELAIDE, Australia Background: Tailoring online content to suit individuals is a suggested means of improving consumer health-protective behaviours. However, implementing tailoring is complex and recent results suggest it may not be effective. Targeting aims instead to present information to groups of users based on common characteristics. This study investigated whether bowel cancer screening decision-stage could be a suitable variable on which to base targeting techniques. Method: N=1,387 participants aged 40 to 78 years of age completed an online questionnaire assessing decision stage with regards to bowel cancer screening, as well as

S63

preferences for different categories of information currently presented in the Australian Bowel Cancer Screening Consumer Information Booklet. Results: The distribution of individuals over PAPM decision stages was: never heard of screening (7%); not considered screening (10%); doesn’t want to screen (1%); unsure about screening (14%); and wants to screen (68%). Decision stage predicted preferences for certain categories of information. For example, individuals who had never heard of screening were twice as likely to preference ‘information about bowel cancer’ over ‘information about screening’ (χ2(3) = 14.14, p=.003; OR=1.98), and also ‘information about screening’ over ‘information about completing a screening test’ (χ2(3) = 19.05, p<.001; OR=1.73). Demographic variables had no effect on preferences. Conclusions: This study provides evidence that individuals of different decision stages prefer to engage with different types of information regarding bowel cancer. Thus, it might prove beneficial to use this variable to refine the content delivered to users of webbased and paper-based educational resources.

O201 EFFICACY OF PSYCHOLOGICAL INTERVENTIONS TO REDUCE DISTRESS IN PATIENTS UNDERGOING INVASIVE CARDIAC PROCEDURES: A META-ANALYSIS C. Protogerou1, N.F. Fleeman2, K.D. Dwan2, M.R. Richardson2, Y.D. Dündar2, R.D. Dickson2 1 University of Liverpool, LIVERPOOL, United Kingdom 2 University of Liverpool, Institute of Psychology, Health and Society., LIVERPOOL, United Kingdom Introduction: Psychological interventions are offered to cardiac patients to counteract distress associated with invasive procedures undertaken. This meta-analysis (a) assessed the efficacy of psychological interventions to reduce distress in patients undergoing cardiac procedures/events; (b) explored study-level predictors of the impact of these interventions. Method: We systematically searched 3 electronic databases for Randomised Controlled Trial (RCT) interventions, published as of July 2013. A random effects meta-analysis was conducted for depression and anxiety outcomes, at post-treatment and follow-up, with further subgroup analyses for: type of usual care; cardiac procedure undertaken; intervention duration; risk of bias; and facilitator type. Results: Twenty studies met the inclusion criteria (2181 participants), 16 reported depression and 12 reported anxiety. Relative to controls, interventions reduced depression and anxiety at post-treatment (SMD -0.66, 95% CI: -1.14 to -0.19 and SMD -0.40, 95% CI: -0.71 to -0.09 respectively) and at follow-up for anxiety (SMD -0.29, 95% CI: -0.47 to -0.10). Reduction in post-treatment depression varied for different types of usual care control (p=0.05), but differences were not significant between these subgroups at follow-up, or for anxiety at any point. Depression reduction differed as a function of type of procedure/event (post-treatment; p=<0.00001, follow-up; p=0.02) and intervention duration (post treatment; p=0.03, follow-up; p=0.0002). Reduction in anxiety differed between different facilitator subgroups (post treatment; p=0.009, follow-up; p=0.02). Kappa for risk of bias was 0.72. Conclusions: RCT psychological interventions had a significant protective effect, reducing depression and anxiety in cardiac patients. Studylevel characteristics moderated the outcomes. CORRESPONDING AUTHOR: Dr. C. Protogerou, University of Liverpool, United Kingdom, [email protected]

O202 EFFECTIVENESS OF A COMMUNITY INTERVENTION ON CHILDREN'S NUTRITION, PHYSICAL ACTIVITY AND WEIGHT J.H. Wiggers1, L. Wolfenden1, K. Gillham2, L. Hardy3, A. Milat4, A.C. Bell5 1 University of Newcastle, WALLSEND, Australia

S64 2

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

Hunter New England Local Health District, WALLSEND, Australia University of Sydney, SYDNEY, Australia 4 NSW Ministry of Health, SYDNEY, Australia 5 Deakin University, GEELONG, Australia

Conclusions: Changes in the level of physical exercise, weight and motivational variables are seen as trends. Findings do not confirm that the intervention with focus on action planning has the potential to evoke substantial differences in self-regulatory cognitions

Introduction.The primary prevention of obesity among children remains an international priority. Limited evidence exists regarding the effectiveness of population level interventions in increasing children’s consumption of healthy food and physical activity and reducing obesity. The aim of this paper is to report on the impact of a community intervention to improve such risks and reduce overweight and obesity among children. Method. A 5 year intervention (Good for Kids Good for Life) was implemented that sought to build the capacity of 7 community settings to promote healthy nutrition and physical activity among children aged 312 years: all schools, child care services, sports clubs, community service organisations, GPs, health services, and Aboriginal health organisations, supported by social marketing. Evaluation of the intervention was undertaken by pre-post collection of parent/self-report and anthropometric data from a random sample of 4,000 children. Results. Significant improvements in the following behaviours were observed: consumption of sweetened drinks and water, fruit and vegetables, and time spent in physical activity. No effect was found regarding the consumption of energy dense, nutrient poor foods. The prevalence of overweight and obesity declined by approximately 1% per year in the intervention region compared to no change in the state. A significant 7.4% reduction in prevalence of overweight and obesity was found for girls aged 5-10 years. Conclusion.The findings suggest that a community intervention has the potential to improve the prevalence of healthy nutrition and physical activity among children, and to reduce the prevalence of overweight and obesity.

CORRESPONDING AUTHOR: W.G. Goehner, Catholic University of Applied Sciences, Germany, [email protected]

3

CORRESPONDING AUTHOR: Prof. J.H. Wiggers, University of Newcastle, Australia, [email protected]

O203 QUASI-EXPERIMENTAL PILOT TRIAL: 6-MONTHS FOLLOW-UP OF A MULTIDISCIPLINARY INTERVENTION PROGRAM FOR OBESE ADULTS W.G. Goehner1, P.W. Wagner2 1 Catholic University of Applied Sciences, FREIBURG I.BR., Germany 2 University of Leipzig, LEIPZIG, Germany Introduction: We evaluated the effects of a one-year multidisciplinary group intervention for obese adults (contents: psychological training, information about healthy diet, physical exercise program). Special focus lies on the volitional aspects of behaviour change (action and coping planning); these are known to be the key components of health behaviour change. Method: Obese persons were invited by public calls to participate in the intervention and were consecutively assigned to the intervention and control group (IG; CG). Both groups received the same intervention, except that we addressed for the IG, in additional to motivational aspects, also volitional aspects of behaviour change. Questionnaires (psychological and behavioural variables, weight) were completed at baseline and 6 months follow-up. Results: 35 obese persons participated either in the IG (n=18) or CG (n=17). At 6 months follow up, both groups lost substantial weight (BMI t1-t2 IG: -2.03, CG -2.11, group differences n.s.). Results yielded that both groups are physically active at t1, but only the IG can increase physical activity at t2 on a descriptive basis (IG t1 2.16hrs/week, t2: 3.23hrs/week; CG t1: 2.23hrs/week, t2: 2.36hrs/week). Between the measurements, the IG showed more increasing motivation in comparison to the CG (group differences n.s.). With regard to action and coping planning both groups show similar values at both measurements.

O204 PRE-INTERVENTION EATING SELF-EFFICACY PREDICTS WEIGHT LOSS IN ASIAN AND PACIFIC ISLANDERS C.K.M. Townsend1, J.K. Kaholokula1, G. Zhang1, D. Paloma2, R. Miyamoto3, I. Kauka4, K. Braun5 1 John A Burns School of Medicine, HONOLULU, United States of America 2 Queens Health System, HONOLULU, United States of America 3 I Ola Lahui, HONOLULU, United States of America 4 Queen Liliuokalani Children's Center, HONOLULU, United States of America 5 University of Hawaii at Manoa, HONOLULU, United States of America Introduction: Asian/Pacific Islanders (APIs) have a high prevalence of overweight/obesity (BMI ≥ 25). Studies found modest weight-loss in overweight/obese APIs who participate in behavioral lifestyle interventions. Identifying behavioral factors that lead to greater weight-loss in APIs can improve the effectiveness of lifestyle interventions. The effects of eating and exercise self-efficacy, locus of weight control, physical activity/functioning, fat intake, and community and family support on weight-loss in overweight/obese APIs were examined. Methods: Data was from 106 APIs who completed a 12-month worksite-based lifestyle intervention. Weight, physical functioning (i.e., 6-min. walk test; 6MWT), and self-report measures of the aforementioned factors were administered at baseline, 3-month, and 12-month. Baseline characteristics: mean BMI=32.2, mean weight=84.2kg, mean age=46.5, 86% female, and 73% college graduates. Results: From baseline to 12-month, significant (p<.01) improvements in weight [-1.9kg], 6MWT [106.8ft], exercise level, fat intake, eating self-efficacy, family support, and external locus of weight control were found. In a multiple regression model, only eating self-efficacy scores at baseline were associated [t=-3.76; p<0.01] with weight-loss at 12-month post-intervention, after adjusting for sociodemographics, baseline weight, and intervention and worksite type. Decrease in fat intake was associated with improvements in eating selfefficacy from baseline to 12-month (r=-0.25, p<.02), but not with weightloss. Conclusion: For overweight/obese APIs, greater eating self-efficacy at the outset of a behavioral lifestyle intervention is associated with greater weight-loss. Because fat intake was not associated with weightloss, it may be that eating self-efficacy leads to dietary improvements other than fat intake, such as decrease in daily caloric intake. CORRESPONDING AUTHOR: C.K.Mm Townsend, John A Burns School of Medicine, United States of America, [email protected]

O205 IDENTIFYING BEHAVIORAL PATTERNS OF RESPONSE TO LIFESTYLE INTERVENTION: A LATENT CLASS ANALYSIS S. Fitzpatrick1, J.W. Coughlin2, L.J. Appel2, C. Tyson3, V.J. Stevens4, G.J. Jerome2, A. Dalcin2, P.J. Brantley5, F. Hill-Briggs2 1 Rush University Medical Center, CHICAGO, United States of America 2 Johns Hopkins School of Medicine, BALTIMORE, United States of America 3 Duke University, DURHAM, United States of America 4 Kaiser Permanente Northwest, PORTLAND, United States of America 5 Pennington Biomedical Research Center, BATON ROUGE, United States of America

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

Examining responders and non-responders to lifestyle interventions may aid in refining and tailoring obesity treatment. PREMIER was a multicenter trial designed to test the effects of an 18-month lifestyle intervention on blood pressure, weight, and diet among adults with prehypertension. The purpose of this secondary analysis was to identify early behavioral responders and non-responders to lifestyle interventions using data from 501 overweight and obese Black and White adults randomized to one of two active lifestyle interventions in PREMIER. Repeated measures latent class analysis (RMLCA) was applied to identify patterns of early response to treatment based on adherence to five daily recommendations at baseline and six months: 5 or more servings of fruits/vegetables; 30% or less energy from total fat; 10% or less energy from saturated fat; 2300 mg or less of sodium; and 180 minutes of moderate-to-vigorous activity. Based on the RMLCA, three distinct latent classes emerged: Responders (45.9%) met zero recommendations at baseline, but all at 6 months; NonResponders (23.6%) met none of the recommendations at baseline or at 6 months; and Early Adherers (30.5%) met three recommendations at baseline, but all at 6 months. Responders and Early Adherers had significantly greater weight loss than Non-Responders at six months, but there was no significant class difference in weight outcomes at 18 months. Early Adherers had significantly higher baseline self-efficacy than NonResponders. PREMIER lifestyle interventions were most effective, behaviorally, among individuals who entered treatment with high self-efficacy. Early behavioral response to lifestyle intervention is associated with greater initial weight loss. CORRESPONDING AUTHOR: Assistant Prof. PhD Fitzpatrick, Rush U n i v e r s i t y M e d i c a l C e n t e r, U n i t e d S t a t e s o f A m e r i c a , [email protected]

S206 INTERVENTION EFFECTS ON DIET AND EXERCISE IN MINORITIES WITH T2D J. Inouye1, J. Yomogida2, J. Davis3, R. Arakaki3, D. Li4 1 University of Nevada Las Vegas, LAS VEGAS, United States of America 2 The Clinic, LAHAINA, United States of America 3 University of Hawaii, HONOLULU, United States of America 4 University of Hawaii at Manoa, HONOLULU, United States of America Introduction: T2D accounts for 90-95% of all diabetic cases (CDC, 2011), with minorities particularly at risk. The aim of this project was to assess the effects of a self-management intervention on diet and exercise. Methods: Using a two-arm randomized clinical trial (n=207), data on dietary and exercise between Asian and Pacific Islanders were analyzed using linear mixed effects model before and after a seven week cognitive behavioral intervention. Measures include a 3-day diet recall and pedometer. Results: Findings revealed significant improvement in the intervention group in trans fat (p=0.0002), sugar (p=0.0333), alcohol (p=0.0007), caffeine (p=0.0008), diet fiber (p=0.0489), and saturated fat (p=0.0005) consumption compared to the control group. Findings also indicated large amount of calories consumed at baseline by both groups (4,721 kcal for the intervention and 5,368 for the control group). While not statistically significant, kilocalories increased a median of 406 in the control compared to 78 in the intervention group (p=0.27) with a greater increase in fat consumption (p=0.08). The intervention group increased the median number of steps by 60 during the study whereas the control group became less active by 790 steps (p=0.31). Older subjects had significantly less caloric intake than younger subjects (p=0.0146). Female subjects had significantly less calories intake and exercise compared to males (p=0.0007) and less steps on average (p=0.0059). Conclusions: Significant differences in gender and age point to research and treatment targeting these groups.

S65

CORRESPONDING AUTHOR: J. Inouye, University of Nevada, Las Vegas, United States of America, [email protected]

S207 WEB-BASED CBT FOR PATIENTS WITH EATING DISORDERS: PRELIMINARY RESULTS OF A RANDOMIZED CONTROLLED TRIAL E.D. ter Huurne1, H.A. De Haan1, M.G. Postel1, C.A.J. De Jong2 1 Tactus Addiction Treatment, ENSCHEDE, Nederland 2 Nijmegen Institute for Scientist Practitioners in Addiction, NIJMEGEN, Nederland Background Cardiovascular diseases (CVDs) are the leading causes of deaths in India and other developing countries. We aimed to assess the longitudinal change in behavioural and clinical risk factors for CVDs over seven years in rural Kerala, India. Methods We conducted a cohort study (2003-2010) among 495 individuals aged 15 to 64 years in a rural area of Kerala, India. We analyzed the data of 451 individuals (91.1%) who participated in the follow-up study. We used the World Health Organization STEPwise approach to surveillance (WHO STEPS) method to collect data at baseline and follow-up. Age-adjusted mean of risk factors was compared using linear regression and age-adjusted prevalence of risk factors was compared using logistic regression. Results The age-adjusted mean of risk factors at baseline showed a significant increase during follow-up for weight (54.1-59.5 kg, p<0.001), body mass index (BMI; 21.8-23.8 kg/m2, p<0.001) and waist circumference (82.5-87.0 cm, p<0.001). The age-adjusted prevalence of risk factors at baseline showed a significant increase during follow-up for current smokeless tobacco use (10.2-17.0%, p<0.001), current alcohol use (20.6-37.0%, p<0.001), BMI ≥23 kg/m2 (36.1-55.2%, p<0.001), BMI ≥25 kg/m2 (20.6-38.5%, p<0.001) and central obesity (51.4-64.1%, p=0.004). Conclusion There was a significant worsening of risk factors for CVDs in this rural sample, which calls for urgent intervention to curtail the impending CVD outbreak in this population. CORRESPONDING AUTHOR: MSc E.D. ter Huurne, Tactus Addiction Treatment, Nederland, [email protected]

S208 MANAGING DEPRESSION AND ANXIETY IN AUSTRALIANS WITH CHRONIC HEPATITIS C B.J.R.S. Stewart1, D. Turnbull1, J.M. Andrews2, A.A. Mikocka-Walus3 1 University of Adelaide, ADELAIDE, Australia 2 Royal Adelaide Hospital, ADELAIDE, Australia 3 University of York, YORK, United Kingdom This paper presents the results of a 4 year, mixed-methods research program aimed at improving management of depression and anxiety in Australian hepatitis C patients. Studies one and two sought to determine the scope of the problem, wherein the prevalence and longitudinal course of depression and anxiety were established. In study one, the rates of anxiety and depression in 395 patients were found to be 1.2 and 2.4 times higher than in the general community. In study two, a sub-set of 61 these patients was followed up 21 to 62 months later, where rates of depression and anxiety increased even further. Study three aimed to examine the perspectives of patients, hepatologists, and counsellors to explore how patients cope with and seek help for mental health problems. Respondents agreed that access to information, social support, and mental health services were critical. Better relationships and communication between patients and doctors were reported to promote access to support, while fears regarding stigma and breaches of confidentiality were noted as significant barriers. Studies four and five aimed to design an acceptable

S66

and effective treatment for depression and anxiety. Study four explored the acceptability of different treatment approaches in 156 patients and found that individual psychotherapy was favoured by the majority (83%). Our research has culminated in the design of C-UP in study five. C-UP is a five week Unified Protocol for hepatitis C patients with depression and anxiety. A pilot trial of this trans-diagnostic intervention is currently underway and will also be reported upon. CORRESPONDING AUTHOR: Mr. B.J.R.S. Stewart, University of Adelaide, Australia, [email protected]

S209 SEROSTATUS DISCLOSURE TO SEXUAL PARTNERS AND CONDOM USE IN PEOPLE LIVING WITH HIV N.P. Nancy1, E. Rodríguez-Estrada2, C. Iglesias-Chiesa2, A.S. Safren3, G. Reyes-Terán2, A. Riveros-Rosas4 1 National University of Mexico, MEXICO CITY, Mexico 2 Center for Research in Infections Diseases, MEXICO CITY, Mexico 3 Massachusetts General Hospital, Behavioral Medicine Service, BOSTON, United States of America 4 School of Business Administration, National University of Mexico, MEXICO CITY, Mexico Serostatus disclosure to sexual partners and condom use are relevant variables for HIV prevention, some studies show contradictory results regarding the effect of disclosure on condom use. The aim of this study was to examine the influence of psychological variables and type of partner on the association between these two variables, by means of structural equation modelling. We conducted a cross-sectional study in a sample of people living with HIV (PLWH) attending the Center for Research in Infectious Diseases in Mexico City. We collected data from 207 PLWH through psychometrical instruments adapted and validated to the Mexican context. Men represented 89% with mean age of 35years (SD=8), 56% reported having a regular partner, 31% had a casual partner and 13% both types of partners; 58% of the sample reported having disclosed their HIV-status to sexual partners, 34% reported no disclosure and 8% had disclosed only to regular partners; inconsistent condom use was reported by 24% of the sample. Results show that disclosure is not associated with condom use. The model for PLWH with regular partner showed satisfactory goodness of fit (x2=31.01, gl=25, p=.189) and revealed that more anxiety and higher treatment adherence predicted inconsistent condom use. The model for PLWH with casual partners also showed a satisfactory goodness of fit (x2=8.78, gl=9, p=.457) and confirmed the result that more anxiety and higher adherence are predictors of inconsistent condom use. The findings emphasize the importance of designing psychological interventions aimed at improving emotional regulation, serostatus disclosure and condom use skills. CORRESPONDING AUTHOR: N.P. Nancy, National University of Mexico, Mexico, [email protected]

O210 INCREASING THE ACCEPTABILITY OF MEDICAL MALE CIRCUMCISION IN ZAMBIA S.M. Weiss1, D.L. Jones1, R. Zulu2, N. Chitalu2, C. Redding3, R. Cook1 1 University of Miami Miller School of Medicine, MIAMI, United States of America 2 University of Zambia School of Medicine, LUSAKA, Zambia 3 University of Rhode Island, PROVIDENCE, United States of America INTRODUCTION: Medical male circumcision (MMC) has been confirmed by three major clinical trials to reduce HIV incidence by 60% in

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

high-prevalence populations. In Zambia only 12% of the male population is circumcised; 80% of the uncircumcised men have expressed no interest in MMC. This ongoing randomized clinical trial seeks to increase MMC among Zambian men, balancing supply and demand for MMC services to make most effective use of scarce resources. METHODS: Uncircumcised, HIV-negative men (n = 958) recruited from VCT services in 12 community health centers (CHCs) attended four weekly sessions on HIV prevention highlighting MMC, or timematched sessions on other health-related issues. Female partners (n = 820) were invited to participate in a comparable women's group. Assessments included readiness to undergo MMC using the Stages of Change model to assess attitudinal and behavioral changes over time (one year post-intervention). RESULTS: At baseline, most men (88%) were in Pre-Contemplation or Contemplation stages. Although those reporting greater readiness at baseline were more likely to undergo MMC (Mantel-Haenszel ChiSquare = 14.97, p = .001), this relationship was not significant in a multivariable model with study condition. The odds ratio for moving to the Action stage (i.e., MMC) was 2.16 (p = .045, 95% CI 1.02, 4.59) in favor of the experimental condition. There was no additional impact of partner participation on MMC decision-making. CONCLUSIONS: Preliminary results of this ongoing study suggest that the intervention significantly increased MMCs without reducing condom use. Policy implications concerning dissemination and implementation strategies will be discussed. CORRESPONDING AUTHOR: Prof. S.M. Weiss, University of Miami Miller School of Medicine, United States of America, [email protected]

O211 ANTIRETROVIRAL THERAPY IN ZAMBIA: DO PARTNERS ON ART ENHANCE ADHERENCE? D.L. Jones1, R. Cook1, S.M. Weiss1, N. Chitalu2 1 University of Miami Miller School of Medicine, MIAMI, United States of America 2 University of Zambia School of Medicine, LUSAKA, Zambia Background: Adherence Antiretroviral therapy (ART) is essential to optimize HIV treatment outcomes. Among individuals on ART, targeted peer support has been found to support adherence. This study of Zambian heterosexual couples examined whether partners would exert a positive influence on each other’s adherence. The study compared adherence between couples in which either one or both members were on ARTs. Methods: Couples (n = 446 participants), in which either or both members were on ART were assessed at baseline, 6 and 12 months. Results: Most participants (64%, n = 263) were on ART; overall, uptake of ART increased to 74% at 12 months. Couples in which both members were on ART increased from 37% to 48% at 12 months. At baseline, most reported near perfect adherence (76%), and at 6 months, 66%, and at 12 months, 70% were adherent. A regression analysis indicated that the decline in adherence did not differ between those couples in which one or neither partner was on ART (F (2,632) = .034, p = .712). Pairwise comparison indicated that adherence primarily decreased between baseline and 6 months (t = 2.95, p = .003), and was stable 6 to 12 months. Conclusions: This study found overall adherence among couples in Zambia to decline over time. In contrast to peer support literature, adherence was not enhanced in couples in which both members were on ART. Partners may not necessarily provide support for adherence to each other without the guidance provided by an intervention. Funded by NIMH grant R01HD058481. CORRESPONDING AUTHOR: Prof. D.L. Jones, University of Miami Miller School of Medicine, United States of America, [email protected]

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

O212 THE EFFECTS OF TEACHERS' AUTONOMY SUPPORT ON STUDENTS' PREVENTION OF H1N1 INFLUENZA IN CHINA D.K. C. Chan1, S. X. Yang2, X. Du3, N. L. D. Chatzisarantis1, M. S. Hagger1 1 Curtin University, PERTH, Australia 2 Sichuan University, CHENGDU, China 3 Hebei University, BAODING, China Chan, D. K. C.1, Yang, S. X.2, Du, X.3, Chatzisarantis, N. L. D.1, & Hagger, M. S.1 1 Curtin University, Australia 2 Sichuan University, China 3 Hebei University, China Introduction: This study adopted a quasi-experimental design to examine if teachers’ autonomy support influenced students’ perception of autonomy support, self-determined motivation and intention of wearing facemasks in the classroom for the prevention of H1N1 during an influenza pandemic. Method: Participants were 705 university students in China (mean age = 20.30) who received a hypothetical scenario about an H1N1 pandemic. In the scenario, they were asked by their teachers to wear facemasks in the classroom for preventing the transmission of H1N1 within the campus. In group 1, the teachers (in the scenario) delivered the health advise in an autonomy supportive manner (e.g., care, support, and rationales), but the teachers in group 2 did so in a controlling style (e.g., school rule, punishment). All participants responded to the items of the study variables according to the given scenario. Results: Two-way ANOVA showed that the perception of teachers’ autonomy support was higher in autonomy-supportive scenario than in the controlling scenario. Variance-based structural equation modeling revealed that the perception of teachers’ autonomy support was a positive predictor of the intention through the mediation of self-determined motivation. Conclusions: The promotion of preventive behaviors for infectious disease (e.g., H1N1) might be more effective by using an autonomy supportive health advise because it appears to be more adaptive to the motivational patterns and intention of the heath behavior. CORRESPONDING AUTHOR: Dr. D.K. C. Chan, Curtin University, Australia, [email protected]

O213 PERSONAL CONTEXT AND CHILDHOOD EXPERIENCES INFLUENCE ADULT VACCINATION BEHAVIOUR A. Wheelock, A. Parand, M. Miraldo, C. Vincent, N. Sevdalis Imperial College London, LONDON, United Kingdom Introduction: Routine vaccination uptake in adults remains suboptimal and poorly understood. In the US, only 66% of adults aged ≥65 years and 47% of younger high-risk adults reported receiving a flu shot in 2012/2013, whereas in the past 10 years, 64% of adults aged 19-64 years and 53% of adults aged ≥65 years received tetanus boosters. The UK achieved higher flu vaccination rates among people aged ≥65 years (6877%), yet coverage among younger high-risk adults was inadequate (5080%). Adult tetanus boosters are inconsistently recorded and not routinely recommended in the UK. Social and psychological factors can significantly influence vaccination uptake. We aimed to gain deeper understanding of how people’s personal context and experiences shape their beliefs and attitudes towards vaccines, particularly flu and tetanus. Methods and results: 40 in-depth, face-to-face interviews were conducted with members of the public (20 in the US, 20 in the UK). Regular flu vaccinators were more likely to trust their doctor, family and friends than intermittent vaccinators and non-vaccinators. Non-vaccinators were more likely to have had a vaccine or health-related adverse psychological experience during childhood. Tetanus vaccinators were more likely to

S67

remember their mother’s warnings during childhood about tetanus’ severity. Tetanus boosters were generally more acceptable than flu shots. Conclusions: Individuals’ context and childhood experiences can both negatively and positively influence their vaccination behaviour. Vaccination decisions should be studied as continuum and not in isolation from each other or from other important health, or lifestyle-related events. Integrated interventions should be considered. CORRESPONDING AUTHOR: Ms A. Wheelock, Imperial College London, United Kingdom, [email protected]

O214 SELF-ESTEEM, DIURNAL CORTISOL AND DAILY HEALTH PROBLEMS IN OLDER ADULTHOOD: AN 8-YEAR LONGITUDINAL ANALYSIS S. Liu, Carsten Wrosch Concordia University, MONTREAL, Canada Introduction: The likelihood of confronting daily health stressors increases with age, thus placing older adults at risk of cortisol dysregulation. Research has shown that self-esteem can buffer increases in cortisol secretion and detriments to physical health. Here we explore the protective role that self-esteem may play in the associations between daily health problems and diurnal cortisol. We expected that older adults with low, but not high, self-esteem would exhibit increased cortisol if they encounter more daily health problems than their normal average. Methods: 100 older adults (aged 60+) completed 5 two-year assessments in the Montreal Aging and Health study. Self-esteem was measured at baseline. In addition, 5 daily measures of cortisol and self-reported daily health problems were collected on 3 different days at each wave. Data were analyzed using hierarchal linear modeling to examine within-person associations between health problems and cortisol and between-person effects of participants’ self-esteem scores. Potential demographic confounds were also assessed. Results: Higher-than-average levels of cortisol were associated with more daily health problems over 8 years. Furthermore, the analysis indicated a cross-level moderation effect of level-2 self-esteem scores on level-1 associations between cortisol and daily health problems (β = -.15, SE = .05, t(93) = -3.29, p = .002). Higher levels of self-esteem ameliorated the within-person association between cortisol levels and daily health problems. Conclusions: These results identify self-esteem as a buffer to older adults’ biological dysregulation and physical health problems. CORRESPONDING AUTHOR: S. Liu, Concordia University, Canada, [email protected]

O215 REMEMBERED PARENTING AND PHYSICAL AND MENTAL FUNCTIONING IN MIDDLE AGED AND OLDER MEN AND WOMEN D.A.I. Groffen1, M.A. Wouters2, M. van den Akker1, H. Bosma1 1 Maastricht University, MAASTRICHT, The Netherlands 2 GGD Midden Limburg, UTRECHT, The Netherlands Introduction. Child abuse and maltreatment are serious threats for health and function across the life course. Less severe, sub-optimal, parenting styles, such as lack of warmth and feeling rejected are also known to affect health in children and adolescents and even in adulthood. We set out to examine whether remembered parenting styles continue to affect health- related function in old age and investigate the influence of participants’ and parents’ educational level on these associations. Methods. Data were from 2,316 Dutch men and women aged 55 thru 90 who participated in the longitudinal Study on Medical Information and Lifestyles Eindhoven (SMILE). The Dutch short version of the EMBU

S68

(i.e. Swedish acronym for own memories concerning upbringing) was used to measure remembered parenting. Logistic regression analyses were used to study associations with health-related function and the one-year changes therein. Results. Particular remembered parenting styles, including emotional warmth and rejection, continue to affect health in and beyond midlife. This effect was independent of participants’ and parental educational level. For example, in a fully adjusted model, participants reporting rejection of the father had an almost 1.5 (95% CI: 1.12-1.95) higher odds of reporting physical dysfunction one year later. Moreover, remembered lack of emotional warmth from both the father and the mother, and remembered father’s overprotection were each related to increasing mental dysfunction between baseline and follow-up. Conclusion. Given the evidence for the effect of remembered parenting style into and beyond midlife, these findings underscore the need for intervention upon sub-optimal, “unhealthy”parenting styles. CORRESPONDING AUTHOR: Dr. D.A.I. Groffen, Maastricht University, The Netherlands, [email protected]

O216 VASCULAR BURDEN AND COGNITIVE FUNCTION IN LATELIFE DEPRESSION M. Zuidersma1, G.J. Izaks1, P. Naarding2, H.C. Comijs3, R.C. Oude Voshaar1 1 University of Groningen, University Medical Center Groningen, GRONINGEN, Nederland 2 GGNet, Department of Old-age Psychiatry, APELDOORN/ZUTPHEN, Nederland 3 VU University Medical Center / GGZinGeest, AMSTERDAM, Nederland Introduction: Vascular burden is associated with cognitive impairments and decline in older persons from the general population, but their relation in depressed older persons remains unclear. Method: Depressed older persons were recruited from general practices and mental health care institutes. Vascular risk profile was assessed with the Framingham Risk Score (FRS), subclinical vascular disease with the ankle-brachial index, and manifest cardiovascular disease (CVD) was defined as a history of a cardiovascular event. Three neurocognitive tasks were administered to evaluate processing speed, working memory, verbal memory and executive functioning. With linear regression analyses, the relation between vascular burden and cognitive function was assessed. Results: A total of 378 depressed older persons were enrolled. FRS related to impaired executive functioning (beta -0.13; p<0.05), but to no other cognitive domain after adjustment for age, sex, education level and depressive symptom severity. Subclinical and manifest CVD were related to slower processing speed (beta -0.14 p<0.05 for subclinical vascular disease and beta -0.17 p<0.01 for manifest CVD), but to no other cognitive domain. In 266 participants without manifest CVD, higher FRS scores related to slower processing speed and poorer executive functioning (both beta -0.15; p<0.05), but not to working memory and verbal memory. Subclinical vascular disease was related to slower processing speed (beta 0.129 p<0.05), but to no other cognitive domain in persons without CVD. Conclusions: Vascular burden relates to slower processing speed and impaired executive functioning in depressed older persons. It is therefore essential to ensure proper cardiovascular prevention strategies in depressed older persons. CORRESPONDING AUTHOR: M. Zuidersma, University of Groningen, University Medical Center Groningen, Nederland, [email protected]

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

O217 QUALITY OF LIFE OF INFORMAL CAREGIVERS: RELATIONS WITH OBJECTIVE AND SUBJECTIVE BURDEN M. Oldenkamp1, M. Hagedoorn1, J.P.J. Slaets1, R.P. Stolk1, R.P.M. Wittek2, N. Smidt1 1 University Medical Center Groningen, GRONINGEN, Nederland 2 University of Groningen, GRONINGEN, Nederland Introduction: Due to an increasing life expectancy and increasing health care costs, informal care becomes a crucial element in our health care system. A good caregiver quality of life is necessary for the provision and continuation of this informal care. In this study we investigate whether caregiver quality of life is mainly determined by the caregiving situation itself (objective burden), or by the experience with this caregiving situation (subjective burden). Method: 424 informal caregivers participated in this crosssectional study (69% female; mean age 63.4 yrs, SD 11.5). The questionnaire contained questions on the following objective burden factors: hours and types of informal care, care recipient’s health, relation between caregiver and care recipient. Subjective burden was assessed with the SelfRated Burden scale (SRB). Quality of life was assessed with a single item question. Ordinal logistic regression analyses were performed, adjusted for caregiver’s age and gender, and care recipient’s age, gender, and living arrangement. Results: Of the factors of objective burden, caring for someone with high multimorbidity was related with low caregiver quality of life (OR .886, 95%CI .750 to .987). In addition, informal caregivers with a high subjective burden had a lower quality of life (OR .799, 95%CI .722 to .878). Conclusions: Both objective and subjective burden are related to the quality of life of informal caregivers. Caregiving interventions aiming at improvement of caregiver quality of life should not only focus on objective or subjective burden, but on both. CORRESPONDING AUTHOR: M. Oldenkamp, University Medical Center Groningen, Nederland, [email protected]

O218 DIFFERENCE IN LIVING FORMS ASSOCIATED WITH POOR DIETARY INTAKE AMONG JAPANESE ELDERLY: THE OHASAMA STUDY. M. Tsubota-Utsugi1, M. Kikuya2, M. Satoh3, R. Inoue4, M. Hosaka3, H. Metoki2, N. Fukushima5, T. Hirose6, K. Asayama7, Y. Imai3, T. Ohkubo8 1 National Institute of Health and Nutrition, TOKYO, Japan 2 Tohoku Medical Megabank organization, Tohoku University, SENDAI, Japan 3 Department of Planning for Drug Development and Clinical Evaluation, Tohoku Univ, SENDAI, Japan 4 Department of Medical Informatics, Tohoku University Graduate School of Medicine, SENDAI, Japan 5 Division of Community Health Nursing, School of Nursing, Fukushima Medical Unive, FUKUSHIMA, Japan 6 Center for Interdisciplinary Research in Biology, College de France, PARIS, France 7 Studies Coordinating Centre, Division of Hypertension and Cardiovascular Rehabil, LEUVEN, Belgium 8 Department of Hygiene and Public Health, Teikyo University School of Medicine, TOKYO, Japan Introduction. With rapid increases in life expectancy, number of elderly people living alone or those forced to change living forms has increased. Previous studies found that poor dietary intake was common in elderly people living alone. However, studies are sparse on dietary intake in elderly people living in other forms, particularly those living with family other than a spouse (nonspouse family) which is common in Japan. Thus

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

we examined the differences in dietary intake by different living forms in elderly Japanese. Methods. We analyzed the data of 1542 healthy residents aged 60 years and over in Ohasama town, who completed self-administrated questionnaires. Dietary intake was measured using a 141-item food frequency questionnaire. Multiple regression models with robust (White-corrected) SEs were individually fitted for foods with living forms. Results. In men, compared with those living with only a spouse, intakes of protein-related nutrients, e.g., pulse, fish, and daily product, were significantly lower in participants living with nonspouse family, while intakes of fruits and vegetables were significantly lower in participants living alone. In women, lower intakes of fruit and protein-related foods were significantly more common in participants living with nonspouse family than those living with only a spouse. Conclusion. Not only elderly people living alone but also those living with family other than a spouse had poor dietary intake, suggesting that strategy improving knowledge for food choices and skills for food preparation could contribute to promote healthy eating in elderly Japanese particularly living in those forms associated with poor dietary intake. CORRESPONDING AUTHOR: M.T. Tsubota-Utsugi, National Institute of Health and Nutrition, Japan, [email protected]

O219 IS THE ASSOCIATION BETWEEN COGNITIVE AND PHYSICAL IMPAIRMENT IN OLDER ADULTS CONTRIBUTABLE TO CARDIOVASCULAR DISEASES? C. Qiu1, Y. Liang2, Z. Yan3, C. Cai4, H. Jiang4, A. Song2 1 Karolinska Institutet, STOCKHOLM, Sweden 2 Jining Medical University, JINING, China 3 Jining First People’s Hospital, JINING, China 4 Xing Long Zhuang Coal Mine Hospital, JINING, China Introduction: Most women develop causal beliefs following diagnosis with breast cancer and these beliefs can guide decisions around care and management. Bilateral mastectomy rates are increasing, although the benefits of this surgery are only established in a small percentage of women. In this study we investigated the relationship between causal beliefs and the decision to undergo bilateral mastectomy. Method: 2,269 women from the Army of Women’s breast cancer research registry completed an on-line survey. Women were asked what they believed caused their cancer and responses were coded into eight causal categories. Participants were also asked about the type of surgery they underwent following their breast cancer diagnosis. The odds ratio for having had bilateral mastectomy were calculated for each causal category using random/bad luck as referent category. Results: Hormonal factors (22%) and genetics (19%) were the most common causal belief, followed by don’t know (19%), environmental toxins (11%), negative emotions (9%), poor health behavior (8%), other (6%) and random/bad luck (6%). Compared to the referent category, the odds ratio of having a bilateral mastectomy was significantly higher in both the genetics and hormonal causal belief groups (OR=2.58 [95%CI: 1.56-4.26] and OR=1.87 [95%CI: 1.133.08] respectively). Conclusions: Beliefs in a genetic cause for breast cancer are common and are associated with high rates of bilateral mastectomy. This is despite evidence that the actual genetic contribution to breast cancer is much lower than perceived and that bilateral mastectomy is, in most cases, unlikely to reduce the future risk of cancer or improve survival. CORRESPONDING AUTHOR: Dr. C. Qiu, Karolinska Institutet, Sweden, [email protected]

S69

INVITED SYMPOSIUM: ABMR symposium: common disease pathways- pleitopic biobehavioural mechanisms

S220 ABMR SYMPOSIUM: COMMON DISEASE PATHWAYSPLEIOTROPIC BIO-BEHAVIOURAL MECHANISMS K.W. Davidson Center for Behavioral Cardiovascular Health/Columbia University, NEW YORK, United States of America Biobehavioral mechanisms affect the outcomes of many common diseases, and yet as scientists we often investigate one specific biobehavioral mechanism in the context of only one specific disease. To contribute to intellectual stimulation and consideration of innovative and alternative ways to approach both our science and our contribution to public health, we review some important and novel biobehavioral mechanisms that may impact on multiple common diseases. Such a scientific approach allows new investigators to consider the ‘pleiotropic’, or multifaceted ways that their mechanism might impact a myriad of diseases. The Academy of Behavioral Medicine Research selected four distinguished scientists to present their insights into novel and innovative ways of approaching the ‘common disease’ research paradigm. Dr. Andrew Steptoe will review the biobehavioral mechanism of emotional triggers, consider their potential mechanistic pathways, and discuss the many disease processes that may be triggered by acute emotions. Dr. Susan Lutgendorf will present how stress primes inflammation, and the variety of common diseases that are impacted by this prevalent biobehavioral pathway. Dr. Heiddis Valdismirdottir will review research on disrupted circadian rhythms, and the ways this impacts upon fatigue and other processes. Finally, Dr. Jerry Suls will discuss ways to better consider biobehavioral mechanisms across common diseases. The best way to improve the pipeline of scholars from diverse backgrounds into behavioral medicine is to offer exciting, new areas of science ready to be investigated. This symposium will accomplish that by presenting three innovative ways that behavioral medicine can impact upon a myriad of common diseases. CORRESPONDING AUTHOR: Prof. K.W. Davidson, Center for Behavioral Cardiovascular Health/Columbia University, United States of America, [email protected]

S221 BIOBEHAVIORAL MECHANISMS UNDERLYING EMOTIONAL TRIGGERS A. Steptoe1 1 University College London, LONDON, United Kingdom

INTRODUCTION: Much biobehavioral research in behavioral medicine is focused on the processes contributing to the long-term development of disease. Such work typically centers on chronic challenges to regulatory processes and how they impact on disease etiology. METHOD: There are a number of health conditions in which acute emotional triggers have a prominent role, eliciting abrupt pathophysiological response that have serious adverse consequences. Examples include emotional triggers of acute myocardial infarction, sudden cardiac death, stroke, asthma attacks, vasovagal syncope, and possibly some autoimmune conditions and the reactivation of latent infections. RESULTS: The biobehavioral mechanisms involved in these circumstances differ from those implicated in chronic etiology in a number of ways. First, the emotional stimuli do not involve exposure to long-term

S70

adversity, but may be brief, ranging from intense short-term episodes of anger, fear or sadness to major acute life events. Second, the biobehavioural responses to emotional triggers typically act against a substrate of pre-existing illness (advanced coronary atherosclerosis, hyperactive airways, etc) rather than healthy tissue. Third, special methodologies are required for investigating emotional triggers, since episodes are frequently unpredictable and transient. Finally, there may be an interplay between the intensity of emotional trigger and severity of underlying pathology, such that relatively minor triggers may elicit responses when the underlying pathology is advanced, while major emotional events are needed to stimulate adverse outcomes in healthier individuals. CONCLUSIONS: Acute emotional triggers are currently studied in relation to several individual health conditions, but cross-talk between areas of behavioral medicine research may reap major benefits. CORRESPONDING AUTHOR: Prof. K.W. Davidson, Center for Behavioral Cardiovascular Health/Columbia University, United States of America, [email protected]

S222 BRIGHT LIGHT THERAPY, CANCER RELATED FATIGUE AND CIRCADIAN ACTIVITY RHYTHM H.B. Valdimarsdottir1, W. Davidson2, W.R. Redd3, L.W. Wu3, S. Ancoli-Israel4 1 Reykjavik University, REYKJAVIK, Iceland 2 Center for Behavioral Cardiovascular Health/Columbia University, NEW YORK, United States of America 3 Mount Sinai School of Medicine, NEW YORK, United States of America 4 University of California, San Diego, SAN DIEGO, United States of America INTRODUCTION. Circadian rhythms (CR), the approximately 24 hour biological activity of behavior and physiology are vital for health. Desynchronized rhythms are considered one of the fundamental underlying mechanisms for several biological, cognitive and psychological disruptions observed in various chronic diseases (e.g., cancer, heart disease) even long after treatment cessation. As CR´s are affected by light the present pilot study sought to examine if bright white light therapy (BWL) restored CR (i.e., circadian activity rhythm {CAR}) among survivors of chronic disease or cancer. In addition we examined if BWL would improve cancer related fatigue (CRF) and if this effects would be mediated by CAR. METHODS. Thirty-one cancer survivors were randomized to either a BWL (n=15) or to a dim red light (DRL; n=16) condition. Assessments were conducted at: baseline, during the second and fourth (and last) week of the intervention and three weeks after the intervention. Participants self-administer light treatment for 30 minutes every morning throughout the four week intervention period. RESULTS. Patients receiving BWL showed improved CAR (p < 0.05), a significant decrease in CRF (p< 0,05), and when CAR was entered into the equation predicting CRF the parameter value for the intervention decreased in significance suggesting CAR’s role as a mediator. CONCLUSIONS. Providing bright white light therapy may provide an non-invasive, innovative way to normalize circadian rhythms in many chronic diseases which are vital for health and well-being. CORRESPONDING AUTHOR: Prof. W. Davidson, Center for Behavioral Cardiovascular Health/Columbia University, United States of America, [email protected]

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

S223 INFLAMMATORY PROCESSES IN CHRONIC PAIN AND CANCER-RELATED FATIGUE S. Lutgendorf University of Iowa, IOWA CITY, IA, United States of America

INFLAMMATORY PROCESSES IN CHRONIC PAIN AND CANCER-RELATED FATIGUE Introduction. Inflammation is a key factor in a number of conditions. The emerging role of non-neuronal glial and microglial cells in the facilitation of nerve impulses has been one of the most exciting recent developments in the field of chronic pain. Pathological expression of pro-inflammatory cytokines can enhance propagation and/or even generation of nociceptive signals. Systemic inflammation can also produce vegetative symptoms characterized by fatigue, sleep disorders, and depression. This is thought to be one mechanism underlying cancer-related fatigue. Stress can exacerbate many of these pathways. This presentation will examine inflammation in the contexts of chronic urologic pelvic pain and cancer-related fatigue. Methods. Attention will be paid to measurement issues, in terms of information that can be gleaned from assessment of stimulated and nonstimulated cytokines, and assessment of cortisol and glucocorticoid resistance. Results. Elevated inflammatory cytokine responsivity to Toll-like receptor-2 and -4 stimulation and flatter diurnal cortisol slopes have been found in chronic urologic pelvic pain, indicating multiple forms of inflammatory dysregulation. In fatigued ovarian cancer patients, elevated inflammatory cytokines and flatter cortisol slopes have been observed, while in fatigued breast cancer patients, specific inflammatory polymorphisms have been identified, suggesting underlying vulnerability for inflammation. Conclusions: Examination of genetic vulnerabilities along with inflammatory and HPA profiles may offer an innovative approach to identifying risk profiles for individuals likely to develop profound fatigue after cancer treatment, or for individuals whose acute pain is likely to turn into chronic pain. CORRESPONDING AUTHOR: Prof. W. Davidson, Center for Behavioral Cardiovascular Health/Columbia University, United States of America, [email protected]

INVITED SYMPOSIUM: The role of placebo mechanisms in medicine and psychotherapy S224 THE ROLE OF PLACEBO MECHANISMS IN MEDICINE AND PSYCHOTHERAPY W.R. Rief, B. Doering University of Marburg, MARBURG, Germany While clinical trials usually try to limit placebo mechanisms, they can play a significant role for outcome in medicine and psychotherapy. The last decade brought significant insight into placebo mechanisms and processes both on a psychological and neurobiological level. These new findings do not only challenge our understanding of the pathways of action, but they also offer new approaches how to optimize treatments. Our first presentation (Rief & Doering, Germany) will highlight the role of placebo mechanisms in clinical trials in general. After demonstrating the significant impact that placebo mechanisms can have, we will outline several approaches how to make use of optimizing expectation and learning processes for the sake of the patient. Investigations of New Zealand (Crichton et al.) demonstrate how placebo and nocebo mechanisms can be induced experimentally to show how these mechanisms contribute to the perception of health risks, loss of efficiency of generic

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

drugs, or the improvement of medical interventions. Our final presentation (Gaab et al., Switzerland) will discuss how placebo mechanisms can be defined in psychotherapy, and how to make use of these mechanisms in psychotherapy to improve outcome. Discussant: Keith J. Petrie, Auckland, New Zealand CORRESPONDING AUTHOR: Prof. W.R. Rief, University of Marburg, Germany, [email protected]

S225 THE ROLE OF THE PLACEBO MECHANISMS IN CLINICAL TRIALS W. Rief, A. Winkler, B. Doering University of Marburg, MARBURG, Germany Meta-analysis of trials investigating pain medication, antidepressants, or antihypertensives reveal large effects in the placebo groups with notoriously smaller incremental benefits in the drug group.Hereby the principle rationale of clinical trials is the assumption of additivity of unspecific and specific factors contributing to the outcome. Methods: Using meta-analysis techniques, we analyzed placebo mechanisms in 50 trials investigating sleep medication. Based on these and other results, we develop paradigms how to make use of the involved placebo mechanisms. Results: Our results confirm strong placebo effects in trials investigating sleeping medication, accounting for major parts of the variance of the outcome. Placebo effects are found both for subjective and objective outcome measures (PSG assessments). Discussion: The substantial role of expectation and conditioning for the development of placebo responses is further confirmed. We therefore propose several new approaches how to make use and optimize these mechanisms (e.g., through pharmacological conditioning paradigms). These results further question the assumption of additivity of specific and unspecific effects in clinical trials. Therefore alternatives to the traditional double blinded 2-3 armed design are needed. CORRESPONDING AUTHOR: Prof. W.R. Rief, University of Marburg, Germany, [email protected]

S226 HARNESSING THE PLACEBO RESPONSE TO COUNTERACT NOCEBO RESPONDING IN THE COMMUNITY F. Crichton, K.J. Petrie University of Auckland, AUCKLAND, New Zealand Introduction: The dissemination of information suggesting adverse health effects may result from exposure to new technologies, such as windfarms, can generate negative expectations leading to symptom reporting in the community. The purpose of the study was to investigate whether creating positive expectations about such technologies, to harness the placebo effect, might reduce or reverse symptom complaints. Method: In the context of a study investigating symptoms during exposure to windfarm sound, 64 volunteers assessed their symptomatic experiences during two discrete sessions, throughout which they listened to wind turbine sound containing audible and sub-audible (infrasound) components. Audio-visual materials incorporating information on the internet about the purported therapeutic effects of infrasound (the positive expectation DVD), as well as television footage about the alleged adverse health effects of exposure to infrasound produced by windfarms (the negative expectation DVD), were shown to every participant. Half the participants viewed the positive expectation DVD before their first exposure session and the negative expectation DVD before their second listening session, while the remaining participants watched the DVDs in the opposite order

S71

Results: During listening sessions, participants were significantly less symptomatic following the delivery of negative expectations, if they had previously received positive expectations about infrasound exposure (p<.05). Further, participants who received positive expectations after the earlier delivery of negative expectations exhibited a placebo response, which reversed the nocebo response seen in their first listening session (p<.001). Conclusion: Results provide promising indications that harnessing the placebo effect may counteract nocebo responding in the community. CORRESPONDING AUTHOR: L Crichton University of Auckland New Zealand [email protected]

S227 PLACEBO EFFECTS IN PSYCHOTHERAPY - THREE RANDOMIZED-CONTROLLED DECEPTIVE TRIALS J. Gaab, M. Meury, R. Cavanesi, J.G. Birkhäuer, J.G. Kossowsky, P. Krummenacher University of Basel, BASEL, Switzerland The study of placebo and its effects highlight the importance of the meaning and the psychosocial context for the understanding of treatments and their effects. However, it needs to be noted that the vast majority of placebo research is focused on placebo interventions with a medical meaning. While this is important on its own right, placebo interventions with a psychotherapeutical meaning are scarce. Therefore, we set out to assess the effects of the meaning and of a friendly and empathetic relationship in the context of placebo interventions. Since the manipulation of the meaning of a given intervention requires the experimental misinformation of participants about the real nature of the intervention and the study, we employed deception designs. Overall, three independent studies with a total 0f 240 subjects were performed. All subjects were randomized to different verbal instructions about the intervention (controls, controls + friendly relationship, plausible meaning and plausible meaning + friendly relationship) and mood was assessed repeatedly before and short- and mid-term after the intervention. The results indicate an overall significant and lasting effect, with specific and incremental improvement in mood over time for groups with a plausible meaning and plausible meaning and friendly relationship over the other conditions. The results show that (1) experimental designs from placebo research can be utilized in the context of psychotherapy research and (2) the significance of the plausibility of an intervention and of a warm, friendly and empathetic relationship is independent of its scientific trueness. CORRESPONDING AUTHOR: Prof. J.G. Gaab, University of Basel, Switzerland, [email protected] Medically unexplained symptoms: the translational pathway S227 MEDICALLY UNEXPLAINED PHYSICAL SYMPTOMS: THE TRANSLATIONAL PATHWAY T. Chalder King's College London, LONDON, United Kingdom Medically Unexplained Physical Symptoms (MUPS) are symptoms for which it is not possible to identify a specific cause. This does not mean the symptoms are ‘all in the mind’ as experiences may be distressing and usually affect people’s ability to undertake daily activities. Examples of MUPS include irritable bowel syndrome, chronic fatigue syndrome, fibromyalgia, chronic tension headaches and chronic pelvic pain. These syndromes often overlap and are associated with lots of consultations, referrals and investigations which can lead to increased levels of disability. This unnecessary activity is very costly to the NHS and some investigations carry their own risks (such as surgery, radiological procedures etc). There is an accumulating body of evidence demonstrating that

S72

cognitive behavioural interventions can reduce levels of symptoms and improve functioning in patients who have MUPS. In 2011 a metaanalysis examined the efficacy of short term psychotherapy for MUPS (Kleinstauber et al 2011) and this drew on ten previously published systematic reviews on psychological interventions as well as new trials. In general there was evidence that CBT was effective for MUPS. In this symposium we will investigate factors associated with the development of symptoms, present new research findings from randomised controlled trials and will present preliminary results of a mediational analysis. Discussion will centre around common mechanisms of change which may be fed back into the translational pathway and targeted in future trials. CORRESPONDING AUTHOR: Prof. T. Chalder, King's College London, United Kingdom, [email protected]

S228 MEDIATORS OF CHANGE IN FATIGUE AND FUNCTION IN THE PACE TRIAL OF TREATMENTS FOR CFS T. Chalder1, K. Goldsmith1, P.D. White2, M. Sharpe3, A. Pickles1 1 King's College London, LONDON, United Kingdom 2 Queen Mary University of London, LONDON, United Kingdom 3 Oxford University, OXFORD, United Kingdom Introduction. In the recent multi-centre PACE trial, cognitive behaviour therapy (CBT) and graded exercise therapy (GET) were found to be more effective in reducing fatigue and improving physical function than adaptive pacing therapy and specialist medical care (White et al., 2011). The aim of this paper was to investigate potential mechanisms of change in the context of this trial in which self-reported unhelpful cognitions were assessed before during and after treatment. Methods. Mediation was assessed using the product of coefficients POC, method. A series of linear regression models were fitted including covariates. The 12 week post-randomisation measurement of each of the putative mediators and the 52 week measure of each of the outcome variables were used. Mediation effect estimates were calculated as a*b, where a = treatment parameter estimate from mediator model and b = mediator parameter estimate from outcome model. Bootstrapping was used to obtain confidence interval estimates for the mediated effects. Differences in the mediator-outcome relationship by treatment were explored using a treatment by mediator interaction term in the outcome model . Results. Cognitive responses had a large mediated effect on both fatigue and physical function for both CBT and GET, with the effect being stronger for GET than for CBT. Conclusion. Changes in cognitions mediated the change in CBT and GET but more for GET. The results strengthen the validity of a theoretical model of CFS by confirming the role of fearful beliefs. CORRESPONDING AUTHOR: Prof. T. Chalder, King's College London, United Kingdom, [email protected]

S229 CBT, SEAL OIL OR BRIEF INTERVENTION. WHAT IS MOST EFFECTIVE IN LOW BACK PAIN TREATMENT? H. Eriksen1, S.E. Reme1, T.H. Tveito1, E.A. Fors2, T. Tangen3, L. Froyland4, T. Chalder5, A. Indahl3 1 Uni Research, BERGEN, Norway 2 Norwegian University of Science and Technology, TRONDHEIM, Norway 3 University of Bergen, BERGEN, Norway 4 National Institute of Nutrition and Seafood, BERGEN, Norway 5 King's College London, LONDON, United Kingdom

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

Objective To evaluate the effectiveness of 4 different interventions; Brief Intervention, Brief Intervention and Cognitive Behavioral Therapy (CBT), Brief Intervention and nutritional supplements of seal oil, and Brief Intervention and nutritional supplements of soy oil. Methods Randomized controlled trial in secondary care. Participants 409 adults aged 18-60 years sick listed for 2-10 months due to low back pain. Interventions Patients were randomly assigned to Brief Intervention (BI) (n=100), BI and CBT (n=101), BI and Seal oil (n=103), or BI and Soy oil (n=105). Brief Intervention is a one-session cognitive, clinical examination program, followed by two booster sessions. The CBT is a 7-session tailored and manualized treatment program, while the nutritional supplements consist of a dosage of 10 grams of either soy or seal oil (capsules) per day for 3 months, in a double blind design. Main outcome measures Primary outcome was return to work at 1-year follow-up. Secondary outcome measures were pain and disability. Results 70 % of the participants reported pain as their main problem. On average, they had 10 different subjective health complaints. 31% of the participants had a current psychiatric disorder. Improvement of pain and disability did not differ significantly between the groups at 3, 6 or 12month follow-up. Conclusion The majority of patients on sick leave due to low back pain have co- and multimorbid subjective health complaints. A large proportion has also psychiatric disorders. The brief intervention seems to be a strong intervention, and adding additional interventions have limited effects on pain and disability. CORRESPONDING AUTHOR: Prof. R. Eriksen, Uni Research, Norway, [email protected]

S230 FATIGUE AND DEPRESSION IN PATIENTS SICK LISTED FOR LOW BACK PAIN H. Eriksen1, S.E. Reme1, T. Tangen2, T. Chalder3, H. Snekkevik4 1 Uni Research, BERGEN, Norway 2 University of Bergen, BERGEN, Norway 3 King's College London, LONDON, United Kingdom 4 Friskvernklinikken, ASKER, Norway Objective: The relationship between fatigue and pain has been investigated previously, but little is known about the prevalence of substantial fatigue in patients sick-listed for low back pain (LBP), and how fatigue is associated with depression, pain, and long-term disability. The aims of the study were to examine the prevalence of substantial fatigue, associations between fatigue, depression, and pain, and whether fatigue predicted long-term disability. Methods: 569 patients, participating in a randomized controlled trial and sick-listed 2-10 months for LBP, were included in the study. Cross-sectional analyses were conducted to investigate the prevalence and independent associations between fatigue, depression, pain, and disability, while longitudinal analyses were done to investigate the association between fatigue and long-term disability. Results: The prevalence of substantial fatigue was 69.7%. Women reported significantly more fatigue than men (p<.001). Those with substantial fatigue had higher pain intensity (p=0.01), more depressive symptoms (p<0.001), and more disability (p<0.001) than those without substantial fatigue. Musculoskeletal pain and depression were independently associated with substantial fatigue. In the longitudinal analyses, fatigue predicted long-term disability at 3, 6, and 12-months follow-up. After controlling for pain and depression, fatigue remained a significant predictor of disability at 6months follow-up. Conclusions: The vast majority of sick listed LBP patients reported substantial fatigue. Those with substantial fatigue had more pain and depressive symptoms, and a significant risk of reporting more disability after 3, 6 and 12 months. Substantial fatigue is disabling in itself, but in addition involves a risk of developing chronic fatigue syndrome and long-term disability.

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

CORRESPONDING AUTHOR: Prof. R. Eriksen, Uni Research, Norway, [email protected] S231 THE USE OF COGNITIVE BEHAVIORAL THERAPY IN IRRITABLE BOWEL SYNDROME V.S. Sveinsdottir1, S.E.R. Reme1, T. Chalder2 1 Uni Research, BERGEN, Norway 2 Department of Psychological Medicine, King’s College Hospital, LONDON, United Kingdom Introduction: This review aims to critically evaluate the effectiveness and current knowledge about the use of cognitive behavioral therapy (CBT) in the treatment of irritable bowel syndrome (IBS). Method: A systematic literature search on all studies published up until January 2014 (EMBASE, PsycINFO and PubMed) was performed. The search string consisted of the following steps: 1) cognitive therapy/ treatment/management/modification/intervention, 2) behavior(al) therapy/treatment/management/modification/intervention, 3) irritable bowel syndrome and 4) randomized controlled/clinical trial/study (all terms including British and American spelling). The review only included randomized controlled trials (RCTs), and interventions were required to include both cognitive and behavioral components. Results: The literature search yielded a total of 136 papers. Preliminary results suggest that CBT treatments are superior to non-active control conditions including discussion forums, self-help books and wait list control, and slightly better than various active control conditions. There is some evidence supporting the effects of electronic and web-based CBT, and the recent upgrowth of these interventions may represent a promising and cost effective alternative to traditional CBT that requires further evaluation. Conclusion: CBT seems to be superior to non-active control conditions in the treatment of IBS, while being slightly better that active control conditions. CORRESPONDING AUTHOR: V.S. Sveinsdottir, Uni Research, Norway, [email protected]

O232 SAFETY INTERVENTIONS IN THE CONSTRUCTION INDUSTRY: A SYSTEMATIC REVIEW OF THE LITERATURE K. Sainsbury1, B.A. Mullan2, Smith L.C.1 1 The University of Sydney, COLLAROY, Australia 2 Curtin University, PERTH, Australia Introduction: Fatal and non-fatal workplace injuries are common in the construction industry. Despite this, few interventions have been designed to improve safety and for those that have their effectiveness is unclear. This systematic review investigated safety interventions in the construction industry and their effectiveness in reducing injury rates and improving safety behaviours. Method: A literature search was conducted across several electronic databases using combinations of the following search terms: construction; injury/accident/safety; occupational/industrial; health promotion; and prevention/intervention. All study designs from peer-reviewed journals using human populations aged above 18 years were eligible for inclusion. Results: Fifteen studies met the inclusion criteria: two reported on both injury rates and the uptake of safety behaviours with the purpose of injury prevention; six reported on injury rates alone; and seven reported on safety behaviours alone. All interventions resulted in improvement in at least one measured variable (injury rates and/or one or more safety behaviours). There was large variability in the study designs, injury types and behaviours targeted, outcome measures, and study quality, making it difficult to draw firm conclusions regarding the factors associated with effectiveness. Theoretical models were used to inform the intervention

S73

targets or methods in five studies, but only one of these measured and tested the components of the theory. Conclusion: All interventions were at least partially effective in improving safety and/or reducing injury rates. More rigorous, theory-driven research is needed to structure interventions and determine the mechanisms of action. CORRESPONDING AUTHOR: Dr. K. Sainsbury, The University of Sydney, Australia, [email protected] O233 ECONOMIC EVALUATION OF COMPREHENSIVE MENTAL HEALTH PREVENTION PROGRAMS FOR WORKERS IN JAPAN S. Iijima1, K. Yokoyama2, F. Kiramura.2, T. Fukuda3 1 Juntendo University, URAYASU-CITY, Japan 2 Department of Epidemiology and Environmental Health, Juntendo University Faculty, TOKYO, Japan 3 National Institute of Public Health. Center for Public Health Informatics, TOKYO, Japan Objective: We constructed a cost-benefit analysis of comprehensive mental health prevention programs in workplaces. Methods: A cross-sectional survey was conducted using a questionnaire targeting staff in charge of programs at 85 companies in Japan. The questionnaire items consisted of a total of 40 questions about the programs’ implementation status and were established on the basis of the Ministry of Health, Labor, and Welfare’s guidelines. The questions included labor, materials, outsourcing costs, overheads. A benefit analysis was conducted according to the workplace perspective. The benefit from absentees’ attendance and attendance of employees with mental discomfort was calculated. In addition, the net benefit and return on investment (ROI) was also calculated. Results: An average of 1532 employees at the 85 companies were targeted in programs. The average annual working hours of a mental program was 861 hours. The implementation rate of the mental health prevention programs was 18 % for primary prevention, 23% for secondary, and 24% for tertiary programs. Companies had an average of 3 attendees with mental discomfort, an average of 7.2 absentee, and an average of 78 days of leave per employee. The average cost for the programs was 4,095 JPY per employee, and the total benefit was 35,326 JPY per employee. The net benefit per employee was 31,231JPY, and the ROI was 7.63. Conclusion: Many employers will receive a potentially significant ROI from comprehensive mental health programs on workplace. Preliminary results of this study were presented at the 9th World Congress of the International Health Economics Association. CORRESPONDING AUTHOR: Prof. S. Iijima, Juntendo University, Japan, [email protected]

O234 PROCESS-EVALUATION OF A PROBLEM-SOLVING INTERVENTION PREVENTING RECURRENT SICKNESS ABSENCE IN WORKERS WITH COMMON MENTAL DISORDERS I. Arends1, Bültmann U.1, K. Nielsen2, W. Van Rhenen3, M.R. De Boer4, J.J.L. Van der Klink1 1 University Medical Center Groningen, GRONINGEN, Nederland 2 Norwich Business School/University of East Anglia, NORWICH, United Kingdom 3 365 Occupational Health Services, UTRECHT, Nederland 4 VU University, AMSTERDAM, Nederland Introduction: Common mental disorders (CMDs) are a major cause of sickness absence and 20-30% of the workers who return to work experience recurrent sickness absence. We developed the SHARP-at work

S74

intervention, a problem-solving intervention delivered by the occupational physician (OP), to prevent recurrent sickness absence in this population. The aim of this study was to conduct a process evaluation, alongside a cluster-randomised controlled trial, to evaluate whether the SHARP-at work intervention was implemented according to the protocol and differed from treatment in the control group. Methods: In the study participated 154 OPs and 158 workers who had returned to work after sickness absence due to CMDs. We collected process data for the intervention and control group on: recruitment, reach, dose delivered, dose received, fidelity, context and satisfaction. Multilevel regression analyses were performed to investigate whether differences between the two groups on these process components were significant. Results: Compared to the control group, participants in the intervention group more frequently had two or more consultations with the OP (OR = 3.2, 95% CI = 1.2-8.8) and completed more assignments (OR = 33.8, 95% CI = 10.4-109.5) as recommended in the intervention protocol. OPs and participants were satisfied with the intervention and rated it as applicable. Conclusion: The process evaluation showed that the SHARP-at work intervention was conducted according to protocol for the majority of the participants, well-received by OPs and participants and differed from treatment in the control group. Overall, the results provide support for implementing the intervention in practice. CORRESPONDING AUTHOR: Dr. I. Arends, University Medical Center Groningen, Nederland [email protected]

O235 IMPROVING WORK FUNCTIONING OF HOSPITAL EMPLOYEES USING A MENTAL HEALTH MODULE FOR WORKERS' HEALTH SURVEILLANCE S.M. Ketelaar1, J.K. Sluiter1, F.R. Gärtner1, L. Bolier2, O. Smeets2, C.Y. Noben2, F. Smit2, K. Nieuwenhuijsen1 1 Academic Medical Center, University of Amsterdam, AMSTERDAM, Nederland 2 Trimbos Institute, Netherlands Institute of Mental Health and Addiction, UTRECHT, Nederland Introduction: Two strategies for a mental health module for workers’ health surveillance (WHS) for hospital employees were developed. Their (cost-)effectiveness regarding work functioning compared to a control group was evaluated in a cluster-randomised controlled trial. Methods: Nurses and allied health professionals working in 1 academic hospital (N=1,731) were approached. Participants of both strategies were screened online on impaired work functioning and impaired mental health and received personalised feedback online. In the e-mental health (EMH) group, participants then received tailored advice and could follow an Internet self-help course. In the occupational physician (OP) group, positively screened participants were invited for a preventive face-toface consultation with an OP. Results: Participation rate was 32%, of which 81% screened positive. Work functioning improved in all three groups. The percentage of participants in the EMH group with relevantly improved work functioning compared to baseline was not significantly different from the percentage in the control group (3 months: both 30%; 6 months: 36% vs 28% respectively, p=0.36). More OP group than control group participants had relevantly improved work functioning (3 months: 46% vs 30%, p=0.02; 6 months: 41% vs 28%, p=0.05). The OP strategy was costeffective and at 6 months follow-up led to a €651 cost reduction per employee and a return on investment of €11 per euro invested. Conclusions: WHS guided by an OP was (cost-)effective in improving work functioning, but WHS using an EMH strategy was not. CORRESPONDING AUTHOR: S.M. Ketelaar, Academic Medical Center, University of Amsterdam, Nederland [email protected]

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

O236 IS 'PSYCHOSOMATIC CONSULTATION IN THE WORKPLACE' EFFECTIVE COMPARED TO REGULAR CARE? E. Rothermund1, R Kilian2, B Reiter3, D Mayer4, M Hölzer5, MA Rieger6, H Gündel1 1 University Clinic Ulm, ULM, Germany 2 Klinik für Psychiatrie II, Universitätsklinik Ulm, ULM, Germany 3 Betriebsärztlicher Dienst der Fa. Cassidian, an EADS Company, ULM, Germany 4 Daimler AG, Werksärztlicher Dienst Health and Safety, SINDELFINGEN, Germany 5 Sonnenbergklinik Stuttgart, ZfP Südwürttemberg, STUTTGART, Germany 6 Institut für Arbeitsmedizin, Sozialmedizin und Versorgungsforschung, Universität, TÜBINGEN, Germany Introduction: Psychosomatic symptoms often appear in the workplace. “Psychosomatic consultation”stands for an initial contact with a psychotherapeutic professional for diagnosis and counselling. Within a PCIW the intervention is transferred from regular care (RC) to “workplace”. Methods: At baseline (T1) and 12 weeks later (T2), work ability (work ability index (WAI) primary outcome) depression, anxiety, somatization, quality of life (QoL, SF-12) and work-related stress effects (Maslach Burnout Inventory (MBI), irritation) were assessed. Effects of setting on change of clinical characteristics were analysed by multivariate linear regression models with membership in setting as independent variables and controlling for selection bias by including propensity scores for different treatment groups. Results: T1: PCIW, n=173, 44.9 years, SD 10.1; 70% male; RC n=179, 39.4 years, SD 11.9; 30% male. T2 is completed in 01/2014, therefore preliminary data (PCIW n=52, RC n=58) are presented. Baseline WAI was better in PCIW (27.33 SD, 8.08) compared to RC (24.59, SD 8.61). Symptoms (depression, anxiety, somatization) were more severe in RC and QoL worse. Both groups improved WAI (PCIW 30.25, SD 8.25; RC26.57 SD 8.86), depression, anxiety, SF-12 mental health and irritation. There was no difference between groups (PCIW vs. RC), nor a group*time effect. No change could be observed for somatization, SF-12 physical health or MBI. Conclusion: Even though the user profile differs in age, gender, and disease severity the effect of the intervention seems to be similar. That implicates that the intervention can be successfully transferred into the context “work”(p<0.05). CORRESPONDING AUTHOR: Dr. E. Rothermund, University Clinic Ulm, Germany, [email protected]

O237 ORGANIZATIONAL/INDIVIDUAL FACTORS INFLUENCING PATIENT-CENTERED CARE K.I Doulougeri1, A. Montgomery1, E. Panagopoulou2 1 University of Macedonia, THESSALONIKI, Greece 2 Aristotle University, THESSALONIKI, Greece Introduction: Patient-centeredness is a key component of Quality of Care (QoC), as identified by the Institute of Medicine. Patientcentered care entails respect of individual patient preferences, needs, and values, and ensures that patient values guide all clinical decisions. The aim of the study is to explore the link between organizational characteristics, health professionals’ (HPs) well-being and the provision of patient-centered care. Methods: A survey was conducted among HPs (physicians, nurses and residents) in Greece. Self- reported data on demographic characteristics, job demands, burnout, job engagement, and patient-centered care were

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

collected. Multiple regression analyses were conducted to identify the factors associated with the provision of patient centered care separately in each group. Results: The sample consisted of 594 participants. Sixty-six percent of the sample consisted of women. The mean age of the total sample was 39,5 years (sd= 8,89). Thirty-eight percent of the sample reported burnout symptoms. Job-demands were significantly associated with burnout and engagement but not with patientcentered care. In physicians, depersonalization was a significant predictor of compromised patient-centered care, whereas in nurses, vigor was positively associated with better patient-centered care. Depersonalization and vigor were both significant predictors of patient-centered care in residents. Conclusion: This study confirms previous findings reporting an association between burnout, engagement and quality of care in HPs. Physicians and residents suffering from depersonalization are more likely to provide less patient-centered care. Preventing burnout in doctors and fostering vigor in nurses and residents can be a strategy to increase patient-centered care in hospitals. CORRESPONDING AUTHOR: K.I. Doulougeri, University of Macedonia, Greece, [email protected]

O238 PATIENT RECRUITMENT RATES IN PSYCHOSOCIAL INTERVENTIONS FOR CANCER PATIENTS: A SYSTEMATIC REVIEW C. Van Scheppingen1, M.J. Schroevers2, J. Fleer1, R. Sanderman1, J.C. Coyne1, A.V. Ranchor1 1 University of Groningen, University Medical Center, Dep of Health Psychology, GRONINGEN, Nederland 2 University of Groningen, University Medical Center Groningen, GRONINGEN, Nederland Introduction Detailed reporting of recruitment procedures for clinical trials facilitates evaluation of the generalizability of findings, design of replication studies, and estimates of likely uptake when interventions are disseminated into routine care. This systematic review examines the completeness of reporting of the recruitment process as well as recruitment rates of randomized controlled trials (RCTs) of psychosocial interventions for cancer patients and factors associated with recruitment rates. Method A systematic search identified 107 eligible psychosocial RCTs from 2000-2011 that included fatigue, pain, distress, depression, anxiety or mood as primary outcome measure. Results Of all studies, 40% reported sufficient information to calculate recruitment rate. For studies that involved recruitment by screening consecutive patients on distress, the mean recruitment rate was 19%; for studies recruiting a consecutive sample without screening 52%; and for studies with a convenience sample 64% (overall range 2.5%-83%). Studies having higher enrollment rates more often included an individual rather than group or couples intervention and used recruitment by a physician. Conclusions Doubts can be raised about the accuracy of exceptionally high reported recruitment rates in the literature. Guidelines, such as CONSORT, should provide clearer definitions and raise awareness of the importance of reporting details of the recruitment process. CORRESPONDING AUTHOR: C. Van Scheppingen, University of Groningen, University Medical Center, Dep of Health Psychology Nederland [email protected]

S75

O239 DEVELOPING COLLABORATIVE DECISION SUPPORT INTERVENTIONS FOR OLDER WOMEN (75 YEARS PLUS) WITH BREAST CANCER K.J. Lifford1, K. Brain1, K. Collins2, M. Burton2, J. Witt1, L. Caldon3, A. Edwards1, L. Wyld3, M. Reed3 1 Cardiff University, CARDIFF, United Kingdom 2 Sheffield Hallam University, SHEFFIELD, United Kingdom 3 University of Sheffield, SHEFFIELD, United Kingdom Background: Primary endocrine therapy (PET) is a recognised alternative to surgery followed by endocrine therapy for a subset of older women with oestrogen receptor positive breast cancer with limited life expectancy due to comorbidity. The best course of treatment therefore depends on individual patient preference. Older breast cancer patients are often conceptualised as passive decision-makers, but little research exists regarding their decision support needs. Methods: Semi-structured interviews were undertaken with older women (75 years plus) offered a choice of PET or surgery at five UK hospital clinics. Women’s information/support needs and preferences for involvement in treatment decisions were explored. Interview transcripts were analysed using the Coping in Deliberation (CODE) framework of decision making and coping in healthcare (Witt et al., 2012). Results: Thirty six interviews were conducted with women aged 75-98 years. Participants did not appear to struggle with the treatment decision, but often did not perceive that there was a choice to be made. Decisions were often based on past experiences and knowledge derived from family and friends (which was sometimes inaccurate). While some followed their doctor’s recommendation, others felt that they could independently make a decision. Many sought information and support from their clinician, family and friends. Conclusions: Many older women with breast cancer make collaborative (rather than passive) decisions, and may benefit from collaborative decision support tools designed to support decision making and coping within and beyond the clinic setting. An option grid (www.optiongrid.org) and an information booklet are being developed and tested with experts and patients. CORRESPONDING AUTHOR: Dr. K.J. Lifford, Cardiff University, United Kingdom, [email protected]

O240 A COUPLES-BASED SKILLS INTERVENTION FOR BREAST CANCER PATIENTS: RESULTS OF TWO RCTS T.Z. Zimmermann Technische Universität Braunschweig, BRAUNSCHWEIG, Germany The cancer diagnosis is a stressful event that poses formidable and enduring challenges to the women and to their intimate partners. A conjoint process of mutual support has been suggested to be an effective coping strategy for patients who are in committed relationships. The primary aim of these studies was to gather data on the potential benefits of a brief psychological intervention (“Side by Side”) designed to assist women and their partners to cope with breast cancer. Two randomized-controlled studies were conducted at two different time points. Study I: N = 72 couples were randomly assigned to either the couples’ component (“Side by Side”), or an information component directly after the diagnosis whereas in study II N = 45 couples were randomly assigned to 'Side by Side' or relaxation after completing the medical treatment. ““Results of study I suggest that women receiving Side by Side showed larger reductions in cancer-related distress and fear of progression post-intervention compared to women in control group. Furthermore, couples in the Side by Side intervention showed less avoidance in dealing with the cancer, more posttraumatic growth, and better communication skills and dyadic coping relative to the control group.

S76

Study II showed a reduction of cancer-related distress in both groups but only in the Side by Side group improvements of relationship satisfaction occured. Short-term changes in cancer-related functioning may be improved by enhancing couples’ dyadic coping skills. The findings highlighted the importance of helping both, the woman and her partner, to effectively cope with cancer. CORRESPONDING AUTHOR: PhD T.Z. Zimmermann, Technische Universität Braunschweig, Germany, [email protected] O241 INVOLVING YOUNG PEOPLE IN THE DEVELOPMENT OF AN ONLINE PSYCHOSOCIAL COUNSELLING CANCER INTERVENTION C. Phelps1, C. Hughes1, A. Baker1, H. French2, M. Minou1, W. Hawkins1, A. Leeuwenberg.1, P. B. Hutchings1 1 University of Wales Trinity Saint David Swansea, SWANSEA, United Kingdom 2 Tenovus the Cancer Charity, CARDIFF, United Kingdom Introduction. Young people are often reluctant to make use of traditional face-to-face counselling but have reported positive experiences of engaging with online counselling (Hanley 2012; King, Bambling, Reid & Thomas 2006). The aim of this project was to work with young people to develop and evaluate a virtual reality counselling intervention for young people affected by cancer. Methods. Phase I involved the development of an avatar-based online intervention, informed by a Young Persons’ Panel of school children recruited from a local school (n= 8) and University students/mentors (n=4). Phase II of this study is a feasibility study exploring the acceptability and effectiveness of this intervention amongst young people aged 13-25, either with a personal diagnosis of cancer or who are caring for someone with cancer. Results. Through monthly group meetings and the use of a closed Facebook page the YPP made a significant contribution to the development of the intervention; testing out the interface, voting on preferences for avatar design, and piloting recruitment materials. Phase II outcomes, due for completion in June 2014, will report acceptability, uptake and retention rates, the demographic profile of participants, and key therapeutic outcomes. Conclusions. This project has raised a number of practical and ethical issues relating to the engagement of young people in intervention development, provision of psychosocial support via online mechanisms, and the protection of both young person and counsellor in an online environment. Phase II data will provide important information on the acceptability and potential effectiveness of this intervention. CORRESPONDING AUTHOR: Dr. C. Phelps, University of Wales Trinity Saint David Swansea, United Kingdom, [email protected]

O242 SEXUAL REHABILITATION AFTER RADIOTHERAPY FOR GYNECOLOGICAL CANCER: CONSENSUS AMONG PROFESSIONALS USING THE DELPHI METHOD R.M. Bakker Leids Universitair Medisch Centrum (LUMC), LEIDEN, Nederland Introduction. This study aimed to reach consensus among professionals on patient information provision and support, regarding sexual rehabilitation after radiotherapy for gynaecological cancers. Methods. An online three-round Delphi study was conducted among 10 radiation oncologists, 10 gynaecological oncologists and 10 oncology nurses from 12 cancer centres providing radiotherapy. We questioned who should provide which instructions and support at what moment, with regard to sexuality and vaginal dilator use. Responses were measured on a

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

seven-point scale varying from 'totally disagree' to 'totally agree'. Consensus was reached when 70% of participants' answers fell within two scale categories with an interquartile range ≤1. Results. The panel agreed that information should be provided by the radiation oncologist prior to treatment start. They should inform sexually active cervical and vaginal cancer patients, younger than 70, whilst more tailored information was recommended for vulvar and endometrial cancer patients, elder patients, and patients without a partner. Preferably, psychological and practical support should be given by specifically trained oncology nurses. Participants recommended vaginal dilation for prevention of vaginal adhesions. The panel advised to start dilation around 4 weeks after treatment, to perform dilation 2-3 times a week, for 1-3 minutes, and to continue dilation for 9-12 months. The most prescribed type of dilators were commercially available plastic dilator sets. Conclusion. Consensus was reached about the content and procedure of patient information provision and support during sexual rehabilitation after radiotherapy for gynaecological cancers. Some patient groups require tailored care. Results were used to develop a sexual rehabilitation intervention. CORRESPONDING AUTHOR: Drs./ MSc. R.M. Bakker, Leids Universitair Medisch Centrum (LUMC), Nederland, [email protected]

O243 EAT: RANDOMISED CONTROLLED TRIAL TO IMPROVE NUTRITION IN HEAD AND NECK CANCER PATIENTS UNDERGOING RADIOTHERAPY B. Britton1, Amanda Baker2, Judith Bauer3, Luke Wolfenden2, Chris Wratten1, Alison Beck2, Patrick McElduff2, Gregory Carter1 1 Hunter New England Health, HUNTER REGION MAIL CENTRE, Australia 2 University of Newcastle, NEWCASTLE, Australia 3 University of Queensland, BRISBANE, Australia Background: Maintenance of adequate nutrition in Head and Neck Cancer(HNC) patients is challenging. The rigours of radiation treatment and the burden of the malignancy make it difficult to maintain sufficient nutrition. However, health behaviour interventions designed to improve nutrition in HNC patients have not yet been evaluated. Aims: The proposed trial builds on promising pilot data and evaluates the effectiveness of a dietitian delivered health behaviour intervention: Eating As Treatment (EAT) to reduce malnutrition in HNC patients undergoing radiotherapy. Research Design: A stepped wedge cluster randomised design will be used.Each site will begin the study as a control hospital providing treatment as usual. Each site will receive two days of EAT training in a randomly generated order and begin contributing to the intervention arm of the study. Intervention: EAT is based on established behaviour change counselling methods, including motivational interviewing, cognitive behaviour therapy and systems change theory. EAT is designed to improve motivation to eat despite a range of barriers and provide patients with practical behaviour change strategies. The EAT intervention will be delivered by dietitians alongside their normal consultations. Methods: 400 HNC patients, aged 18+ will be recruited from radiotherapy departments in five Australian sites. Assessments will be conducted at five time points by project officers blinded to the intervention allocation. Assessments include a nutrition assessment and self report depression screen. Treatment interruption and completion, number and length of hospital stays and dietetic contacts will be retrieved from medical records. Clinical covariates will be documented and analysed. CORRESPONDING AUTHOR: Dr. I.P. Britton, Hunter New England Health, Australia, [email protected]

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

O244 THE EFFICACY OF DIFFERENT NUTRITION LABELS FOR IMPROVING HEALTHIER FOOD CHOICES R. McNeill, E. Odishoo, D. Seesaengnom University of Auckland, AUCKLAND, New Zealand Introduction. The current evidence for the effect of nutrition lables on people's food selection is mixed. The best evidence currently supports interpretative labels, such as a traffic light system, over non-interpretative systems but the strength of this evidence is poor. Methods. This paper presents the results of two studies that have used an online randomised experimental design to assess the efficacy of different types of nutritional labelling on food choices in a hypothetical dining scenario. The context of the first study was a non-fast-food restaurant and the second study was based in a fast-food restaurant. Differences in food consumed was analysed using an ANOVA model with the label type as the independent variable and various nutritional content variables as the dependent variable, including calories, fat, and sugar content. Results.The results showed that nutritional information altered food selection more in the non-fast-food setting and that the effect was specific to people with certain characteristics. In the areas where the effect of the nutritional information was significant, the size of the effect was small. Conclusions. Although the results partially support the efficacy of nutritional information on food selection, the difference in overall energy consumption would be unlikely to result in any real health improvement. These studies provide more evidence of the relative effect of different food labelling systems and the overall conclusion that they are not likely to play a key role in the fight against obesity. CORRESPONDING AUTHOR: Dr. R. McNeill, University of Auckland, New Zealand, [email protected]

O245 A BRIEF PSYCHOLOGICAL INTERVENTION TO PROMOTE HEALTHY EATING AND REDUCE UNHEALTHY EATING IN EMOTIONAL EATERS D.B. O'Connor University of Leeds, LEEDS, United Kingdom Background.Stress may indirectly contribute to cardiovascular disease risk by producing deleterious changes in diet such as increasing high fat food intake and reduce healthy food intake. Emotional eaters are particularly vulnerable to these effects. This paper reports two studies, using a randomised controlled, daily diary design, that tested the effectiveness of Self-Generated Volitional Help Sheet (VHS) to promote healthy snacking and to reduce unhealthy snacking. Methods.Emotional eaters (n=75; n=79) were randomised to complete a self-generated VHS with explicit instruction to link stressful situations with healthy snack alternatives (experimental condition) or a VHS condition without a linking instruction (active control condition). On-line daily diary reports of daily stressors and unhealthy and healthy snacks were completed for 7 days post-intervention Results. In Study 1, hierarchical linear modelling found a cross-level interaction such that there was a significant main effect of condition such that participants in the experimental condition reported eating more healthy snacks and less unhealthy snacks during the next 7 days compared to participants in the active control condition. These findings were replicated in Study 2. Conclusions. The self-generated VHS has been found to be an effective intervention to promote healthy snacking and to reduce unhealthy snacking. Further work is required to identify the value of this tool in the longer term and to understand the mechanisms of action. CORRESPONDING AUTHOR: Prof. B. O'Connor, University of Leeds, United Kingdom, [email protected]

S77

O246 DOES LIFESTYLE COACHING OF OVERWEIGHT PATIENTS IMPROVE AUTONOMOUS MOTIVATION FOR PHYSICAL ACTIVITY AND HEALTHY DIET? G.M. Rutten1, J.J.M. Meis1, M.R.C. Hendriks.1, C. Veenhof2, S.P.J. Kremers1 1 Maastricht University, MAASTRICHT, Nederland 2 Netherlands Institute for Health Services Research (NIVEL), UTRECHT, Nederland Introduction.In Combined lifestyle interventions (CLIs) to reduce overweight and obesity, being autonomously motivated enhances the odds of maintaining higher levels of physical activity (PA) and healthier dietary behaviour. We examined the shift in types of motivation, based on SelfDetermination-Theory (SDT), to increase PA and healthy dieting among participants of a primary care CLI, and the contribution of lifestyle coaching to changes in motivational quality. Methods.Participants were recruited from 29 general practices in the Netherlands that implemented a CLI named ‘BeweegKuur’. Questionnaires measuring demographics, lifestyle coaching and motivation were sent at baseline and after 4 months. Quality of Motivation was assessed with the Behavioural Regulation and Exercise Questionnaire (BREQ-2) and the Regulation of Eating Behaviour Questionnaire (REBS). T-tests and multivariate regression analysis were used to identify predictors motivational change. Results.For physical activity, participants shifted towards a more autonomous type of motivation (i.e. controlled types of motivation decreased and autonomous types increased). Autonomy supportive coaching was generally found to predict a larger shift in autonomous types of motivation. For healthy dietary behaviour, however, no favourable changes in types of motivation were observed, and autonomy supportiveness of the Lifestyle coach did not induce any positive changes. Conclusions.Motivation of participants of a real life CLI had changed towards a more autonomous motivation after 4 months of intervention. Autonomy-supportive lifestyle coaching contributed to this change with respect to physical activity. Lifestyle coaching for healthy diet requires thorough knowledge about the problem of unhealthy dieting and solid coaching skills. CORRESPONDING AUTHOR: Dr. G.M. Rutten, Maastricht University, Nederland, [email protected]

O247 SOCIAL IMPACT OF PARENTS AND PEERS ON CHILDREN'S FRUIT AND VEGETABLE INTAKE E.L. Lehto1, C. Ray1, A. Haukkala2, E. Roos1 1 Folkhälsan Research Center, HELSINKI, Finland 2 University of Helsinki, HELSINKI, Finland Background.Infrequent fruit and vegetable (FV) intake is common among school-children in Finland. The aim was to examine whether, in addition to the parental influence, also modeling FV intake of peers at baseline predicts children’s higher FV intake in the follow-up. Methods. 424 Finnish 11-year-old children participated in the PRO GREENS project both at baseline in 2009 and in the follow-up 2010 (response rate 77%). Children filled in validated food frequency questionnaire assessing FV intake (times/day) and a validated questionnaire about social influence including modeling FV intake of mother, father, and peers, parental verbal encouragement and demand to eat FV in both study years. Associations were examined with separate linear regression analyses. Findings. Modeling FV intake of parents and peers as well as parental encouragement and demand were all associated with both children’s fruit and vegetable intake at baseline. Positive change in fruit intake was associated with modeling higher fruit intake of father and peers, higher level of encouragement from mother and parental demand. Significant

S78

predictors of positive change in vegetable intake were modeling higher vegetable intake of father and peers as well as higher level of parental demand. Both peers’ and parents’ impact on FV intake was generally stronger among girls than among boys. Discussion. Interventions which aim to increase children’s FV intake should take into account the importance of modelling peers in addition to the family’s social environment. CORRESPONDING AUTHOR: E.L. Lehto, Folkhälsan Research Center, Finland, [email protected] O248 MULTIPLE-BEHAVIOUR PERSPECTIVE ON HEALTHY EATING: RELATIONSHOP BETWEEN HEALTHY EATING AND HEALTH-, WORK AND FAMILY-RELATED BEHAVIOURS J.P. Presseau, H. Brown Newcastle University, NEWCASTLE UPON TYNE, United Kingdom Introduction: Eating healthily requires resources including time and energy for self-control. However, eating does not occur in a vacuum; individuals engage in multiple other behaviours in everyday life, each vying for limited time and energy alongside eating healthily. We hypothesised that facilitating behaviours (physical activity) help individuals eat healthily by contributing energy resources, whilst behaviours which compete for available resources, including health-related (smoking, binge drinking), work-related (employment, commuting), and family-related (childcare) behaviours undermine healthy eating generally, and particularly in dieters, who expend additional energy in selfcontrolling their eating. Methods: We used data from the 2007 and 2009 Household Income and Labour Dynamics of Australia (HILDA) population survey of adults (18-65 years). Using a random effects generalised least squares model (N=11191), controlling for demographic and social factors, we tested whether health-related, work-related and family-related behaviours were associated with healthy eating, and whether this differed between dieters and non-dieters, and between sexes. Results: In men, physical activity positively (B=0.16, SE=0.02), and binge drinking (B=-0.14, SE=0.02) and smoking negatively (B=-0.33, SE=0.02) predicted healthy eating. The same predictors applied for women, as did working (B=-0.05, SE=0.03) and having young children (B=0.10, SE=0.03). Childcare (for women) and employment (for men) particularly undermined healthy eating in dieters compared to non-dieters. Conclusions: Multiple facilitating and competing behaviours may help and hinder healthy eating, particularly for individuals actively selfcontrolling their eating such as dieters. Eating cannot be considered in isolation; a multiple behaviour change approach is advocated to account for competing and facilitating behaviours in everyday life. CORRESPONDING AUTHOR: Dr. J.P. Presseau, Newcastle University, United Kingdom, [email protected] O249 THE FEASIBILITY OF USING ELECTRONIC INTERVENTIONS TO SUPPORT CHILDCARE SERVICES' IMPLEMENTATION OF OBESITY PREVENTION PROGRAMS S.L. Yoong1, Luke Wolfenden1, Meghan Finch2, John Wiggers2 1 University of Newcastle, CALLAGHAN, Australia 2 Hunter New England Population Health, WALLSEND, Australia Background: Approximately 25% of Australian children are overweight or obese. As a substantial proportion of children attend childcare services, supporting services to deliver nutrition and physical activity promoting practices is a recommended public health strategy for obesity prevention. Electronic-based systems interventions represent a novel way to support

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

childcare services in implementing these practices. To inform the development and implementation of these interventions in childcare settings, this study assessed the current use of technology, predictors of organisational intentions to use such systems and which features would be useful in supporting service to implement obesity prevention practices. Methods:A computer assisted telephone interview (CATI) was conducted with authorised supervisors from childcare services. The CATI assessed i) services’ use of electronic devices; ii) determinants of organisational intentions to use electronic systems; and iii) what features would be useful to include in a web-based program. Associations between constructs from the Technology Acceptance Model (TAM) (perceived usefulness and perceived ease of use) and intentions to use was also assessed using 11-item questionnaire. Results: Overall 214 services (77%) consented to participating. All services except one reported using computers, whereas 40% used iPads/tablets. In the multiple regression analyses, both constructs of TAM were significantly associated with intentions to use. None of the other examined characteristics were significant. More than 80% of services rated the inclusion of education and resources including menu assessment tools and interactive videos as useful/very useful. Conclusions:This study provides novel information to inform the development and implementation of electronic interventions in childcare services. CORRESPONDING AUTHOR: S.L. Yoong, University of Newcastle, Australia, [email protected] O250 HEALTH RISK BEHAVIOR AMONG THAI YOUTH: NATIONAL SURVEY 2013 T. Sirirassamee1, Buppha Sirirassamee2 1 Faculty of Medicine, Srinakharinwirot University, NAKHONNAYOK, Thailand 2 Institute for Population and Social Research, Mahidol University, NAKHONPATHOM, Thailand Background.Although risk-taking behavior is considered a major problem among youth, there are only few youth risk survey has been reported in Thailand. The objectives of our study were to establish the prevalence of health risk behaviors among Thai youth and to compare the health risk behaviors between genders, age groups, educational status and regions. Methods.This study analyzed data from a population-based, nationally representative, cross-sectional survey of 938 youth aged between 13-24 years, sampled from Bangkok and four regions of Thailand. The 2011 Youth Risk Behavior Surveillance System questionnaire was used to measure youth risk behaviors. Results. During 30 days before the survey, 35% of respondents had rarely or never wore a motorcycle helmet and the similar percentage had rarely or never wore a seat belt.Twenty-six percent of respondents had driven when drinking alcohol one or more times. During the 12 months before the survey, more than 15% of respondents were engaged in at least one physical fight and 10% of respondents had been electronically bullied.The prevalence of current cigarette smoking, current alcohol drinking and current marijuana use were 21%, 29% and 3% respectively. Thirty-two percent of respondents were currently sexually active. Among those who ever had sexual intercourse, 56% used condom and 13% did not prevent pregnancy during their most recent sexual intercourse. Conclusions. The prevalence of health risk behaviors among Thai youth were considered high, including behaviors that contribute to unintentional injuries and violence, unsafe sexual behaviors, cigarette and alcohol consumptions. CORRESPONDING AUTHOR: Dr. Sirirassamee, Faculty of Medicine, Srinakharinwirot University, Thailand, [email protected]

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

O251 PEER AND CYBER VICTIMIZATION: DO THEY PREDICT HEALTH RISK BEHAVIORS AMONG ADOLESCENTS? A.M. La Greca, S.F. Chan University of Miami, CORAL GABLES, United States of America Introduction. Peer victimization (PV) is a critical problem linked with psychological risk. Little research has examined the health risks of PV, especially cyber PV. We examined: concurrent and prospective associations between PV (including cyber) and adolescents’ (lack of) exercise and sedentary behaviors, which are risk factors for obesity and other health problems, and whether these health risks predicted subsequent PV. We examined adolescents from predominantly Latino backgrounds, who often have high rates of obesity. Method. 1067 adolescents, 13-19 years (57% female), completed the Revised-Peer Experiences Questionnaire, CyberPeer Experiences Questionnaire, and exercise and sedentary behavior items from the Youth Risk Behavior Survey on 2 occasions, 3 months apart. Regressions examined concurrent and prospective associations between PV and health risks, controlling for gender and age. Results. Greater cyber PV was associated with less vigorous exercise (p=.02) and more sedentary behavior (e.g., hours on the computer; p=.01). Greater relational PV (e.g., social exclusion) was associated with less exercise (e.g., fewer PE classes; p=.046). Greater overt PV (e.g., being pushed, shoved) was associated with more sedentary behavior (e.g., hours watching television; p=.001). PV did not predict increased health risks; however, greater participation in groupbased exercise and computer use predicted subsequent cyber PV (p’s<.02). Conclusions. Overtly and cyber victimized adolescents may withdraw from peer-oriented physical activities. Alternatively, sedentary teens may be likely PV targets. Active adolescents may also be targets of cyber PV, suggesting that sports and PE are important settings for monitoring peer interactions. Further research should examine PV’s impact over longer time periods. CORRESPONDING AUTHOR: Professor M. La Greca, University of Miami, United States of America, [email protected] O252 MEASURING SCHOOL-RELATED STRESS AND ITS ASSOCIATION WITH SUICIDAL IDEATION IN ADOLESCENTS: THE EFFORTREWARD IMBALANCE MODEL J.L. Li1, J. Siegrist2 1 University of Düsseldorf, DÜSSELDORF, Germany 2 Senior Professorship, University of Düsseldorf, Life-Science Center, DÜSSELDORF, Germany Introduction: We attempted to apply the model of Effort-Reward Imbalance to school settings in order to measure students' psychosocial stress and analyze its association with suicidal ideation in adolescents, which were rarely investigated so far. Method: A total of 1004 Chinese students (468 boys and 536 girls) from Grade 7-12 were recruited into our cross-sectional survey. Results: Satisfactory internal consistencies for the scales for effort (5 items) and reward (11 items) were obtained (Cronbach’s alpha coefficients > 0.70). Factor analysis replicated the theoretical structure of the ERI construct (Goodness-of-Fit Index > 0.90). In this sample Chinese adolescents, 10.86 % students reported suicidal ideation during the last 6 months, which was found to be significantly related to both effort and reward. Remarkably, effort-reward imbalance was associated with elevated risk of suicidal ideation (odds ratio = 1.77, 95 % confidence interval = 1.34-2.35). Conclusions: The Effort-Reward Imbalance Questionnaire - School Version is a valid instrument for identifying sources of stressful experience among Chinese adolescents, which is observed to be associated with suicidal ideation. Further studies with prospective design and in other cultural settings are called for. CORRESPONDING AUTHOR: J.L. Li, University of Düsseldorf, Germany, [email protected]

S79

O253 ADOLESCENT SCHOOL ABSENTEEISM AND SERVICE USE IN A POPULATION-BASED STUDY K. Gärtner1, K. M. Stormark2, S. Haugland2, T. Bøe2, M. Hysing2 1 Norwegian Institute of Public Health, BERGEN, Norway 2 Centre for Child and Adolescent Mental Health and Welfare, West, Uni health, BERGEN, Norway Introduction.School absenteeism is linked to a range of health concerns, health risk behaviors and school dropout. Therefore, it is important to evaluate the extent to which adolescents with absenteeism are in contact with health care and other services. The aim of the current study was to investigate service use in adolescents with high absenteeism, both rate and type of contact in comparison to their peers. Methods. The study is based on data from a population-based study targeting all pupils in upper secondary education in Hordaland County, Norway (the youth@hordaland study) in 2012. 10 220 adolescents between the ages of 16 and 18 participated in the study. Information on service use was based on adolescents’ self-report and data on absence came from an administrative registry provided by Hordaland County Council. Results. A total of 1819 participants (17.7%) had high absence (defined as being absent 15% or more). Compared to participants with no absence, those with high absence were more likely to be in contact with all services, such as specialist mental health care (OR 3.2), public health center for youth (OR 2.3) and their GP (OR 1.9). Further, contact with several services was significantly more common among adolescents with high absence and the contacts were more frequent. Conclusion. Adolescents with high absence had increased service use and were more frequently in contact with services. This finding suggests a potential for these services to address school absenteeism and play a role in preventing school dropout and associated adverse consequences. CORRESPONDING AUTHOR: K. Gärtner, Norwegian Institute of Public Health, Norway, [email protected] O254 INJURY AND SOCIAL CORRELATES AMONG IN-SCHOOL ADOLESCENTS IN THE CARIBBEAN K.F. Peltzer, S. Pengpid Mahidol University, SALAYA, Thailand Introduction, The aim of this study was to determine estimates of the prevalence and social correlates of injury among adolescents in Caribbean countries. Methods, Cross-sectional national data from the Global School-based Health Survey (GSHS) included 11571 students primarily at the ages from 13 to 16 years from 7 Caribbean countries chosen by a two-stage cluster sample design to represent all students in grades 6-10 in each country. Results, The percentage of adolescents reporting one or more serious injuries within the past 12 months was 54.3% for all countries, ranging from 43.1% in Dominica to 59.5% in Jamaica. By major activity of all survey participants, “fall”(11.4%) was the leading external cause of injury, followed by fighting with someone (5.0%), “something fell on me or hit me”(4.9%), vehicle accident (4.3%), and attacked or assaulted or abused by someone (1.9%). The injury sustained by most students of all surveyed involved a cut, puncture, stab wound (9.5%), followed by broken bone/dislocated joint (7.8%), concussion/head injury (2.7%), burn injury (1.5%) and had a gunshot wound (1.4%). In multivariate regression analysis it was found that being male, having hunger (as an indicator for low socioeconomic status), substance use (smoking and alcohol use), psychological distress (anxiety or worried and suicide ideation) and truancy were found to be associated with annual injury prevalence rates. Conclusion, A high prevalence of injury was observed. Risk factors identified should be considered in an integrated approach to injury

S80

etiology in planning injury prevention and safety promotion activities among school children. CORRESPONDING AUTHOR: K.F. Peltzer, Mahidol University, Thailand, [email protected] O255 SWEDISH CHILDREN OF PRISONERS' MENTAL HEALTH AND QUALITY OF LIFE IN COMPARISON TO POPULATION NORMS A.H. Berman1, B. Liu1, M. Löfgren2, N. Koivumaa2, K. Engström1 1 Karolinska Institutet, STOCKHOLM, Sweden 2 Bryggan, STOCKHOLM, Sweden Introduction: Children are exposed to significant life stressors when a parent is sentenced to prison. This study investigated mental health and quality of life status for Swedish children of prisoners who participated in the FP7 EU COPING study, in comparison to recently established population norms. Method: Children of prisoners aged 7-17 (n=50) responded to a survey about having a parent in prison, including the Strengths and Difficulties Questionnaire (SDQ) and the KIDSCREEN-27. Parental assessments were available for children 14 and younger (n=36). Parents answered the WHOQOL-BREF regarding their own quality of life. Population norms were available from a random population sample of children 1116 years old (n=201) and their parents (n=202), who responded to the SDQ and KIDSCREEN-27 in self-report and parental versions. Sociodemographic variables were included. Results: Children of prisoners showed significantly worse mental health in comparison to the population sample, regarding hyperactivity, emotional symptoms, conduct problems, peer problems and Total difficulties (p”0.02). Quality of life was significantly lower regarding physical and psychological wellbeing, autonomy and parent relations, school environment and total raw score (p”0.05). Boys differed to a greater extent than girls from population respondents. Conclusions: Despite sampling limitations, Swedish children of prisoners generally can be said to have lower mental health and quality of life status than their randomly selected peers. Although Sweden is regarded as a country with extensive support resources for children with social deficits, these services do not appear to compensate for the negative effects of coping with a parent in prison. CORRESPONDING AUTHOR: Associate Prof. H. Berman, Karolinska Institutet, Sweden, [email protected] O256 POPULATION-BASED PRECONCEPTION CARRIER SCREENING: ATTITUDES AND INTENTIONS OF POTENTIAL USERS M. Plantinga, A. Buitenhuis, E. Birnie, M.A. Verkerk, K. Abbott, I.M. Van Langen, A.V. Ranchor UMCG, GRONINGEN, Nederland Background The UMCG (Groningen, NL) has developed a next-generation sequencing (NGS) test for Preconception Carrier Screening (PCS). This is the first population-based PCS test to be offered to prospective couples in the Netherlands. It covers approximately 60 severe, autosomal recessive diseases simultaneously. So far, little is known about how potential users view a population-based PCS test covering several diseases simultaneously. Before the PCS test will be offered in a pilot trial, it is important to investigate potential users’ attitudes and intentions. This study aims to receive a better understanding of the intentions of potential users towards PCS, which factors influence this intention, and to what extent different ways of formulating information influence people’s intentions. Methods Data is collected by means of an online survey among 600 potential users (300 couples). Based on the Theory of Planned Behaviour (TPB) (Ajzen, 1991) we examined potential users’ intentions to undergo the test and the factors that influence their intentions. Included determinants are: attitudes towards the offer of PCS, towards having a sick child

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

and towards terminating pregnancy; subjective norms; response efficacy; perceived barriers and perceived benefits. Information about PCS will be manipulated in a 2*2 design (positive/negative and factual/emotional). Results The data will be collected in three phases by recruitment company SSI in February. Each participant will receive a questionnaire, including the items specified above, before manipulation of information, after manipulation of information, and finally, after couples have had a discussion about the PCS test. The results will be available in April. CORRESPONDING AUTHOR: M. Plantinga, UMCG, Nederland, [email protected] O257 ANTICIPATED PRESENTATION TO HEALTH PROFESSIONALS WITH OVARIAN CANCER SYMPTOMS IN GENETIC RISK VERSUS POPULATION SAMPLES S. Smits1, J. Boivin1, U. Menon2, K. Brain1 1 Cardiff University, CARDIFF, United Kingdom 2 UCL, LONDON, United Kingdom Background: In the absence of an ovarian cancer (OC) screening programme, symptom awareness is a potential route to earlier symptomatic presentation. However, materials to support this strategy may need to be tailored according to risk. The current study explored this using the Health Belief Model. Methods: Surveys were completed by 283 women at increased genetic risk of OC and 1,043 women drawn from the general population via the International Cancer Benchmarking Partnership. Questions covered symptom awareness, anticipated presentation, health beliefs and OC worry. Structural equation modelling (SEM) was used to identify determinants of anticipated presentation in the two risk groups. Results: SEM revealed a latent construct ‘perceived threat’ that determined perceived risk (r=.35***) and OC worry (r=.98***) in women at increased risk, but only perceived risk (r=.64*) in the general population sample. High perceived threat was associated with earlier anticipated presentation for those at increased risk (-.18**), but not in the general population (r=-.03). Endorsing more benefits than barriers to presentation was associated with earlier presentation in both increased risk (r=-.34***) and general population (r=.17***) samples. Discussion: The nature and mechanism of action of health beliefs differs according to disease risk. Perceived threat comprised both affective (worry) and cognitive (susceptibility) aspects in women at risk, but only cognitive aspects in the general population sample. In those at increased genetic risk, the affective component of risk perception predominated and may override cognitive aspects of help seeking behaviour. Information materials should be tailored to minimise worry and avoid inappropriate help seeking behaviour. CORRESPONDING AUTHOR: Miss E. Smits, Cardiff University, United Kingdom, [email protected] O258 HOW DO HEALTH PROFESSIONALS COMMUNICATE RISK D U R I N G C A R D I O VA S C U L A R D I S E A S E S C R E E N I N G CONSULTATIONS? C. Keyworth1, P.A. dr Nelson1, C.A. prof. Chew-Graham2, K. Kane1, C.J. Pearce1, A. Chisholm1, C.E.M. Griffiths3, L. Cordingley4 1 Dermatology Research Centre, University of Manchester, MANCHESTER, United Kingdom 2 Research Institute, Primary Care and Health Sciences, Keele University, KEELE, United Kingdom 3 Dermatology Centre, Salford Royal NHS Foundation Trust and MAHSC, MANCHESTER, United Kingdom 4 Dermatology Research Centre, Health Psychology,University of Manchester, MANCHESTER, United Kingdom

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

Introduction. Psoriasis is associated with risk factors for cardiovascular disease (CVD) which are linked to lifestyle factors. Knowledge about their own CVD risk is important for patients who may need to make lifestyle changes to reduce that risk. Communicating information about risk remains a challenge for health professionals and little is known about the specific methods used during CVD screening consultations. We examined current methods of risk communication used by primary care practitioners when undertaking CVD screening consultations. Methods. CVD risk screening consultations (n=44) between practitioners and patients with psoriasis across 10 primary care practices in NW England were audio-recorded and analysed using content analysis. A coding frame was designed to record specific techniques used to communicate CVD risk. Results. Practitioners favoured combining verbal descriptors with numerical expressions of risk communication (n=27), over verbal descriptors alone (n=15). No examples of numerical risk communication methods alone were found. Only two participants were given personalised risk calculations. Risk information was commonly presented solely as a blood pressure reading (n=17). Risk communication was rarely followed by discussions about lifestyle changes to reduce risk (n=25). Conclusions. There was little consistency in how CVD risk information was communicated to patients. Given the current literature suggests a conflict in preferred methods of risk communication between practitioners and patients, further research should examine the most optimal methods for communicating information about CVD risk to improve patients’ perception of risk. CORRESPONDING AUTHOR: Mr. C. Keyworth, Dermatology Research Centre, University of Manchester United Kingdom [email protected] O259 PREDONATION COGNITIONS AND POSTDONATION HEALTHRELATED QUALITY-OF-LIFE OF LIVING KIDNEY DONORS: A PROSPECTIVE MULTICENTER STUDY G.A.A. Wirken1, H. van Middendorp1, C.W. Hooghof2, L.B. Hilbrands2, A.J. Hoitsma2, A.W.M. Evers1 1 Leiden University/Radboudumc, LEIDEN, Nederland 2 Radboudumc, Department of Nephrology, NIJMEGEN, Nederland Introduction: Living kidney transplantation is the preferred treatment for patients with end-stage renal disease, because of better long-term recipient and graft survival, shorter waiting list times, and better recipient quality of life than after postmortal kidney transplantation. The presurgery health-related quality of life of most donors is high, but a small percentage experiences physical or psychosocial problems after donation. Only limited research has examined predonation cognitions (motivation, expectancies, and worries) and health-related quality of life both before and after donation in living kidney donors. Method: Of an ongoing prospective multicenter study in seven Dutch transplantation centers, predonation cognitions were examined with a newly developed questionnaire, together with predonation and 6-month postdonation health-related quality of life in 127 donors. Results: Improved wellbeing of the recipient and idealistic incentives were the most reported motivations for living kidney donation. About 25% of donors reported moderate to severe worries about the wellbeing of the recipient, whereas most donors did not worry about the medical screening, surgery, or physical consequences of donation. At six months after donation, a minority experienced physical or psychosocial problems, such as pain, fatigue, concentration problems, or psychological distress. Conclusions: The wellbeing of the recipient is most prevalent both as motivation and source of worry for living kidney donors. For the subgroup of donors who experience problems in health-related quality of life six months after donation, it is important to determine predictors for their health-related quality of life after donation. CORRESPONDING AUTHOR: G.A.A. Wirken, Leiden University/ Radboudumc, Nederland, [email protected]

S81

O260 IMPACT OF LUNG CANCER SCREENING IN A HIGH-RISK POPULATION: THE UK LUNG SCREENING (UKLS) TRIAL KE Brain1, K Lifford1, B Carter1, F McRonald2, G Yadegarfar2, D Baldwin3, D Weller4, DM Hansell5, SW Duffy6, JK Field2 1 Cardiff University, CARDIFF, United Kingdom 2 University of Liverpool, LIVERPOOL, United Kingdom 3 Nottingham University, NOTTINGHAM, United Kingdom 4 University of Edinburgh, EDINBURGH, United Kingdom 5 Imperial College, LONDON, United Kingdom 6 Queen Mary University, LONDON, United Kingdom Introduction: UKLS is a population-based pilot trial of low dose CT screening in individuals at high lung cancer risk. Identifying the impact of participation can help to inform a potential UK lung screening programme. Method: A random sample of individuals aged 50-75 years was contacted using a two-stage postal survey to assess their lung cancer risk. High risk participants completed psychosocial outcome measures including cancer distress, anxiety/depression and decision satisfaction, before and two weeks after randomisation (control arm) or initial CT screening result (intervention arm). Results: Of 247,354 individuals approached, 4060 high risk individuals were randomised to CT screening or usual care. Outcome analyses indicated adverse effects of trial allocation at short-term follow-up, with higher cancer distress scores in the intervention group (p≤.001) and higher depression scores in the control group (p≤.001), although scores remained within the normal range and the absolute differences were small. Follow-up questionnaire completers were more likely to be in the intervention arm, older, male, ex-smokers, more affluent, and to have lower baseline psychosocial scores (p≤.01 for each) compared to noncompleters. A greater proportion of control arm participants were not very satisfied with their decision to participate (66%) compared to intervention arm participants (58%) (p≤.001). Conclusions: Short-term negative consequences of lung screening were not clinically significant, but this may reflect the selected nature of the sample and potentially the impact of being identified as high risk. These findings will be interpreted in light of comparative data from nonparticipants regarding the barriers to UKLS uptake. CORRESPONDING AUTHOR: Dr. E. Brain, Cardiff University, United Kingdom, [email protected] O261 THE RELATIONSHIP BETWEEN PARTICIPANT EVALUATIONS OF FAECAL OCCULT BLOOD TESTS (FOBT) AND REPARTICIPATION IN SCREENING A. Duncan1, D. Turnbull1, C. Wilson2, S.R. Cole2, J. Osborne2, I. Flight3, I. Zajac3, G.P. Young2 1 The University of Adelaide, ADELAIDE, Australia 2 Flinders Centre for Innovation in Cancer, Flinders University, BEDFORD PARK, Australia 3 Preventative Health Flagship, CSIRO, ADELAIDE, Australia Background: Adherence to regular faecal occult blood test (FOBT) screening reduces colorectal cancer (CRC) incidence and mortality. The majority of CRC research has focused on increasing initial screening uptake, however achieving high rates of ongoing participation is also necessary. Satisfaction following initial screening has been shown to predict re-participation for mammography, however the strength and duration of this effect for FOBT screening is unknown. The aim was to determine whether satisfaction with initial screening predicted reparticipation in next and subsequent rounds of CRC screening Method: Participants were 1,925 men and women, residing in South Australia and aged 50-74 years at recruitment, who were invited annually to participate in a free program of 3 rounds of FOBT screening. After each round, participants rated their overall satisfaction with the FOBT, and the extent that they agreed with several positive (e.g., worthwhile) and negative

S82

(e.g., embarrassing) statements about the screening. Analyses explored the relationship between satisfaction after each screening round and reparticipation in subsequent rounds. Results: Those who gave higher negative ratings in round 1 were significantly more likely to refuse participation in round 2. Similarly, negative experiences in round 2 predicted refusal in round 3. Analyses of patterns of participation over three rounds showed that those who gave higher negative ratings in round 1 were also significantly more likely to refuse participation in round 3. Ratings of the positive aspects of FOBT screening did not significantly predict re-participation. Conclusions: A negative screening experience has both an immediate and ongoing influence on re-participation. CORRESPONDING AUTHOR: C. Duncan, The University of Adelaide, Australia, [email protected] P262 ARE LONELINESS AND SOCIAL NETWORK SIZE ASSOCIATED WITH COGNITIVE FUNCTION IN DEPRESSED OLDER PERSONS? J.S. Kuiper1, N. Smidt1, R.C. Oude Voshaar1, S.U. Zuidema1, E.R. van den Heuvel1, H.C. Comijs2, M.L. Stek2, M. Zuidersma1, R.P. Stolk1 1 University Medical Center Groningen, GRONINGEN, Nederland 2 VU University Medical Center Amsterdam/GGZinGeest, AMSTERDAM, Nederland Introduction: Social network and loneliness are two well-established risk factors for cognitive impairments among healthy older persons. However, this association has rarely been investigated in a specific population of depressed older persons, even though depressive symptoms are associated with both social relations and cognitive function. The aim of this study is to investigate the association between social network size and loneliness in a well-identified sample of depressed older persons. Method: Data was used from the NESDO study; a case-control study, including 378 depressed older persons (60 - 90 years). Participants completed questionnaires on loneliness (de Jong Gierveld Loneliness Scale) and social network size (Close Person Inventory). Three tests of cognitive function were performed; a modified version of the Auditory Verbal Learning Test, the Stroop Colour-word test, and the subtest Digit Span from the Wechsler Adult Intelligence Scale (WAIS). Scores were summarized in four cognitive domains; memory, executive function, working memory, and processing speed. Analyses were performed adjusting for age, gender, education, alcohol use, physical activity, and severity of depressive symptoms. Results: Neither social networks size nor loneliness were significantly associated with cognitive function among this specifically identified sample of depressed older persons. Results ranged from B=0.000 to B=0.035 for loneliness and from B=-0.050 to B=0.302 for social network size (all not significant). Conclusions: Although many studies suggest that a small social network and loneliness are associated with cognitive impairment in older persons, we could not confirm this in a sample of depressed older persons. CORRESPONDING AUTHOR: J.S. Kuiper, University Medical Center Groningen, Nederland [email protected] P263 AGE-RELATED DIFFERENCES IN THE CONTROL OF WEIGHTSHIFTING WITHIN THE SURFACE OF SUPPORT E A De Vries, S.R. Caljouw, M.J.M. Coppens, K. Postema, G.J. Verkerke, C.J.C. Lamoth University Medical Center Groningen, University of Groningen, GRONINGEN, Nederland Introduction: An important reason for falling in elderly is incorrect weight-shifting. In many daily life activities quick and accurate weightshifting is needed to maintain balance and to prevent from falling. The

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

present study aims to gain more insight in age-related differences in the control of weight-shifting. Methods: Nine healthy older adults (70.3±6.9 years) and twelve young adults (20.9±0.5 years) participated in the study. They performed a weight-shifting task by moving the body’s center of pressure, represented by a red dot on a screen, in different directions, towards targets of different sizes and at different distances projected on a screen. Movement time (MT), fluency (counts on goal target and dwelling time around goal target (CoG, DT)) and accuracy (maximal deviation (MD) and number of peaks (nP) in the performed path) of the movement were determined. Results: Significant main effects of target size, target distance and age on all outcome measures were found. With decreasing target size, increasing target distance and increasing age, MT significantly increased and fluency and accuracy significantly decreased (nP, MD, CoG, DT increased). In addition, significant interaction effects of size*age and distance*age were found. Conclusion: Older adults needed more time to perform the weight-shifting task and their movements were less fluent and accurate compared to younger adults, especially with increasing task difficulty. This indicates that elderly might have difficulties with executing an adequate adaptation to a perturbation in daily life. CORRESPONDING AUTHOR: E.A. De Vries, University Medical Center Groningen, University of Groningen Nederland [email protected] P264 SOCIAL FACTORS AND COGNITIVE DECLINE AND DEMENTIA: A SYSTEMATIC REVIEW J.S. Kuiper, M. Zuidersma, R.C Oude Voshaar., S.U. Zuidema, R.P. Stolk, N. Smidt University Medical Center Groningen, GRONINGEN, Nederland Introduction: A wide variety of social factors are shown to be associated with cognitive decline and dementia, but it remains unclear to what extent social factors play a causal role in cognitive decline and dementia. In this systematic review we will determine to what extent social factors influence cognitive decline and dementia. Method: Data sources were MEDLINE, EMBASE and PsycINFO. Eligible studies consisted of longitudinal, observational studies investigating cognitive function (i.e. Neuropsychological tests or dementia diagnosis by specialist) over time (minimum 6 weeks follow-up) in healthy adults and measured the relation between social factors and cognitive function at follow-up. Two reviewers independently performed the quality assessment using an adjusted version of QUIPS (6 items) and data-extraction of all included studies. Results: The search identified8527 unique records. Finally 46 studies were included in the review of which 30 studies focusing on cognitive decline (total sample size=58517/62.1% women/mean age at baseline=72.2 years) and 14 studies on dementia (total sample size=22099/59.3% women/mean age at baseline=76.2 years). Two studies focused on both outcomes (total sample size=3259/mean age at baseline=79.9 years). The methodological quality of the included studies showed that most studies were hampered by differences between participants who completes the study and those who dropped out, and by the appropriate adjustment for certain covariates in the analyses. At conference, the results will be presented. Conclusions: Our findings will provide implications to whether it will be worthwhile to involve social factors in interventions on cognitive decline and in the evaluation, screening and treatment of dementia. CORRESPONDING AUTHOR: J.S. Kuiper, University Medical Center Groningen, Nederland, [email protected] P265 HOW TO REACH OLDER ADULTS FOR THE GRIP&GLEAM INTERVENTIONS AIMED AT IMPROVING WELL-BEING AND LONELINESS M.M. Goedendorp, R. Sanderman, N. Steverink University Medical Center Groningen, University of Groningen, GRONINGEN, Nederland

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

Introduction: The GRIP&GLEAM [Dutch: GRIP&GLANS] interventions are offered in various forms to moderately frail older adults. The target group consists of older adults who are vulnerable at psychosocial domains such as depressive mood and loneliness. However, it is difficult to reach the target group. We explored which factors influence the intention to participate in the G&G group-intervention for older women. Method: Older women were approached in malls of eight cities with short leaflets about the intervention. Of the 301 approached women 204 completed the questionnaire and interview. Based on the social marketing theory and the theory of planned behavior various factors were assessed to investigate factors associated with intention. Results: Of the women 28% indicated that they had the intention to participated. Regression analyses showed that a lower life satisfaction, a smaller social network, positive attitude, higher perceived control, and acceptable costs of the intervention, were factors associated with intention. Conclusions: Part of the target group can be reached. Women with lower life satisfaction and smaller social network seem to identify with information on the leaflet. The main message was that the intervention is for women who would like to add more ‘gleam’ to their lives. The additional information focused on coping with deterioration of health and losing a dear one. Framing the information to a more positive message might attract a different segment of the target group., e.g. women with higher life satisfaction who want to take preventive actions to avoid isolation. This will be investigated in follow-up research. CORRESPONDING AUTHOR: Dr. M.M. Goedendorp, University Medical Center Groningen, University of Groningen Nederland [email protected] P266 PACE AND PREDICTORS OF INITIAL IMPLEMENTATION OF THE G&G GROUP INTERVENTION FOR OLDER ADULTS D. Kuiper, M.M. Goedendorp, R. Sanderman, S.A. Reijneveld, N. Steverink University Medical Center Groningen, GRONINGEN, Nederland Introduction.The GRIP&GLEAM (G&G) interventions are evidencebased in improving self-management ability and well-being in older adults. Implementation of new EBPs is challenging. Arrays of factors operate on different stakeholder levels and evidence which factors are important at what phase is scarce. The aim of this study is (1) to assess how many professionals attain initial implementation and at what pace (2) to identify predictors of initial implementation in context. Method.The study is performed in routine health and social care. Theoretically informed, factors at two hierarchical levels (i.e. professionals nested in organizations) are assessed. Pace of initial implementation is described as the mean time in months and its natural variation. Predictors of initial implementation are determined by applying single- and multilevel techniques with first use (yes/no) at the professional level as the dependent variable. Results.Two third of the professionals attain initial implementation. Their mean pace is 7.5 months (SD 4.2). Six percent are early implementers (within 3 months) and 8% are laggards (longer than a year). Predictors at the professional level are ‘ownership’, ‘relative advantage’, ‘support’ and ‘compatibility’. Predictors at the organizational level are ‘size’ and ‘innovation-task orientation fit’. When analyzed in context ‘compatibility’ is shown to be the only significant predictor. Conclusion. Initial implementation of the G&G group intervention was relatively successful and done at an average pace. Compatibility at the professional level (i.e. being able to combine implementation efforts with other designated tasks) is an important predictor of initial implementation, regardless of characteristics of the organization. CORRESPONDING AUTHOR: Msc D. Kuiper, University Medical Center Groningen, Nederland, [email protected]

S83

P267 SHARED DECISION MAKING IN LONG-TERM ELDERLY CARE K. Kuipers1, C.H.C. Drossaert2, A. Dijkstra1 1 NHL Hogeschool, LEEUWARDEN, Nederland 2 University of Twente, ENSCHEDE, Nederland Background: Although Shared Decision Making (SDM) is increasingly being advocated in medical practice, little is known about client involvement in decision making in long-term care. The aim of the present study was to explore decision making and client involvement in decisions about care and daily life in so called “care plan meeting”in long-term care settings. Methods: Care plan meetings (n=27) were audio-taped and transcribed verbatim. Decision moments were selected and coded for subject of decision (open coding), category (care, treatment, living conditions and well being) and the level of SDM (using the Observer-OPTION-5-item scale) by two independent coders. In addition, a Delphi-study with focus groups of clients, their family members and care professionals was conducted to explore which factors facilitate or hinder client involvement. Results: Care plan meetings contained a mean number of 2.9 decision moments (SD 1,8) with a range from 0 till 8 decision moments per care plan meeting. In total, 77 decision moments were selected and further analyzed. Most decision moments were about care delivery (37%), followed by well-being (30%), treatment (21%) and living conditions (11%). According to the Observer-OPTION-5-scores, SDM was not common. The focus groups are currently being conducted. Preliminary results suggest that care professionals mention other barriers to SDM than clients and their family. Conclusions: Decision moments do take place in care plan meetings, but the level of SDM is low. Limitations in care budgets make SDM in elderly care even more important. Interventions to increase SDM in long-term care should be implemented. CORRESPONDING AUTHOR: drs. K. Kuipers, NHL Hogeschool, Nederland, [email protected] P268 IS THERE A LINK BETWEEN EDUCATION, RISK PERCEPTION, HEALTH OUTCOMES IN DIABETES IN THE ELDERLY? M.P. Padua, J.S. Santos, H.H. Horta Instituto Superior Técnico, LISBON, Portugal Background.The association between education, risk perception and health outcomes in secondary/ tertiary intervention is well-documented but under-analyzed in primary prevention context, particularly for the elderly. Objective. To examine among the elderly with low literacy, one of the groups more at risk of developing diabetes, whether there is a link between education, risk perception, preventive behavior, health outcomes. Methods. A cross-sectional study was carried in vulnerable urban areas of Lisbon, Portugal. Interviews with 356 subjects were conducted to assess their perceptions on risk and knowledge on diabetes. The risk of having this diseaseand habits with regard to modifiable risk factors were assessed based on Finish diabetes risk score. The age of the sample was M=62, 25; SD=15, 64. 16% of the subjects were illiterate, 73% had a basic education and 10.7% had a secondary education. Results. The risk of developing diabetes was found to correlate significantly and negatively with education (p<0.01, r=0.265). Knowledge on diabetes varied according to level of education and was low, specially for risk factors. All subjects scored low on behaviours associated with the reduction of risk and neither knowledge nor risk perception account for these (except when the subjects perceive that they are at risk they engage in health screening practices and more modifiable risk behaviours). Conclusions. Formal education levels affect health outcomes concerning diabetes. The adoption of risk modiable behaviours associated with

S84

diabetes does not seem to be associated with educational level or risk perception (except when people at risk). CORRESPONDING AUTHOR: Dr. M.P. Padua, Instituto Superior Técnico, Portugal, [email protected] P269 DETERMINANTS OF CARE DEPENDENCY B.H. Dijkstra1, A Dijkstra2, E.J. Finnema2, F.L.P. van Sonderen3, R. Sanderman3 1 NHL University AS / University of Groningen-UMCG, LEEUWARDEN, Nederland 2 NHL University AS, LEEUWARDEN, Nederland 3 Dep. of Health Sciences, Section Health Psychology, University of Groningen-UMCG, GRONINGEN, Nederland Introduction:Between 2000 and 2050, the proportion of the world's population over 60 years will double from about 11% to 22%. An associated rising prevalence of age related (chronic) diseases and consequent need for care can be seen. We face the challenge to get grip on healthcare costs and, at the same time, maintain quality of life for elderly. Function-oriented prevention of care dependency can presumably make a valuable contribution on this issue. This raises the question on which aspects we should focus when designing function-oriented preventive interventions. Purpose of this preliminary study is to map possible determinants that can predict substantial differences in care dependency levels over time. Method:As a first step in a more comprehensive literature review, the PubMed database was used to retrieve a total of 1102 articles by constructing three broad aimed searches. Search strings contained terms like care dependency, predict*, determin* and elderly. Results:A quick scan of articles showed some studies describing a relationship between care dependency and risks on falls and malnutrition. A particular study indicated care dependency and age as predictors for twoyear-survival. A few studies seem to point into the direction of care dependency at T1 predicting a change in care dependency levels at T2. Conclusions:We see clues towards the notion that care dependency levels, overall and on specific items, point towards a change of care dependency levels over time. CORRESPONDING AUTHOR: B.H. Dijkstra, NHL University AS / University of Groningen-UMCG, Nederland, [email protected] P270 LOW SOCIO-ECONOMIC STATUS AND JOB STRESS AMPLIFY THE EFFECTS OF CYTOMEGALOVIRUS INFECTION ON IMMUNOSENESCENCE J.A. Bosch1, J. Dowd2, A. Loerbroks3, P. Moss4, J.E. Fischer5, J.C. Fischer3 1 University of Amsterdam, AMSTERDAM, Nederland 2 University of Michigan, ANN ARBOR, United States of America 3 University of Dusseldorf, DUSSELDORF, Germany 4 University of Birmingham, BIRMINGHAM, United Kingdom 5 Heidelberg University, MANNHEIM, Germany Hallmark of an aging immune system (immunosenescence) is the gradual accumulation of so-called ‘terminally differentiated’ CD8+ T cells (tdCTL). This accumulation is a partial cause of declining immunity that accompanies increasing age. Infection with Cytomegalovirus (CMV) is a key determinant of tdCTL accumulation, due to the repeated reactivation of this virus which exhausts the immune system. CMV-reactivation is known to be stress-sensitive and shows a socioeconomic gradient. Hence, the present study investigated if socioeconomic status (SES) and job stress differentially predict tdCTL accumulation in CMV-infected (CMV+) versus noninfected (CMV-) individuals. Participants were 486 employees of a manufacturing company (88% male; mean age 47; range 30-65 years). tdCTL were determined by flow

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

cytometry. SES was measured as education, income and occupational status. Job stress was assessed as work-load, effort-reward imbalance, and low social support at work. Among CMV+ individuals (38%), the number of tdCTL was negatively associated with education and job status (r > -.23, p<.01, adjusting for age, gender and life style variables). Employees in the lowest educational category showed 45% higher tdCTL numbers as compared to employees in the upper category (p<.01). No such associations were found in CMVparticipants, yielding a significant Infection by SES interaction (p<.05). A comparable but weaker interaction (p<.07) was also observed for low social support at work. The current findings suggest that low SES may amplify the effects of CMV-infection on immune-aging. These findings present a novel PNI pathway, showing that infection history can moderate the associations between psychosocial factors, immunity and aging. CORRESPONDING AUTHOR: Dr. A. Bosch, University of Amsterdam, Nederland, [email protected] P271 NON-PHARMACEUTICAL DELIRIUM PREVENTION ON GERIATRIC INTERNAL WARD - EVALUATION OF INTERDISCIPLINARY VOLUNTEER-BASED WHOLESOME CONTACT PROJECT W. Kalwak1, Olesia Ivchenko1, Krzysztof Rewiuk2 1 Jagiellonian University, KRAKOWA, Poland 2 Collegium Medicum Jagiellonian University, KRAKOWA, Poland INTRODUCTION The aim of this presentation is to introduce the audience to aim and detailsand to evaluate initial two years of Wholesome Contact Projectwork. Wholesome Contact Project is an innovative, interdisciplinary, volunteer-basedprogramme for non-pharmaceutical delirium profilactics on geriatric internal ward. Delirium, which issuffering itself andmarker of poor outcome and higher mortality, is most frequent complication of hospitalization in older patients. METHODS The programme is evaluated with the use of both qualitative and quantitative methodology. The experience of care-giving over older patients of 20 volunteers (age 20-25, medical and psychology students), safety and efficiency of their voluntary work and their motivations, costs and reasons for drop-out were evaluated with the use of qualitative methodology. The efficiency and cost-effectiveness of delirium profilactics will be evaluated retrospectively (medical documentation) and in the future the randomized clinical trial is also planned. RESULTS Initial evaluation of the programme leads to the conclusion about the necessary need for supervision and psychological support for volunteers due to great emotional burden connected with the care-giving over patients in geriatric ward and frequent incidents of confrontation with death, severe ilness, difficult aging and traumatizing hospital environment. CONCLUSION During the presentation results of two-years qualitative study concerning volunteers' experience of care-giving and safety and quality of their work will be presented with the results of quantitative evaluation of programme's efficiency that are not available yet. The research design for future clinical trial will also be discussed during the session. CORRESPONDING AUTHOR: W. Kalwak, Jagiellonian University, Poland, [email protected] P271 LIVED EXPERIENCE WITH DIABETES: A PERSON-CENTERED APPROACH C. Laranjeira Piaget Institute, VISEU, Portugal Introduction:Chronic illness, such as diabetes, results in a transition process involving a variety of changes in both bodily function and

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

conditions for living, and requiring broad knowledge and understanding in order to meet new demands. The aim was to illuminate the transition of living with diabetes during the first year after diagnosis, in order to reach an understanding of how better to create more person-centred care. Design:The study has a lifeworld phenomenological approach with a transversal, descriptive and interpretive design. Twelve participants were interviewed one year after initial diagnosis. Results: Three themes were found; ‘taken over by a new reality,’ ‘the body plays a role in life’, and ‘the health care service as a necessary partner’. For a healthy transition when living with diabetes, the ability to interact with others in order to share their experiences was crucial and contributed towards participants understanding themselves and their life world. Discussion: In order to be able to interact, participants had to understand their body as a subject, interlaced with the self and the life, the lived body. Being able to listen to the body and interpret body cues and circumstantial information also contributed to a healthy transition. Patient education in a group setting, with the goal of sharing experiences in a learning process, will be meaningful only if the person has the ability to interact with others. CORRESPONDING AUTHOR: Prof. C. Laranjeira, Piaget Institute, Portugal, [email protected] P272 A THEORY DRIVEN QUALITATIVE EXPLORATION OF BEING OVERWEIGHT; COMPARING OLDER AND YOUNGER ADULTS B. Mullan1, A. Robertson2, J. Todd2 1 Curtin University, PERTH, Australia 2 University of Sydney, SYDNEY, Australia Young adults experienced the greatest increase in weight, BMI and waist circumference between 2000 and 2005 but remain underrepresented. The purpose of this study was to explore the specific factors that are relevant to weight control in overweight and obese young adults compared to older adults,. A qualitative methodology with purposive sampling was used. Semistructured interviews were conducted with 23 overweight or obese participants. The research was informed by thematic analysis. A mixed deductive-inductive approach that was structured around but not limited to TPB constructs was applied. Themes mapped onto the TPB well, with additional themes of motivation as well as knowledge and experience emerging. Differences across groups included motivators to weight loss (e.g. appearance and confidence for young adults, health for older adults), importance of social influences, and perceptions of control (e.g. availability and cost for young adults, age and energy for older adults). Similarities across groups included attitudes towards being overweight and losing weight, and the value of preparation and establishment of a healthy routine. Finally, across both groups, knowledge and confidence in ability to lose weight appeared adequate, despite failed attempts to do so. Overweight and obese young adults report some unique factors relevant to weight loss and control, and therefore may benefit from a tailored weight loss intervention. The different experiences identified can be used to inform future tailored weight-loss interventions and the TPB could provide a useful framework. Additional intervention strategies, such as improving behavioural routine and improving self-regulation also warrant further investigation. CORRESPONDING AUTHOR: Assoc. Prof. A. Mullan, Curtin University, Australia, [email protected] P273 SALIENT BELIEFS ABOUT FASTING IN BRITISH MUSLIMS WITH DIABETES: A QUALITATIVE STUDY N.R. Patel1, D. Reeves1, C. Blickem1, A.P. Kennedy2, C.A. Chew-Graham3 1 University of Manchester, MANCHESTER, United Kingdom 2 University of Southampton, SOUTHAMPTON, United Kingdom 3 Keele University, KEELE, United Kingdom

S85

Introduction: Ramadan, the ninth month of the Islamic calendar, represents one of the five pillars of Islam, during which Muslims across the world fast during daylight hours. Little is known about the beliefs of fasting in Muslims with diabetes, and the changes to their diabetes regimen. We explored the beliefs and experiences of managing diabetes during Ramadan in Muslims with diabetes, and perceived roles of General Practitioner (GP) or Practice Nurse (PN). Methods: Semi-structured interviews with 23 UK Muslims recruited using random and purposive sampling. Data were analysed thematically using a constant comparison approach. Results: Personal models about diabetes during Ramadan include the passion and motivation, despite having health problems, which was believed to make the patient a stronger person, with the grace of God. This decision involved religious, social and psychological tensions, including perceived pressure from family members, and guilt for not fasting. Participants described feeling happier, energetic, but were less likely to access health services. There was a reluctance to disclose fasting in consultation due to fears of being discouraged, with beliefs that the GP/ PN did not understand the importance of fasting. Conclusion: Participants reported fasting without discussion with their GP/PN, and with limited understanding of how this might affect their diabetes control. This has potentially serious consequences for diabetes control and future health. Research into how health-care professionals can best support Muslim patients who wish to fast is required to ensure patients can discuss fasting openly and receive best care to manage their diabetes safely whilst fasting. CORRESPONDING AUTHOR: Dr. N.R. Patel, University of Manchester, United Kingdom, [email protected] P274 CYNICAL HOSTILITY, OUTCOME BELIEFS, SELF-EFFICACY AND WEIGHT CHANGES IN A FIVE YEAR FOLLOW-UP A. Haukkala1, Hanna Konttinen1, Markku Peltonen2 1 University of Helsinki, HELSINKI, Finland 2 National Institute for Health and Welfare, HELSINKI, Finland Introduction: Obese persons have been found to have higher cynical hostility scores, but there are only few prospective studies on this topic. We examined how cynical hostility is related to outcome beliefs of behavior change and health-related self-efficacy and weight changes in five year follow-up. Method: Participants aged 45-74 years from the FINRISK 2002 Study (n=1109) were invited to follow-up study 5 years later. 80% of the high risk group for Type 2 diabetes (n=432) and 84% of the other group (n=477) participated in the follow-up phase. Weight and height were measured in a health examination. Cynical hostility (8 items), selfefficacy (5 items) and outcome beliefs (3 items) were based on selfreported questionnaires. Results: At the baseline, participants with higher cynical hostility scores had more negative outcome beliefs of behavior change (r=-.12, p<.001), but there was no association with self-efficacy. After five years, 9% in the low risk group and 14% in the high risk group lost weight more than 2 BMI units and 10% and 12% increased more than 2 BMI units. Cynical hostility, self-efficacy or outcome beliefs scores did not predict significantly weight gain or loss. There were no significant interactions with T2D risk group or gender either. Conclusion: Cynical attitudes toward others are more common among those with higher BMI, and subjects with higher hostility have more negative attitudes toward outcome beliefs of health behavior changes. However, cynical hostility do not predict weight changes among middle aged individuals with high or normal risk for Type 2 diabetes. CORRESPONDING AUTHOR: PhD A. Haukkala, University of Helsinki, Finland, [email protected]

S86

P275 DEVELOPMENT OF A FOOD RELATED QUALITY OF LIFE QUESTIONNAIRE FOR INFLAMMATORY BOWEL DISEASE L. Hughes.1, L. King2, J. O. Lindsay3, M. C. Lomer2, M. Morgan2, S. Ayis2, K Whelan2 1 Institute of Psychiatry, King's College London, LONDON, United Kingdom 2 King's College London, LONDON, United Kingdom 3 Barts Health NHS Trust, LONDON, United Kingdom Introduction: Diet and inflammatory bowel disease (IBD) has been well researched leading to improved treatment and prognosis. However, little is understood about how IBD impacts food related quality of life (FRQoL). Following qualitative studies describing the food related experience across IBD patients, FRQoL questionnaire for people with IBD was developed. Methods: 150 FRQoL items were generated from semi-structured interviews. 100 IBD patients ranked the importance of each item. Items were removed based on ceiling/floor effects and high inter-item correlations. The 41 highest ranking items were retained with a 5-point Likert (disagree-agree) response scale. Subsequently, 287 IBD patients, 100 asthma patients and 117 healthy volunteers completed the FRQoL questionnaire alongside clinical measures, MUST nutritional scoring, food satisfaction and generic and disease-specific quality of life questionnaires. Psychometric testing of the FRQoL questionnaire was carried out. Results: Principle components analysis identified seven components explaining 68.35% of variance with high internal reliability (alpha=0.96). FRQoL sumscore (higher scores indicate worse FRQoL) correlates with younger age (r=-0.12), higher MUST score (r=0.17) and worse BMI score (r=-0.13). Worse FRQoL scores were associated with female gender (p<0.001), Crohn’s Disease (p<0.05), surgery (p<0.05) and high MUST risk score (p<0.05). Conclusion: An FRQoL questionnaire has been developed to identify issues around eating for people with IBD at all stages of the disease process. Preliminary psychometric testing indicates that the questionnaire is related to clinical characteristics indicative of poor FRQoL. Further testing will determine the validity and reliability of the questionnaire. CORRESPONDING AUTHOR: Dr. D. Hughes, Institute of Psychiatry, King's College London United Kingdom [email protected] P276 PATIENTS AND PARTNERS' PSYCHOLOGICAL VARIABLES IN TYPE 2 DIABETES M. Pereira1, Ana Almeida2 1 School of Psychology, BRAGA, Portugal 2 University of Minho, BRAGA, Portugal Introduction: This study focused on differences between two assessment moments regarding variables of the theory of planned behavior concerning self-monitoring of blood glucose, satisfaction with health care, adherence to medication, metabolic control, in patients, and trust in physician, illness representations and beliefs about medicines, in patients and partners. Method: 133 diabetes patients and partners were assessed at T1 i.e. no longer than a year prior to their diagnosis (3-4 sessions of standard care for diabetes) and at T2, 4 months later (1-2 more sessions later). The instruments used were: Revised Summary of Diabetes Self-Care Activities Measure, Medication Adherence Report Scale, Planned Behavior Questionnaire, Trust in Physician Scale, Patient Satisfaction Questionnaire, Questionnaire on Beliefs about Medicines, Illness Perception Questionnaire-Brief. Metabolic control was assessed by glycosylated hemoglobin. Results: There was a significant increase from T1 to T2, in patient’s attitudes and timeline representations, and a decrease in negative beliefs about medicines and illness representations i.e. seeing diabetes as less

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

threatening regarding consequences, treatment control, comprehension and emotional response in patients and partners, and personal control, only in partners. Partners revealed higher beliefs about toxicity and overuse of medicines, and increase in representations of personal control, treatment control, concerns and emotional response, when compared with patients. Conclusion: This study showed that in recently diagnosed patients, standard education programs are not making a difference in TPB variables except attitudes. In terms of illness representations, patients perceive the disease as less threatening and the partners revealed more concerns and emotional response towards diabetes. CORRESPONDING AUTHOR: Prof. M. Pereira, School of Psychology, Portugal, [email protected] P277 FAMILY VARIABLES ON ADHERENCE AND QUALITY OF LIFE OF ADOLESCENTS WITH TYPE 1 DIABETES M. Pereira1, Ana Almeida2, Engracia Leandro3 1 School of Psychology, BRAGA, Portugal 2 University of Minho, BRAGA, Portugal 3 Universidade Católica Portuguesa, BRAGA, Portugal Introduction: This study analyses the influence of family support, family functioning and family illness’ representations on adherence and quality of life of adolescents with type 1 diabetes.Method: Sample included 100 adolescents with type 1 diabetes and a respective family member. Adolescents answered the following instruments: Self-Care Inventory (SCI) (La Greca et al., 2005), Diabetes Family Behavior Support (DFBS) (McKelvey et al., 1993) and Diabetes Quality of Life (DQoL) (Ingersoll & Marrero, 1991) and the family member, the Family Assessment Device (FAD) (Epstein, Baldwin & Bishop, 1983) and the Brief Illness Perception Questionnaire (IPQ-Brief) (Broadbent, Petrie, Main & Weinman, 2006).The results: Family Support (warmth-caring) and illness representations (personal control) predicted adolescent’s adherence to diabetes self care, explaining 31,2% of the variance. Personal control of family illness representations mediated the relationship between family support (warmth-caring) and adolescent’s adherence. Family functioning moderated the relationship between adherence and quality of life, i.e., when family functioning is low, the relationship between adherence to diabetes and quality of life is stronger.Conclusion: Family variables such as family support, family functioning and family illness’ representations are important resources that influence type 1 diabetes outcomes, in adolescents. Therefore, intervention programs should include families’s participationin order to improve adolescents’ adherence to self care and quality of life. CORRESPONDING AUTHOR: Prof. M. Pereira, School of Psychology, Portugal, [email protected] P278 DYNAMIC FATIGUE AND GLUCOSE LEVELS IN DIABETES: JOINT TRAJECTORIES OF PATTERNS AND THEIR PREDICTIVE ABILITY C. Fritschi, C. Park, L. Quinn, H. Park, A. Richardson, E.G. Collins, R. Mermelstein University of Illinois at Chicago, CHICAGO, United States of America Introduction: In patients with type 2 diabetes (T2DM), fatigue symptoms are common and have been associated with poor quality of life, decreased self-care behaviors, and most recently, shortened survival time. Fatigue is thought to result from uncontrolled glucose levels, but evidence from cross-sectional studies has never supported this relationship. Because fatigue symptoms and glucose levels vary across time, we hypothesized that this relationship might be revealed through use of continuous and concurrent measurement of glucose and fatigue across multiple days.

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

S87

Methods: 64 subjects with T2DM (mean age: 57.9, male: 50.8%) wore a continuous glucose monitor (CGM) and wrist accelerometer able to capture self-reported fatigue scores (0-10) every 2 hours throughout the day, for 5 days. Group-Based trajectory modelling was used to identify subgroups with similar patterns of fatigue and glucose levels. Determinant factors for each trajectory were estimated with multinomial logistic regression. Results: Those in the high glucose trajectory group were 7 times more likely than those in the low glucose group to be in the high fatigue trajectory group (odds ratio = 7.6, p = 0.006). Baseline self-reported physical activity, BMI, sleep did not affect these relationships. HbA1c was the only significant predictor of membership in the high glucose trajectory group. Conclusions: Over time, glucose and fatigue levels fall into distinct trajectory patterns which covary in patients with T2DM. These are among the first data to reveal that patients who fall into a high glucose trajectory are at a higher risk to also be in the high fatigue trajectory.

choice (incongruent). The difference between incongruent and congruent trials, the flanker effect, reflects the level of distraction, or conflict, experienced. Responses to different food groups (high/low fat, high/low sugar) were compared with a “low fat/ low sugar”condition serving as a baseline. Results: Results showed that (1) overall reaction times correlated with emotional eating and negative affect, (2) significant flanker effects of a similar size were found for all food group comparisons, and (3) the flanker effect magnitude positively correlated with positive affectand negatively correlated with external eating habits. Conclusion: This shows that external eaters were more effective at ignoring distraction from irrelevant food stimuli. We propose this may indicate that over time external eaters develop a controlled cognitive strategy, focusing more strongly on the target food, to help inhibit distraction from palatable foods. If this suggestion is true, this strategy would require additional cognitive resources and may be subject to failure when resources are low, as seen by periods of over consumption.

CORRESPONDING AUTHOR: PhD Fritschi, University of Illinois at Chicago, United States of America, [email protected]

CORRESPONDING AUTHOR: Ms E. Husted, University of Surrey, United Kingdom, [email protected]

P279 DIFFERENCE ON AGE AND BINGE EATING DISORDER IN PATIENTS CANDIDATES FOR BARIATRIC SURGERY F.T.P. Trotter, V. Assadi, C. Barros, V. Francesetti, M. Ríos CTO-UC, SANTIAGO, Chile

P281 ADHERENCE TO DIETARY GUIDELINES DURING PREGNANCY IS MOTIVATION ENOUGH? C. Kothe, S McPhail, B. Hill, H. Skouteris, M. Fuller-Tyszkiewicz Deakin University, BURWOOD, Australia

Binge eating disorder appears between 2% and 5% of the general population; in case of persons who have signed up for some type of program to lose weight, this sum rises to 30% (de Zwann, 2001). This research evaluated the mean differences in the BES scale in relation to the development stages of Erikson in a sample of candidates for Bariatric Surgery at the Obesity Treatment Center of Universidad Católica de Chile. A descriptive-correlational research on a purposive sampling has been conducted. 959 CTO-UC patients were recruited, with an age range of 14-74. The BES scale was used to evaluate the binge eating disorder, and the development stages of Erik Erikson were used for the categorization by age. The results showed an inversely proportional correlation according to age and the binge eating disorder with a high significance (r Pearson - 0.213 and SIG 0.01). In addition, candidates on the stage of old adulthood development presented less indicators of binge eating disorder than, middle, young adults and adolescents, according to the BES scale (Adolescents: m=15.88, Young Adults: m=15.00, Middle Adults: m=13.08 and Old Adults: m=10.16). It is concluded, that the old adult is more protected from the binge eating disorder than young adults and adolescents.

Introduction: Understanding the factors that determine women’s diet during pregnancy is necessary for the development of effective interventions to improve maternal and foetal health. Recent conceptual models suggest that motivation plays a central role in determining whether pregnant women engage in healthy nutrition behaviours; however the extent to which women are able to translate motivation into behaviour is currently unclear. Methods: We investigated the extent to which motivation to adopt/ maintain a healthy diet in pregnancy was successfully translated into adherence to dietary guidelines regarding dairy, fruit, and vegetable intake. Participants were drawn from longitudinal datasets that included pregnant women from Victoria, Australia. Women indicated their motivation to adopt/maintain a healthy diet at 16-18 weeks and adherence to dietary guidelines was measured at 32 weeks. Results: Motivation was operationalized as high scores on measures of importance, confidence and readiness; 74% of women were highly motivated to adopt/maintain a healthy diet during pregnancy However, despite high levels of motivation, these women frequently failed to translate their motivation into behaviour. Indeed, 90% of motivated women failed to meet at least one of the dietary guidelines assessed in the current study. Conclusions: Level of motivation to initiate/maintain a healthy diet during pregnancy does not necessarily translate into healthy dietary behaviours. These findings raise questions regarding the potential for interventions that are designed to improve diet during pregnancy by increasing motivation. More research is needed to understand the psychosocial factors that may underlie pregnant women’s difficulty in adhering to dietary guidelines.

CORRESPONDING AUTHOR: Prof F.T.P. Trotter, CTO-UC, Chile, [email protected] P280 EXTERNAL AND EMOTIONAL EATING DIFFERENTIALLY EFFECTS PERFORMANCE IN FLANKER TASK USING FOOD STIMULI M. Husted, A.P. Banks, E. Seiss University of Surrey, GUILDFORD, United Kingdom Introduction: External eaters demonstrate an automatic bias towards food stimuli and increased motivational drive to consume palatable foods, leading to over-consumption and weight gain. Methods: A series of experimental studies investigated external eaters’ cognitive processes when exposed to conflicting food images. A modified Flanker task using food pictures was undertaken by participants. Participants responded quickly and accurately to a central target food picture while trying to ignore flanking food pictures. The flankers either facilitated the response choice (congruent) or conflicted with the response

CORRESPONDING AUTHOR: J. Kothe, Deakin University, Australia, [email protected] P282 ATTACHMENT, ADHERNCE TO DIETARY RECOMMENDATIONS AND WEIGHT CHANGE ONE YEAR AFTER GASTRIC BYPASS SURGERY C. Hinnen1, F. Aarts1, R. Geenen2 1 Slotervaart ziekenuis, BAARN, Nederland 2 Universiteit utrecht, UTRECHT, Nederland

S88

Introduction: There is a small but considerable proportion of patients who are unable to benefit optimally from a gastric bypass operation in terms of weight loss, which is dependent on the degree to which patients are able to adopt healthy dietary recommendations. The main aim of the present study was to investigate the mediating role of adherence to dietary recommendations between: on the one hand, current and past psychiatric comorbidity, habitual interpersonal relations as reflected in attachment representations; and on the other hand, weight reduction one year after gastric bypass surgery. Materials and Methods: This longitudinal study included 105 patients applying for a Roux-en-Y gastric bypass operation. Current and past psychiatric comorbidity and patients’ attachment representations before surgery were assessd and dietary adherence at 6 and 12 months after surgery were assessed. Patients’ weight and height were collected from medical records. Multiple linear and logistic regression analyses and mediation analyses using resampling procedures were conducted. Results: Attachment anxiety was the strongest predictor of dietary adherence at both 6-months (p=.009) and 12-months (p=.006) after surgery. Moreover, bootstrapping analyses showed that dietary adherence at both 6 months (B =.48; 95% CI =.17 to .97) and 12 months (B =.29; 95% CI =.03 to .90) mediated the relation between attachment anxiety and weight loss expressed as loss of body mass index (BMI). Conclusion: More anxiously attached patients were found to report less adherence to dietary recommendations after gastric bypass surgery which was associated with less weight loss in the first year after surgery. CORRESPONDING AUTHOR: Dr. S.C.H. Hinnen, Slotervaart Ziekenuis, Nederland, [email protected] P283 RELIGION, SPIRITUAL HEALTH LOCUS OF CONTROL, AND HEALTH RISK BEHAVIORS IN AFRICAN AMERICANS C.L Holt1, D.L. Roth2, E.M. Clark3 1 University of Maryland, COLLEGE PARK, United States of America 2 Johns Hopkins University, BALTIMORE, United States of America 3 Saint Louis University, SAINT LOUIS, United States of America Introduction: Spiritual health locus of control reflects a person’s beliefs about the role of a higher power in one’s health and can reflect an active or a passive perspective. The purpose of this study was to examine the moderating role of active and passive spiritual health locus of control beliefs in health risk behaviors, including smoking status and heavy or problem drinking, in a national sample of African Americans. Method: Study participants (N=2,370) completed a telephone survey assessing religious involvement, active and passive spiritual health locus of control beliefs, and alcohol consumption and smoking status. Results: Higher religious behaviors were associated with lower odds of alcohol use in the last 30 days, and this effect was even more pronounced for those high in active spiritual health locus of control. Moderation findings differed for men and women. For men, lower religious beliefs and higher passive spiritual health locus of control were associated with more days of 4/5 or more alcoholic drinks. Conclusions: Findings have implications for pastoral counseling around negative religious beliefs and spiritual struggle, in the context of faithbased alcohol abuse programs. CORRESPONDING AUTHOR: Associate Prof. L. Holt, University of Maryland, United States of America, [email protected] P284 THE HEALTHY EATING AND LIFESTYLE IN PREGNANCY (HELP) TRIAL: LOGIC MODEL, THEORY AND MEDIATION EFFECTS S.A. Simpson, Elinor John, Dunla Cassidy, Rebecca Playle, The HELP study team Cardiff University, CARDIFF, United Kingdom

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

Maternal obesity is associated with pregnancy and birth complications. Excess weight gain during pregnancy can lead to long-term obesity. This study tested a theory based weight management intervention for obese pregnant women, which targets physical activity and healthy eating. We developed a logic model based on the theoretical pathways of intervention effects. This paper describes the logic model and associated mediation analyses. This cluster RCT recruited 598 pregnant women. Women in the intervention group attended a weekly weight management group. Control participants received usual care. Women were followed up at 36 weeks gestation and 6 weeks, 6 months and 1 year postpartum. The primary outcome is BMI at 1 year. The main results are reported elsewhere. We measured key hypothesised mediators associated with weight loss which included: self-efficacy, social support, self regulation and motivation. Other potential mediators were also explored in the qualitative analyses. We undertook mediation analyses to see if identified pathways to effect were associated with weight loss at one year. Outcomes were predicted using a hierarchical model controlling for baseline randomisation variables. This analysis was used to assess the theoretical model underlying the intervention which suggests that the group based intervention provides social support thereby boosting motivation and self-efficacy and enabling participants to set goals and plans relating to behaviours that ultimately result in weight loss. The findings of this work will contribute to the knowledge base regarding key intervention components that impact on diet and exercise behaviours in pregnant women, thereby facilitating successful post-partum weight loss. CORRESPONDING AUTHOR: Dr. S. Simpson, Cardiff University, United Kingdom, [email protected] P285 WEIGHT LOSS MAINTENANCE IN ADULTS: THE WILMA TRIAL MAIN RESULTS S.A. Simpson1, Rachel McNamara1, Chris Shaw1, Mark Kelly1, David Cohen2, Liz Randell1, The WILMA team1 1 Cardiff University, CARDIFF, United Kingdom 2 University of South Wales, WALES, United Kingdom Introduction:Weight loss maintenance interventions have had limited effectiveness with weight regain common. The purpose of this study was to assess a theory based 12-month individually tailored intervention based on Motivational Interviewing (MI) and self-monitoring. Methods:The study was a 3-arm RCT comprising 2 intervention arms which differed in amount of MI only, and a control. Individuals were followed up at 6 and 12 months. 170 obese adults aged 18-70 who had lost at least 5% body weight during the previous year were recruited. The primary outcome is BMI at 1 year. Secondary outcomes include waist circumference; waist to hip ratio; physical activity; proportion maintaining weight loss; diet; quality of life; health service resource usage; binge eating, well being and duration of participation and drop out from intervention. Results:Baseline characteristics were broadly similar between groups. Mean BMI was 34 in the control group and 35 in the intervention groups. Over 80% in all groups were weighing themselves at least once weekly. Participants indicated fairly high degrees of motivation and confidence regarding maintaining their weight. 82% in the intensive MI group attended all face to face sessions and 89% in the less intensive arm. Follow-up rates at 12 months were 90%. Conclusions:This is the first trial of motivational interviewing in a community based sample of individuals who have lost weight. The study data are currently being analysed and results are embargoed by the funder until report submission in February 2014, but will be available for the conference. CORRESPONDING AUTHOR: Dr S. Simpson, Cardiff University, United Kingdom, [email protected]

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

P286 WEIGHT LOSS MAINTENANCE IN ADULTS (WILMA) TRIAL: MEDIATORS OF WEIGHT MAINTENANCE OUTCOMES S.A. Simpson, Chris Shaw, Rachel McNamara, Mark Kelly, The WILMA team Cardiff University, CARDIFF, United Kingdom Introduction: Weight loss maintenance interventions have had limited effectiveness with weight regain common. The purpose of this study was to assess a 12-month individually tailored intervention based on Motivational Interviewing (MI) and self-monitoring. This is the first trial of a weight loss maintenance intervention in the UK. Methods: WILMA was a 3-arm individually randomised controlled trial comprising 2 intervention arms which differed in amount of therapist contact only, and a control arm (usual care). Individuals were followed up at 6 and 12 months post-randomisation. 170 obese adults aged 18-70 with a current or previous BMI of 30+ who had lost at least 5% body weight during the previous year were recruited. The primary outcome was BMI at 12 months and results are reported elsewhere. Mediators associated with weight change were also assessed and included: self-efficacy, social support, self-monitoring, habit formation and intrinsic motivation. Results: We undertook mediation analyses to see if identified pathways to effect measured at six months were associated with maintenance at one year. Outcomes were predicted using a hierarchical model controlling for baseline randomisation variables. This analysis was used to assess the theoretical model underlying the intervention which suggests the MI counselling process boosts motivation and self-efficacy, enabling participants to set goals and implementation intentions relating to behaviours that ultimately result in maintenance of weight loss. Conclusions: The findings from this work will contribute to the knowledge base around key intervention components and functions that effect diet and exercise behaviours, thereby facilitating successful weight loss maintenance. CORRESPONDING AUTHOR: Dr. S. Simpson, Cardiff University, United Kingdom, [email protected] P287 THE SOCIAL DYNAMICS OF GIVING UP SMOKING C.E. Smith University of Edinburgh, EDINBURGH, United Kingdom BACKGROUND: Recent observational studies suggest that family and friends may play a role in smoking cessation. Stop smoking interventions that attempt to increase social network support have not, however, improved quit rates, suggesting that such interventions fail to harness the underlying social processes. OBJECTIVES: To explore the social processes that shape quit attempts among smokers. DESIGN: A qualitative study, involving in-depth interviews with clients of NHS stop smoking services in three Scottish towns, exploring quitters’ experiences of stopping smoking, with an emphasis on their social relationships. Thirteen participants from diverse socio-demographic backgrounds were interviewed four weeks after quitting, and a thematic analysis undertaken. RESULTS: Participants experienced a range of reactions to their quit attempt from family and friends. Rather than passively responding to these reactions, quitters adopted three main strategies in attempting to successfully navigate their social networks and sustain their quit attempt: (1) seeking to make sense of family and friends’ responses, both in relation to their specific actions (e.g. not/smoking in their presence) and their wider social relationships; (2) managing their social networks, variously taking control of their social world (e.g. establishing a home smoking ban), and altering their interactions with these social environments (e.g. avoiding smokers); (3) using available social resources,

S89

drawing on lay stories of quitting as well as modelling their behaviour on, and judging their success against, other quitters. CONCLUSIONS: A deeper understanding of the social processes underpinning smokers’ quit attempts can inform existing accounts of smoking cessation and offer insights into improving quit interventions. CORRESPONDING AUTHOR: Mrs C.E. Smith, University of Edinburgh, United Kingdom, [email protected] P288 MOTIVATIONAL INTERVIEWING (MI) FOR HEALTH BEHAVIOUR CHANGE IN PRIMARY CARE SETTINGS: A SYSTEMATIC REVIEW K. Morton1, M. Beauchamp2, S. Spencer-Bowdage1, C. Pedlar1 1 St. Mary's University College, TWICKENHAM, United Kingdom 2 University of British Columbia, VANCOUVER, Canada Introduction: Motivational interviewing (MI) is a patient-centered approach to behaviour change that has increasingly been applied to public health settings with a focus on health promotion. This review examined the evidence for MI interventions in primary care settings with nonclinical populations to achieve behaviour change for physical activity, dietary behaviours and/or alcohol intake. We also sought to explore the specific behaviour change techniques (BCT’s) included in MI interventions. Methods: Electronic databases were searched for relevant articles and 35 papers met inclusion criteria and were included in this systematic review. Results: Approximately 50% of the included studies demonstrated positive effects. The efficacy of MI is unclear given the inconsistency of MI descriptions and intervention components. A key issue that remains unresolved is the identification of mediators of MI-based interventions, and a significant limitation is the lack of reporting and/or attention paid to training and fidelity. It appears that BCTs focusing on goals and planning, the provision of social support and feedback, and a comparison of outcomes (via a discussion of pros and cons) appear to be essential active ingredients of effective MI-based interventions across all three health behaviours targeted. Conclusions: More rigorous and systematic development and evaluation of MI interventions is required. This includes (a) greater attention to developing and piloting the intervention content using standardised descriptors and language, (b) improved study designs that isolate the effects of the intervention, (c) the use of objective measures of behaviour, and (d) the inclusion of process measures to enhance the validity of findings. CORRESPONDING AUTHOR: Dr. L. Morton, St. Mary's University College, United Kingdom, [email protected] P289 UNDERSTANDING ‘HARD-TO-REACH' MEN'S ENGAGEMENT IN WEIGHT MANAGEMENT PROGRAMMES: WHAT WORKS? L. Lozano Leeds Metropolitan University, LEEDS, United Kingdom INTRODUCTION: Men’s weight problems are a serious public health concern. However, both research and practice have failed to address the complexity of the issue, meaning that many 'hard-to-reach' men have been left with long histories of weight problems. Understanding the process by which HTR men engage in weight management interventions is therefore crucial to tackle health inequalities. METHODS: Twenty-two ‘hard-to-reach’ men who were taking part in a male-specific weight management programme were purposively sampled through a socio demographics questionnaire (N=75). Ethnographic observations and semi-structured interviews were conducted to explore how men learned about, adopted and stuck to the weight management programme. RESULTS: Most men learnt about the programme through their wives. Once the men became aware of the programme, they decided to attend

S90

when they anticipated that it would be a friendly programme and would enhance their social interaction. The flexibility of the programme (e.g. free of charge, different venues) and the fun aspect of it (e.g. “we have a good laugh”) helped the men to stick to it. My being (a) exotic (Spanish) and (b) able to play football, were among the key elements that helped to build a close relationship with the men. CONCLUSIONS: Although recruiting men into weight management interventions still remains a challenge, this study offers positive insights on how to attract men to these activities. Self-reflection of this ethnography, identified factors not normally shown to successfully engage men in health-related topics. CORRESPONDING AUTHOR: Miss L. Lozano, Leeds Metropolitan University, United Kingdom, [email protected] P290 THE IMPACT OF IMPLEMENTATION INTENTIONS FOR INDIVIDUALS WITH DIFFERING LEVELS OF PLANNING ABILITY E. Kothe, F. Poci Deakin University, BURWOOD, Australia Introduction: Prompting the formation of implementation intentions (or if-then plans) has been shown to result in significant improvements in health behaviour across a number of contexts. However, the efficacy of this intervention strategy may be dependent on individuals’ pre-existing planning skills and capacity. This study considered the impact of implementation intentions to decrease breakfast skipping for individuals with differing levels of planning ability. Method: Twenty-six young adults completed a computerised measure of planning ability, the Tower of London task, and formed implementation intentions to decrease the frequency of breakfast skipping. Breakfast consumption frequency was measured at baseline and one week follow-up. Results: Across the sample, there was a significant decrease in breakfast skipping behaviour over the course of the study (p = .01). Individuals with better planning ability at baseline demonstrated a decrease in breakfast skipping between baseline and follow-up (p = .03). However, no change in breakfast skipping behaviour was observed among individuals with poorer planning ability at baseline (p = .19). Conclusion: Implementation intentions have generally found to be an effective method of acheiving behaviour change, the benefits of these interventions do not appear to be evenly distributed across the population. Implementation intentions were only effective for those individuals who already had relatively good planning ability. This suggests that those most in need of assistance (poorer planners) may be missing out. Future research should investigate whether these interventions can be modified to better serve individuals with relatively poor planning ability. CORRESPONDING AUTHOR: E. Kothe, Deakin University, Australia, [email protected] P291 PARTICIPANT-GENERATED IMPLEMENTATION INTENTIONS: INTERVENTION FIDELITY AND EFFICACY E. Kothe, V. Tan Deakin University, BURWOOD, Australia Introduction: Implementation intentions have shown promising results in a range of health behaviours, including dietary behaviour. However, whether or not participants follow instructions on how to form them is rarely considered when evaluating of their efficacy. This study aimed to investigate whether participants adhere to instructions when forming implementation intentions, in the context of breakfast consumption. To do this the types of plans they made and number of plans they formed were examined, and the effects these factors had on intervention efficacy were analyzed.

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

Methods: Seventy-four participants completed an online questionnaire assessing their breakfast frequency during the previous week and generated implementation intentions to increase their breakfast consumption. One week later, participants completed a follow-up questionnaire about their breakfast consumption frequency. Results: A large proportion of participants did not fully adhere to instructions to form implementation intentions in the stipulated ‘if-then’ format. Despite this, participants significantly increased their breakfast consumption between baseline and follow-up. There were no significant associations between intervention efficacy and the extent of participants’ adherence to instructions. Conclusions: The results of this study suggest that implementation intentions are effective regardless of participants’ adherence to instructions. As the efficacy of implementation intentions is not attributable to reasons previously assumed, the findings raise questions regarding the mechanisms underlying these effects. However, the results of this study also suggest that implementation intentions are highly versatile, and may therefore be a promising intervention for large-scale application. CORRESPONDING AUTHOR: E. Kothe, Deakin University, Australia, [email protected] P292 PARENTAL PRACTICES AND ALCOHOL CONSUMPTION RISK BEHAVIORS IN A SAMPLE OF CHILEAN ADOLESCENTS A.H. Herrera1, P. Repetto2 1 Faculty of Dentistry, University of Chile, SANTIAGO, Chile 2 Faculty of Psychology, Pontifical Catholic University of Chile, SANTIAGO, Chile Introduction: Researchers have shown the importance of parental practices to risk behavior in adolescents. The goal of this research was to study the influence of parental practices: mother and father support, behavioral control (monitoring and presence of rules) and psychosocial control on alcohol consumption. Method: A sample of 470 adolescents attending schools in the Metropolitan Region of Santiago, Chile participated. Participants answered a selfadministered questionnaire in which they were asked about parental practices and frequency of alcohol consumption (“ever had an alcoholic drink?”, 3 or more, or 5 or more drinks in a couple of hours). Analysis were performed using descriptive and inferential statistics, using multiple and logistic regressions. Results: For those who had “ever had an alcoholic drink”: adolescents who reported psychological control by the parents are 45% more likely to have ever drunk alcohol than those who report low psychological control. But those who report a higher father support are 1,44 times more likely to have never drunk alcohol than those who report low father support. For both questions relating to the frequency of alcohol consumption, only support of the mother was a protective factor. Conclusions: Psychological control is a risk factor of alcohol consumption and father support is a protective factor. Moreover, support of the mother predicts alcohol consumption of 3 or more, or 5 or more drinks in a couple of hours. Thus, parental practices influence this behavior and reveal the importance they have for its prevention and in the development of behavioral self-regulation. CORRESPONDING AUTHOR: Prof. A.H. Herrera, Faculty of Dentistry, University of Chile, Chile, [email protected] P293 IT'S NOT AS SIMPLE AS BEING GLUTEN-FREE! PSYCHOSOCIAL PREDICTORS OF ADJUSTMENT TO COELIAC DISEASE K.J. Burke, Dominica Kazan CQUniversity, ROCKHAMPTON, Australia Coeliac disease, while incurable, is not associated with high mortality rates, incapacitation or extended periods within hospital or respite

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

settings; a strict gluten-free diet being the only available medical treatment. However, given the lifelong nature of the disease, the requirement for significant alterations to health and lifestyle behaviours, and potential for reduced quality of life, identification of mechanisms to promote positive psychosocial adjustment for sufferers is imperative. This study examined the utility of coping and health locus of control in predicting benefit finding and health-related quality of life in individuals with Coeliac disease. In a self-report online survey, 186 respondents had biopsy confirmed Coeliac disease, and had been on a gluten-free diet for a minimum of three months. Adherence to a gluten free diet did not predict adjustment, however, the participant’s relationship with their doctor did. Individuals attributed greater responsibility for determining their health status to their own behaviour (higher internal locus of control, lower avoidant coping) and lower accountability to sources such as luck, fate, chance or others. These findings suggest that a biomedical focus alone is limited in assisting individuals adjust to having conditions such as Coeliac disease. Adherence to a gluten free diet, while reducing physical symptoms, is unlikely to directly promote positive psychosocial health. Accordingly, alongside traditional medical intervention; a more holistic approach incorporating reduction of maladaptive coping; reinforcement of problemfocussed coping; and the building of collaborative relationships with health practitioners; is necessary to promote positive adjustment for individuals with Coeliac disease. CORRESPONDING AUTHOR: Dr. J. Burke, CQUniversity, Australia [email protected] P294 HPV VACCINATION KNOWLEDGE, BELIEFS, AND PRACTICES AMONG MEXICAN MOTHERS IN THE U.S. AND MEXICO R.B. Bastani1, Y.N. Flores2, B.A. Glenn2, J. Salmerón3 1 UCLA, LOS ANGELES, United States of America 2 UCLA Fielding School of Public Health and Jonsson Comprehensive Cancer Center, LOS ANGELES, United States of America 3 UIESS. Instituto Mexicano del Seguro Social, CUERNAVACA, Mexico Purpose. To compare knowledge, beliefs, and practices regarding human papillomavirus (HPV) vaccination between mothers of vaccine eligible girls in Cuernavaca, Mexico and California, U.S. Method. Data were collected via in-person interviews with Mexican-born mothers of girls (9-18 years) recruited from clinics in Cuernavaca (n=200) and California (n=200). Results. Mothers in Mexico were more knowledgeable about HPV and the vaccine than U.S. mothers, with a greater proportion having heard of HPV (94.5% vs. 65%), being aware of the link with cervical cancer (89.5% vs. 60.5%) and knowing that the vaccine is most effective prior to sexual debut (91% vs. 64%). Mothers in Mexico were less likely than mothers in the U.S. to believe that the vaccine causes more harm than good (1.5% vs 9.5%), worry about side-effects (31% vs 74%), or express concerns about future fertility problems for daughters (1% vs 19%). Vaccine initiation rates were lower among girls in Mexico (39.5%) versus the U.S. (48.5%). In multivariate analyses, having discussed the HPV vaccine with their daughter's health care provider was the most important predictor of vaccine uptake in both samples. Conclusion. Although U.S. mothers had less knowledge and more negative attitudes towards the vaccine compared to their Mexican counterparts, vaccine uptake rates were higher in the U.S. sample. This suggests that factors other than knowledge and beliefs, (such as system, clinic or provider factors), may be the main drivers of vaccine receipt in our samples. CORRESPONDING AUTHOR: Prof. R.B. Bastani, UCLA, United States of America, [email protected]

S91

P295 ASSESSMENT OF INTERNALIZED STIGMA AMONG PATIENTS WITH MENTAL ILLNESS IN TRIVANDRUM DISTRICT, KERALA James Thekkekuruvinal1, V. Raman Kutty2 1 District Mental Health Programme, TRIVANDRUM, India 2 Sree Chitra Tirunal Institute for Medical Sciences and Technology, TRIVANDRUM, India Background: This study aimed to examine the prevalence of, and factors associated with, high internalized stigma among patients with mental illness, and to compare the prevalence between the community-based care and psychiatric hospital-based care in Trivandrum district, Kerala. Methods: A cross sectional survey was conducted among 290 patients with mental illness attending the follow-up outpatient clinics of the government psychiatric hospital and 13 randomly selected mental health integrated primary care clinics in Trivandrum district, Kerala. A pretested interview structure and an adapted Malayalam version of Internalized Stigma of Mental illness Inventory (ISMI) were used for data collection. Results: The prevalence of high internalized stigma in the total sample was 36.2 percent. Treatment in community-based care (95 percent CI: 1.07-3.45), unemployment (CI: 1.21-3.92), absence of long-term friendship (CI: 1.42-4.42), no hope of cure with medication (CI: 1.66-7.91), no change in job after having mental illness (CI: 1.04-3.26), family history of mental illness (CI: 1.05-3.23), presence of other illness (CI: 1.11-4.17) and number of hospitalizations due to mental illness ≥ 4 (CI: 1.05-3.55) were significantly associated with high internalized stigma. The Malayalam version of ISMI had a Cronbach’s alpha of 0.90, split half reliability of 0.86, and a construct validity of -0.16, p=0.006. Conclusion: The findings call for urgent attention by the District Mental Health Programme to widen its coverage of rehabilitation centres, and to design its awareness campaigns and counselling sessions to focus more on addressing internalized stigma. CORRESPONDING AUTHOR: Ms. James, Thekkekuruvinal District Mental Health Programme, India, [email protected] P296 BLOOD VS FOBT: THE IMPACT OF PSYCHOLOGICAL VARIABLES AND SCREENING-DELIVERY MODELS ON CONSUMER PREFERENCES I.T. Zajac1, V. Dimitrakopoulos2, A. Duncan2 1 CSIRO, ADELAIDE, Australia 2 University of Adelaide, ADELAIDE, Australia Background: Faecal Occult Blood test screening is yet to attract optimal participation rates in Australia and elsewhere. The development of bloodbased methods may improve participation by removing test-specific barriers, such as faecal sampling. Studies have shown a preference for blood sampling but they have not considered the added complexities of blood screening methods and their potential impact on screening using this methodology. Method: N=289 community volunteers completed a survey designed to measure factors associated with FOBT screening. The survey also collected data on preferred screening method (blood or FOBT) and preferred screening delivery with participants indicating the likelihood that they would participate in each of four alternate screening scenarios for each screening method. These models ranged from a complete at-home screening model, through to a full practitioner based model.Results: Participants displayed an overwhelming preference for blood (80.8%) over stool (19.2%) sampling and both psychological (social influence, barriers, response-efficacy and susceptibility) and demographic/background variables (age and previous experience with FOBT) were associated with sample preference at both the univariate and multivariate levels. Furthermore, the complexity of the screening delivery scenario (e.g., at-home versus physician-based scenarios) significantly affected likelihood of participating in both screening technologies.

S92

Conclusions: This study extends on previous findings by exploring the psychological predictors of screening preference as well as the impact of different screening scenarios on intended participation. Although the majority of participants preferred blood sampling, the data show that intended participation in blood screening varies significantly depending upon the complexity of the screening delivery model. CORRESPONDING AUTHOR: Dr. T. Zajac, CSIRO, Australia, [email protected] P297 GENOME-WIDE DNA METHYLATION PROFILING ON SMOKING AND RHEUMATOID ARTHRITIS A.J.S. Svendsen University of Southern Denmark, ODENSE C, Denmark Although it has been well established that cigarette smoking is a major risk factor for the development of rheumatoid arthritis (RA), the molecular basis for the connection is not yet well understood. We conducted a genome-wide epigenetic profiling on identical twin pairs discordant for RA with smoking as a covariate to look for epigenetic markers differentially regulated in smoking and non-smoking twins in their development of RA. Genome-wide DNA methylation patterns were analyzed using Illumina Infinium Human Methylation 450K BeadChip allowing simultaneous quantitative monitoring of more than 480,000 CpG sites. Our data analysis identified 407 CpG sites differentially methylated in RA patients in connection with smoking with p<5e-08. Biological pathway analysis revealed a large number of functional clusters of genes involved including macrophage activation, sphingomyelin metabolic process, apoptotic cell clearance, histone demethylase activity, etc. Overall, our results indicated profound epigenetic modification in response to cigarette smoking during the development of RA. CORRESPONDING AUTHOR: A.J.S. Svendsen, University of Southern Denmark, Denmark, [email protected] P298 ARE PRACTITIONERS EQUIPPED TO SUPPORT LIFESTYLE BEHAVIOUR CHANGE IN PSORIASIS PATIENTS ? A QUALITATIVE STUDY P.A. Nelson1, L. Cordingley2, C. Keyworth1, A. Chisholm1, C.J. Pearce1, C.E.M. Griffiths3, C. Bundy1 1 Dermatology Research Centre, University of Manchester, MANCHESTER, United Kingdom 2 Dermatology Research Centre, Health Psychology,University of Manchester, MANCHESTER, United Kingdom 3 Dermatology Centre, Salford Royal NHS Foundation Trust and MAHSC, MANCHESTER, United Kingdom Introduction. Psoriasis, a long-term inflammatory skin condition is associated with unhealthy behaviours which can trigger onset/flares and are known risk factors for cardiovascular disease. It is unknown whether health-care practitioners managing people with psoriasis are familiar with the role of lifestyle factors and lifestyle behaviour change (LBC) techniques, or are confident to incorporate this information into routine consultations. The study aimed to elicit practitioners’ views about addressing LBC in psoriasis patients. Method.In-depth, qualitative interviews were carried out with 23 practitioners managing psoriasis patients in UK primary and secondary care, stratified by discipline. Interviews explored practitioners’ knowledge of and skills in LBC to support people with psoriasis and their attitudes to this work, including any perceived barriers. Data were analysed using Framework Analysis to identify key issues. Results. Practitioners viewed LBC as important in managing psoriasis but commonly believed such work was outside their role, citing obstacles such as organisational constraints and prioritisation of other goals. Data

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

suggest that limited knowledge and skills about lifestyle factors in psoriasis underpinned these beliefs, resulting in LBC being rarely addressed in consultations. Conclusions. This study identified low levels of knowledge, skills and confidence among professionals about addressing lifestyle issues in psoriasis patients and a lack of structured support in both primary and secondary care for this work. Findings will inform the development of training for staff and support materials to improve lifestyle in people with psoriasis. CORRESPONDING AUTHOR: Dr. P.A. Nelson, Dermatology Research Centre, University of Manchester United Kingdom [email protected] P299 PLANNING CONTRIBUTES TO FILL THE INTENTIONBEHAVIOUR GAP VIA ACTION CONTROL B. Reyes Fernández, R. Schwarzer Freie Universität Berlin, BERLIN, Germany Introduction: Action and coping planning are some of the most promising variables to fill the so called intention-behaviour gap. More proximate variables to health behaviour have been studied, such as action control. The degree by which planning exerts its effect on behaviour through action control is still not enough studied. It is hypothesized that for the case of physical exercise, action control mediates the effects of action and coping planning on behaviour. Methods: A longitudinal study was conducted in Costa Rica with 497 participants (mean age = 18.7 years, SD = 2.8) from humanities courses at the university. Structural Equation Modelling was used to test the model proposed. A 3 steps-mediation was specified where T1 intention and T1 action planning were expected to precede T2 coping planning, T2 action control and T2 exercise. Intention, action planning, coping planning were expected to exert their effects on exercise through action control. T1 exercise, sex and age were specified as covariates on T2 exercise. Results: The model proposed showed a good fit to the data, which indicates that intention, action planning, coping planning have indirect effects on exercise via action control. Conclusions: The effects of intentions and planning on exercise are explained via action control. CORRESPONDING AUTHOR: B. Reyes Fernández, Freie Universität Berlin, Germany, [email protected] P300 SURVEILLANCE OF SEXUAL AND REPRODUCTIVE HEALTH OF PORTUGUESE WOMEN F.M Ferreira1, S. Ferreira2, N. Ferreira2, J. Duarte1 1 Polytechnic Institute of Viseu, VISEU, Portugal 2 Hospital of Tondela-Viseu, VISEU, Portugal Background: The increase in malignancies in women is a concern, both due to the impact on their quality of life, as in health indicators; and costs. It is therefore vitally important that women attain early screening. Objectives: Our study aims to identify factors associated with womens' survelliance of their sexual and reproductive health; and examine to what extent the knowledge about self-examination and cervical cancer are related to surveillance of reproductive and sexual health. Methods: This study is of quantitative, non-experimental, descriptive, correlational and cross-sectional nature, with a sample of 522 women aged between 18 and 67 years who attend primary health care in Portugal. Results: Women who participated in the study have an average age of ≤ 30 years (25.7%), most have secondary education(38.5%), are married or in de facto unions (62.8%), are of Portuguese nationality (97.7%), are living with a partner (72.8%), and the majority have 2 children (33.9%). We verified that age groups, marital status, household, educational

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

qualifications, occupation and number of children were predictors of adherence to surveillance of reproductive and sexual health. Performing breast self-examination and knowledge about cancer of the cervix (cervical cancer) were shown to be predictors of adherence to surveillance of sexual and reproductive health. Conclusion: Healthcare professionals should adopt screening practices and promote informed and safe decisions for the health of women. Medical recommendations of self breast examination and cytology, as well as sociodemographic characteristics, should be considered by health professionals to promote the sexual and reproductive health of women. CORRESPONDING AUTHOR: F.M. Ferreira, Polytechnic Institute of Viseu, Portugal, [email protected] P301 FACTORS AFFECTING FAILURE TO QUIT SMOKING AFTER EXPOSURE TO PICTORIAL CIGARETTE PACK WARNINGS IN THAILAND N. Sirichotiratana, D. Sujirarat, P. Silpasuwan, C. Viwatwongkasem Faculty of Public Health, Mahidol University, BANGKOK, The Netherlands Objective: This study was carried out to determine whether health warning pictures (HWP) affect smoking cessation for smokers in workplaces. Method: Data from 1-year longitudinal follow-up of attempt-to-quit employees was obtained to determine if HWP affect tobacco cessation rates. Stratified simple random sampling, and Structured Equation Modeling (SEM) were employed. Results: Approximately 20% of intending-to-quit smokers were successful. The integrated model, combining internal, interpersonal factors and health warning pictures as external factors, fit the fail to quit pattern of the model. Conclusion: HWP was an external factor in the decision to stop smoking, but direct and indirect causes of failure to quit smoking were the influence of the family members. Recommendation: Future HWP should include information about factors which stimulate smokers to quit successfully. Keywords: health warning pictures, workplace cessation, failure to quit smoking, structured equation modeling. CORRESPONDING AUTHOR: Assoc. Prof. N. Sirichotiratana, Faculty of Public Health, Mahidol University, The Netherlands, [email protected] P302 RCT: COMPARING EFFECTS OF A MULTIDISCIPLINARY INTERVENTION MODEL TO TAU, IN PATIENTS WITH MUSCULOSKELETAL PAIN R.B. Brendbekken1, A.H. Harris2, T.T. Tone Tangen3, A.G. Grasdal4, H.R.E. Eriksen5 1 Sykehuset Innlandet hospital trust, OTTESTAD, Norway 2 Department of health Promotion and Development, University of Bergen, BERGEN, Norway 3 Department of Clinical Medicine, University of Bergen, BERGEN, Norway 4 Institute of Economics, University of Bergen, BERGEN, Norway 5 Uni Health, Uni Research, BERGEN, Norway Introduction. Musculoskeletal pain is common in the general population. These conditions are multicausal. This suggests that the optimal treatment may be multidisciplinary. The scientific literature on such treatment is, however, inconclusive, mainly due to high diversity in the content of the multidisciplinary programs but also lack of high quality studies. The main outcome of this project is to test if a new multidisciplinary intervention (MDI) will be more effective than treatment as usual (TAU) within the specialist health care, in helping patients sick-listed with unspecific musculoskeletal pain conditions back to work. Secondary

S93

outcomes are effects on health complaints, coping, function level, activity of daily living. Methods. The study is a randomized controlled clinical trial including 300 patients into two groups: The MDI-group with the new tool ISIVET: Interdisciplinary Structured Interview with use of a Visual Educational Tool, and the TAU-group. The new psychosocial instrument ISIVET plays a key role in the MDI. It is a practical instrument where the patient and the therapists collaborate to visually map different problems the patient experience into a star plot. The visualisation is the foundation for a dialogue between patient and therapist on possible actions for the patient. The ISIVET promotes shared-decision-working. Results. The new MDI-method has significant effects on anxiety and depressive symptoms, sensitivisation and function level and coping at 3 months compared to TAU. The effects on return to work by 12 months follow-up was not significant. The results indicates that ISIVET may represent an important contribution in the multidisciplinary, therapeutic work. CORRESPONDING AUTHOR: Chief Physician R.B. Brendbekken, S y k e h u s e t I n n l a n d e t h o s p i t a l t r u s t N o r w a y, [email protected] P303 HOW CAN A BRIEF INTERVENTION CONTRIBUTE TO COPING WITH BACK PAIN? A FOCUS GROUP STUDY E.R. Ree1, A. Harris2, A. Indahl3, T.H. Tveito1, K. Malterud4 1 Uni Research, BERGEN, Norway 2 University of Bergen, Department of Health Promotion and Development, BERGEN, Norway 3 Vestfold Hospital Trust, Department of Research and Development, STAVERN, Norway 4 University of Bergen, Department of Global Public Health and Primary Care, BERGEN, Norway Introduction: A brief back pain intervention based on non-directional communication of a non-injury model reduces sick leave, but little is known about why it is perceived as helpful by the participants. This study explores how individuals who had participated in the intervention, perceived connections between the intervention and their subsequent coping. Methods: Three focus-group discussions were conducted with a sample of ten employees, who had participated in a brief intervention for back pain, perceived the intervention as helpful, and had returned or remained at work subsequent to the intervention. Systematic text condensation was used for the analysis. Preliminary results: The participants considered information delivered in a comprehensible way by trustworthy experts as helpful aspects of the intervention. Understanding why they felt pain and that it was not a sign of a serious disease, changed the participants’ perception of how they could live with it. The participants now exceeded their previous limits, and dared to do activities they previously had avoided due to fear. Conclusion: Information based on a non-injury model delivered in a comprehensible way, with health professionals perceived as trustworthy experts, helped participants to cope with their pain. These seem to be important aspects of brief back pain interventions. CORRESPONDING AUTHOR: M.D. E.R. Ree, Uni Research, Norway, [email protected] P304 MEDICAL STUDENTS AND DISSECTION CADAVERS - ATOUGH RELATIONSHIP TO MASTER V.T. Tananska1, Stefana Y. Uzunova2 1 Medical University-Plovdiv, PLOVDIV, Bulgaria 2 BA, MSc, PLOVDIV, Bulgaria D-r Valeria Tananska (MD, MSc, PhD), Department of Anatomy, Histology and Embryology, Medical University-Plovdiv, Bulgaria Stefana Uzunova (BA, MSc), Researcher

S94

Introduction: Learning from the dead to benefit the living is THE ‘Grand Tour’ for medicine. Unlike its cultural counterpart however, the process is neither pleasurable, nor relaxing. It is a rough awakening to the realities of the Profession, away from normal social interaction. Some students faint, others become uncommonly aggressive, while a third group skips anatomy classes altogether. Behavioral issues of this kind endanger learning outcomes. Research was therefore launched to uncover the level of anxiety in the dissection room. Method: Test subjects were 414 first and second-year medical students at Medical University-Plovdiv. The evaluation method was self-reporting psycho-somatic tools (HAM-A, Tananska’s Tests on Anatomy Dissection - TADT) and biochemical probe taking (salivary α-amylase and chromogranin A). Results: Only a third of all testees fell within normal anxiety levels. Revealed were three reasons why: no pre-medical program, lack of comprehensive entry exams (beyond biology and chemistry) and missing in-university guidance. Conclusion: Better knowledge acquisition and higher quality of postgraduation medical services require changes in the way EU's medical education is being conducted. CORRESPONDING AUTHOR: MD, MSc, PhD V.T. Tananska Medical University-Plovdiv Bulgaria [email protected] P305 EFFECTIVENESS OF A VIDEO- AND TEXT-BASED COMPUTERTAILORED SMOKING CESSATION INTERVENTION AMONG LSES AND HSES SMOKERS N. Stanczyk1, Catherine Bolman2, Jean W.M. Muris1, Math JJM Candel1, Hein De Vries1 1 Maastricht University, MAASTRICHT, Nederland 2 Open University of the Netherlands, HEERLEN, Nederland Introduction In western countries smoking prevalence rates are high among unmotivated people and people with a lower socioeconomic status (LSES). Multiple computer tailoring and the use of visual aids may improve such interventions and increase cessation in LSES smokers. This study assessed the 12-month effectiveness of a video- and text-based computer-tailored intervention, and potential differential effects for subgroups varying in socioeconomic status (SES) and motivation to quit. Methods A randomized controlled trial was used in which smokers were allocated to the video-based condition (N=670), the text-based condition (N=708) or the control condition (brief generic text advice) (N=721). After 12 months, self-reported prolonged abstinence was investigated and biochemically verified in respondents who indicated to quit smoking. Multiple imputation analyses were used as primary analyses whereas complete case and a negative scenario (respondents lost to follow-up considered as smokers) were conducted as secondary analyses. Results Primary analyses with a multiple imputed dataset revealed that the video-based computer-tailored intervention was more effective on prolonged abstinence compared to the control condition (OR=1.897, p=.005) and the text-based condition (OR=1.712, p=.011). No differential effects were found for people with different SES levels and different motivational levels. Results were confirmed in the complete cases and negative scenario analyses. Conclusion The video-based computer-tailored intervention is effective to help smokers in achieving long-term smoking abstinence. CORRESPONDING AUTHOR: E. Stanczyk, Maastricht University, Nederland, [email protected] P306 Patient Experiences Participating in Primary Care-based Care Management J. Holtrop1, G. Piatt2, Q. Chen3, D. Hershey3, K. Adams3 1 University of Colorado Denver, AURORA, United States of America 2 University of Michigan, ANN ARBOR, United States of America 3 Michigan State University, EAST LANSING, United States of America

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

Introduction: Care management is a new complex intervention is US primary care. We sought to understand how patients perceived this new intervention and those who accept versus decline care management. Methods: First, we surveyed patients at 11 practices who had been offered care management in the past year and were present for a care visit during the 2 week study recruitment period. We sought to determine if patients in the two groups were different in the Patient Assessment of Chronic Illness Care (PACIC) and the Patient Activation Measure (PAM). Second, we selectively interviewed patients to better understand their experience. Statistical and qualitative analysis were conducted. Results: 271 surveys were completed (73% response rate) and 29 interviews. A key finding was that patients were confused. Forty percent of those who had care management did not know they had it and 37% thought they had it and did not (as confirmed by medical record review). Those who thought they had care management rated their chronic illness care higher (4.01 versus 3.68; p=.014) than those who did not, but there was no difference in patient activation. Reasons patients reported not participating in care management were lack of awareness that they had been offered it; feeling it was not needed, and that they wanted to work on their chronic condition by him/herself. Patients participating in care management generally had positive experiences and reported making improvements in self-management. Conclusion: Clarity in the role and function of the care manager within primary care is needed. CORRESPONDING AUTHOR: Associate Prof PhD Holtrop, University of Colorado Denver, United States of America, [email protected] P307 THE EFFECTS OF ILLNESS CONCEPTIONS ON THE PATIENTS' QUALITY OF LIFE S.K. Kreitler1, F. Barak.2, M.M. Kreitler1 1 Tel Aviv University, TEL-AVIV, Israel 2 Barzilai Medical Center, ASHKELON, Israel Background: Illness conceptions are patients' opinions in regard to medical diagnoses, e.g., concerning the origin, course, and effects of diseases. These conceptions reflect socially prevalent ideas rather than medical facts, and have gained in importance due to the increased emphasis on patients' empowerment. They are of special significance for cancer patients whose disease still has negative social connotations. Purpose: To explore relations between illness conceptions and quality of life in cancer patients. Methods: Participants were 210 patients of both genders, with mean age of 56.4 years, mean disease duration of 3.7 years, and different cancer diagnoses (disease stages I-IV). They were administered the Illness Conceptions Inventory which assesses attitudes concerning attribution of diseases, its manifestations and effects, and the Multidimensional Quality of Life questionnaire (QOL), which yields an overall score and scores for 15 scales (e.g., social, emotional functioning). . Results: The following Illness Conception variables were related negatively to QOL: the number of domains affected by cancer; beliefs that cancer occurrence depends on the individual’s psychological tendencies, that it is a total phenomenon, persisting beyond symptoms, that its course is unpredictable, and that it has a “personal character”. The following variables were related positively to QOL: beliefs that disease course depends on the individual’s health behaviors, that recovery is possible and that the major healing power lies with the doctor. Conclusions: The evidence that patients’ beliefs affect their QOL indicates the importance of communicating to patients beliefs and information supporting positive effects on their QOL. CORRESPONDING AUTHOR: Prof S.K. Kreitler, Tel Aviv University, Israel, [email protected]

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

P308 WHICH DETERMINANTS EXPLAIN SMOKING INTENTIONS AMONG BOYS AND GIRLS OF HIGH AND LOW SOCIOECONOMIC STATUS? P. Cremers, L. Mercken, H. De Vries, A. Oenema Maastricht University, MAASTRICHT, Nederland Introduction.Socioeconomic status and gender are known to be associated with the intention to smoke. The present study aims to identify potential differences in socio-cognitive determinants among boys and girls of a high socioeconomic status (HSES) and a low socioeconomic status (LSES). Method. Data for this longitudinal study were gathered at baseline and one year follow-up of a smoking prevention intervention study. A total of 1,801 children (aged 10-12 years) completed a web-based questionnaire concerning smoking related factors. To test for potential differential effects of SES and gender on the predictor variables (i.e. attitude, social influence and self-efficacy expectations) interaction terms were included in the analyses. Results. A 3-way interaction effect was present (OR: 0.19; 95% CI: 0.041.00), and therefore subgroup analyses were warranted. HSES boys had a higher intention to smoke if they had a positive attitude toward smoking, whereas HSES girls had a higher intention to smoke if they perceived less disadvantages of smoking (OR: 0.42; 95% CI: 0.22-0.82). LSES boys’ intention to smoke was predicted by their perceived social norm (OR: 0.49; 95% CI: 0.25-0.93), for LSES girls the smoking behavior of people in their environment was related to the intention to smoke (OR: 5.55; 95% CI: 2.81-10.93). Conclusions. To prevent youth smoking, HSES children may benefit from interventions that address attitude beliefs (i.e. pros and cons of (non-)smoking), whereas LSES children may benefit most from smoking prevention interventions that provide strategies to resist the influence of smoking models in their environment. CORRESPONDING AUTHOR: H.P. Cremers Maastricht University Nederland [email protected] P309 POPULATION-BASED, WHOLE OF COMMUNITY INTERVENTIONS TO PREVENT POPULATION WEIGHT GAIN: A META-ANALYSIS L. Wolfenden1, R. Wyse1, M. Nicholas2, S. Allender.2, L. Millar2, P. McElduff1 1 The University of Newcastle, NEWCASTLE, Australia 2 Deakin University, GEELONG, Australia Introduction: Population-based, whole of community interventions addressing multiple determinants of obesity have been recommended.A systematic review was conducted to assess the effectiveness of population-based whole of community interventions in preventing excessive population weight gain. Methods: Electronic databases, and hand searches of key journals, references lists of included trials were conducted and data was extracted by 2 reviewers independently. Population-based, whole of community interventions were defined as those targeting the weight status of a population characterised along geographical boundaries. Trials with an objective measure of adiposity and a parallel control group were included. Results: Eight trials were identified as eligible. All trials targeted children or adolescents and utilised quasi-experimental designs. Most trials reported a significant intervention effect on at least one measure of adiposity. Meta-analysis revealed a a small reduction in BMI z-score among participants in intervention communities (mean difference (MD) -0.09; 95% confidence interval (CI) -0.16 to -0.02). Conclusion: The review suggests such interventions achieve modest reductions in population weight gain among children. CORRESPONDING AUTHOR: Dr. L. Wolfenden, The University of Newcastle, Australia, [email protected]

S95

P310 FUNCTIONING ASSESSMENT VERSUS CONVENTIONAL MEDICAL ASSESSMENT: CONSEQUENCES OF A NEW HEALTH CARE PARADIGM H.A. Stallinga1, P.F. Roodbol1, C. Annema1, G.J. Jansen2, K. Wynia1 1 University Medical Centre Groningen, GRONINGEN, Nederland 2 Hanze University of Applied Sciences, GRONINGEN, Nederland Introduction: In chronic diseases, pathogenic-oriented healthcare falls short in generating all the information required for determining healthcare provision to promote health. A broader, so-called salutogenic approach, by using the International Classification of Functioning, Disability and Health (ICF), focussing on how to stay healthy, rather than what causes diseases, seems more appropriate.The purpose of this study was to compare a functioning assessment based on the ICF with a conventional medical assessment, in terms of their respective consequences for health professionals’ clinical decision-making and on the fit with patient’s own perspective of health. Method. Data about patient problems and professional healthcare activities were collected from 81 patients with severe Multiple Sclerosis who were randomly assigned to one of two groups: the ICF group, assessed with a functioning assessment (n=43), and the medical group, assessed with a conventional medical assessment (n=38). Results. A functioning assessment resulted in registration of significantly more patient problems in health components ‘Participation’ and ‘Environmental Factors’, as well as significantly more professional healthcare activities befitting these components. The ICF group had a significant positive correlation between registered problems by health professionals and patients’ self-reported problems whereas the medical group had several negative correlations. Conclusions. A functioning assessment resulted in a care plan which was not only broader and more complete but which also reflected the patients’ self-reported problems more closely than a medical assessment. This corresponds to a new paradigm of health care especially for the chronically ill focusing on how to stay healthy, rather than what causes diseases. CORRESPONDING AUTHOR: H.A.S. Stallinga, University Medical Centre Groningen, Nederland, [email protected] P311 PSYCHOSOCIAL SUPPORT AND BURNOUT AT PHYSICIANS ATTENDING ADVANCED CANCER PATIENTS: THE IMPACT OF BALINT TRAINING O. Popa-Velea1, D. Pamfile2, I. Popp3 1 University of Medicine and Pharmacy „Carol Davila„, BUCHAREST, Romania 2 Department of Consultation-Liaison Psychiatry, University Hospital of Lausanne, LAUSANNE, Switzerland 3 University Hospital of Freiburg, FREIBURG, Germany Introduction: Along with medical responsibilities, offering psychosocial support (PS) can represent a difficulty for untrained health professionals. The aim of this paper was to evaluate the impact of participating in Balint groups on physicians working daily with advanced cancer patients, in terms of (a)their willingness to offer PS, (b)the way they define PS and (c)their own perceived psychosocial support and burnout. Method: Participants were 89 physicians addressing advanced cancer patients (42♂,47♀; mean age=53.58, SD=6.19), with 38 (20♂,18♀), regularly attending 20 Balint meetings in the last 3 years (the Balint subgroup). Three rounds of a Delphi process were run, in order to provide converage on their offer of PS and the keywords best describing PS. All participants completed the Maslach Burnout Inventory and the DukeUNC Functional Social Support Questionnaire, to test (via analysis of variance and t-tests) the differences attributable to Balint affiliation. Results: Offering PS was significantly more frequent in the Balint group (89,47% vs. 54,90%, p<.05). Five categories of keywords defining PS

S96

were identified, with Balintian physicians providing a significantly higher number of keywords related to communication issues (p<.05) and social support (p<.01). Perceived psychosocial support was significantly higher (p<.05) and scores of two components of burnout were significantly better (depersonalization and low personal accomplishment: p<.01, emotional exhaustion: p<.05) at Balint groups attendees. Conclusion: Results support the use of Balint training in order to enable physicians managing advanced cancer patients to provide psychosocial support, use more efficiently psychosocial support resources and be exposed to a smaller risk of burnout. CORRESPONDING AUTHOR: Dr. S. Popa-Velea, University of Medicine and Pharmacy Carol Davila Romania, [email protected] P312 USE OF NORMALIZATION PROCESS THEORY TO DESCRIBE INTEGRATION OF CARE MANAGEMENT INTO PRIMARY CARE PRACTICE J. Holtrop1, L.A. Green2, G. Potworowski3, L.A. Fitzpatrick4, A. Kowalk5 1 University of Colorado Denver, AURORA, United States of America 2 University of Alberta, EDMONTON, Canada 3 SUNY/University at Albany, ALBANY, United States of America 4 Michigan State University, EAST LANSING, United States of America 5 University of Michigan, ANN ARBOR, United States of America Introduction: Care management is a new complex intervention in the US within primary care. The purpose is to help patients with chronic disease(s) to better self-manage and reduce their risk of disease complications. We sought to understand how care management structure (centrally organized at an administrative level versus managed at the practice team level) affects how much care management gets utilized and if it becomes a routine practice. We used normalization process theory (NPT) to help understand elements of routinization. Methods: At 25 purposefully selected practices (out of 52 in 5 physician organizations or POs) we conducted interviews and observations; transcripts were placed into atlas.ti and qualitatively analyzed using the NPT. Coded segments with summary notes were analyzed by PO, within PO by practice, and by program structure (e.g., centralized versus practicebased). Extensive group review was used within the framework of the four NPT constructs, and the four components of the Collective Action construct were focused upon to examine structure within settings. Results: There was a greater degree of normalization where the care manager was embedded in the practice as part of the care team. An illustrative figure will be shown to demonstrate how the NPT elements of contextual integration, relational integration, and skill set workability facilitated relational integration, which in this case, meant routine use of care management. Normalization suffered when any one of these elements were absent or problematic. Conclusion: NPT was useful for illuminating how care management program structure affects utilization in practice. CORRESPONDING AUTHOR: Associate Prof PhD Holtrop, University of Colorado Denver, United States of America, [email protected] P313 QUALITY OF LIFE AND PUBLIC HEALTH PROGRAMS IN MARGINAL COMMUNITIES IN MEXICO CITY C.C. Renaud Universidad Nacional Autonoma de Mexico, MEXICO CITY, Mexico Introduction: In this paper we proposed to the Public Policy on Health designers to use the Quality of Life Index (QLI) as a requirement for the design and implementation of programs of health promotion and disease prevention, specifically in communities with a high index of marginalization. Methods: From a random sample of 77 inhabitants of both genders with: a) ages between 18 and 70; b) low level of education; c) average monthly

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

income of USD 290; d) 48% have 5+ members per household; e) all of them live in a poor community of Tlalpan in Mexico City, we applied an instrument (INCAVI, 2011) to measure the quality of life on a scale from 1 to 10 in five domains and 25 items: Health, Economy, Education, Security and Personal Life. The index results were: 6.85, 5.25, 5.71, 4.73 and 7.27 respectively, while the global index was 5.96. Results: Our findings suggest that both the community and the Government officials must collaborate in the design of public health programs and the enforcement of public policies in order to reconstruct the social fabric of these communities which have been significantly hurt by violence and social inequality. Public policy-making should focus on placing people in an environment that promotes satisfaction with their lives. CORRESPONDING AUTHOR: Ph. D C.C. Renaud Universidad Nacional Autonoma de Mexico Mexico [email protected] P314 A FOCUSGROUPSTUDY TO ASSESS HOW ASTHMA AND COPD PATIENTS FEEL ABOUT USING A PATIENT-WEB-PORTAL E.I. Metting University of Groningen, University Medical Center Groningen, GRONINGEN, The Netherlands Introduction: Patient-web-portals (PWP) have been developed by several healthcare providers to increase transparency and enhance selfmanagement. PWPs provide access to personal medical records and can include self-management tools or other applications. However, studies that have evaluated the effect of a PWP on self-management show mixed results. To increase the efficacy of PWPs, patients should be involved in the development process. Method: A focus group study among 36 asthma and COPD patients (52% men, mean age 64±11) to obtain more detailed information about their opinion, emotions and needs towards a PWP. Patients were divided in 3 groups according to their living area. Three focus group meetings were organized for each group and 4 main topics were addressed: 1)online access to medical information, 2)communication with caregiver, 3)applications and 4)self-management. Results: Patients have stressed the importance of having insight in the development of their disease over time. They think that a PWP can enhance insight and improve the communication with their physician and make them more equal partners. A PWP can also enhance communication between different caregivers. However, patients are afraid of fragmentation caused by different portals. Most patients were not interested in applications and self-management tools in the PWP except for making appointments online. Conclusion: Patients were very open and explained that they want to have access to a PWP and see their medical records if presented in an understandable way. However to make a PWP effective, primary and secondary care physicians should have access to the same portal. CORRESPONDING AUTHOR: E.I. Metting, University of Groningen, University Medical Center Groningen, The Netherlands, [email protected] P315 MOVING BEYOND EFFECTIVENESS: EXTENT TO WHICH BEHAVIROAL INTERVENTIONS REPORT OTHER TYPES OF EVIDENCE OF SUSTAINABILITY S. Ramondt, L.D. Cameron, S. Millhoff, P. Brown University of California, Merced, MERCED, United States of America Introduction: Ensuring that successful interventions aimed at changing health behaviors are sustained after the research concludes is a primary goal of behavioral medicine. In a previous study, we argued that sustainability maybe enhanced by greater engagement of organizations or individuals in the research project conclusion of the research and providing

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

other types of information, include cost and return on investment. This study reviews existing literature to identify the extent to which interventions aimed at changing health behaviors provide evidence on sustainability. Method: We reviewed all articles from 2010 to 2013 in 5 major behavioral medicine journals (JBM, BM, AMB, IJBM, and TBM) to identify whether studies reporting the results for a intervention aimed at changing health behavior report i) effectiveness measures, ii) costs, iii) funding sources, and iv) involvement of community organizations, individuals, or providers in the research. Results: Of the 180 articles that fit the criteria, 35% reported the cost per participant. Only 6% of the articles reported cost to the participants, organization or health care system. All studies reported effectiveness information, though types of information varied. Of those studies reporting funding, over 90% were from grants or temporary funding sources. Very few studies gave evidence that the program would be sustained by members of the community or participating organizations. Conclusion: Sustainability of interventions may be facilitated by behavioral medicine researchers expanding the type of information they collect and report as part of their studies. CORRESPONDING AUTHOR: S. Ramondt, University of California, Merced, United States of America [email protected] P316 EXPLORING CURRENT PAIN MANAGEMENT IN BURULI ULCER AND DESIRES FOR A TAILORED CLINICAL PRACTICE GUIDELINE J. de Zeeuw, M Alferink, T.S. van der Werf, M. Plantinga, A.V. Ranchor, Y. Stienstra University Medical Center Groningen, GRONINGEN, Nederland Introduction Buruli ulcer (BU) is a skin disease, highly prevalent in remote areas in West Africa. Wound care dressings play an important role in the treatment. Clinical observations showed pain during wound care among BU patients, however a clinical practice guideline is lacking. This study aimed to gain insight in factors influencing pain management in BU treatment settings, and to explore the need for a clinical practice guideline for treating pain in BU. Methods Data were collected in Ghana and Benin between November 2012 and August 2013. A mixed methods approach was used, including interviews and reviewing medical records. 11 semi-structured interviews were conducted among health personnel, regarding perceptions and beliefs on barriers and facilitating factors in pain management. 149 medical records were reviewed on pain medication prescription. Results Reported barriers included; difficulties in identifying mild pain; difficulties in pain assessment; fear of side effects; lack of availability of pain medication; shortage of time of personnel, suppression of pain expression by patients; financial constraints to pay for pain and patients’ expectations about pain management. Reported facilitating factors included the current practice of assessing pain by the personnel; the willingness to improve pain management and strategies used to help the patient to cope with pain. Conclusion Several barriers exist that should be intervened upon to enhance implementation of a clinical practice guideline. Facilitating factors, such as the willingness to improve treatment are likely to enhance the implementation of a clinical practice guideline for treating pain in BU. CORRESPONDING AUTHOR: Msc J. de Zeeuw, University Medical Center Groningen, Nederland, [email protected] P317 PARTICIPATORY ACTION RESEARCH (PAR) IN DEVELOPING A NATIONAL HEALTH INSURANCE (NHI) POLICY IN SOUTH AFRICA G. Setswe Human Sciences Research Council, PRETORIA, South Africa

S97

Introduction: PAR is a way of collecting information for organizing that honours, centres and reflects the experiences of people most directly affected by issues in a community. People use it to learn more about their material conditions. Community participation is an open and accountable process through which individuals and groups within a community can exchange views and influence decision-making. Methods: The PAR approach in this project used the Plan, Act, Observe and Reflect model. In the Plan phase, we conducted a desktop review to understand the health context and participation in health matters. In the Act phase, we conducted consultations to support communities to engage with the materials on the NHI and give input on its implementation. During the Observe and Reflect phases, we conducted an evaluation to show whether the project enhanced people’s participation in the NHI. Findings and Discussion: People are experts in their own experiences and control the gathering and use of information about their communities. They gather information to inform their actions for change. People reflect on the information they've gathered and the way in which they are gathering it throughout the process. They are active and not passive participants in the process and agree on principles and values that will guide their information gathering and stay accountable to them. Conclusion: PAR is useful when where there is little or no information available through government or academic research that reflects people’s experiences of access and quality to health care. CORRESPONDING AUTHOR: Prof. K.G. Setswe, Human Sciences Research Council, South Africa, [email protected] P318 COMMUNITY CONSULTATIONS IN THE DEVELOPMENT OF THE NATIONAL HEALTH INSURANCE (NHI) IN SOUTH AFRICA G. Setswe1, J. Witthuhn2 1 Human Sciences Research Council, PRETORIA, South Africa 2 Monash University, JOHANNESBURG, South Africa Introduction: The aims of the community consultations were to test a public policy engagement using the NHI, to conduct community consultations to guide development of the NHI, to enable communities to contribute to the development and implementation of the NHI, and to enable communities to hold government accountable for the implementation of the NHI Methods: Fifteen Focus Group Discussions (FGDs) were conducted with representatives from community-based organisations (CBO) and consultations were conducted at community sites and at workplaces. Findings and Discussion: The majority of participants had never heard about the NHI. The few that had heard about it had no idea about the content. People said they were generally interested to learn about the NHI. Some groups brought out their questions easily while others had to be probed. Questions were asked on membership and access to the NHI, financial implications for beneficiaries and for the public health sector and structure of the NHI and its leadership. Conclusions: Community members are interested to learn about issues that affect their daily lives with the NHI. It is essential to bring new information as close as possible to the people as it is generally not picked up in traditional ways. It is also essential to have good preparation for the staff before reaching out to the community to ensure that they feel confident. Participants hope for a serious change in health care with the NHI but have reservations in raising their hopes. CORRESPONDING AUTHOR: Prof. K.G. Setswe, Human Sciences Research Council, South Africa, [email protected] P319 AUTISM SPECTRUM DISORDER AND HEALTH ISSUES M. Bolduc, N.A. Abouzeid, N.P. Poirier University of Quebec in Montreal, MONTREAL, Canada

S98

Introduction: Autism Spectrum Disorder (ASD) is a neurodevelopmental disorder affecting nearly 1 % of the population. Individuals with ASD present distinct health particularities. Medical conditions and psychiatric disorders, as well as deficits in maintaining healthy lifestyles, may be associated with ASD. Objective: The aim of this study is to expose the variety of health issues experienced by individuals with ASD. Method: Participants: 52 parents Instrument: Self-administered questionnaire and semi-structured interviews Analysis: Quantitative (frequencies and means) and qualitative Results: Numerous health conditions such as gastrointestinal disorder, epilepsy and allergies were present in these individuals. Attention Deficit Disorder with or without Hyperactivity (ADHD) and Anxiety Disorder was also revealed by parents. Difficulties with toilet training, sleeping and eating are also common within the sample. More than half of these individuals are taking medication. Conclusion: It is clear that individuals with ASD are presented with other health issues that need to be addressed. Despite the considerable challenges, and the medication taken, parents still consider their child as being healthy. Research and interventions are required in this area. CORRESPONDING AUTHOR: M. Bolduc, University of Quebec in Montreal, Canada, [email protected] P320 KIDNEY TRANSPLANTATION AND PREGNANCY A.J Andrade1, R. Alves1, M.S.J Pais2, M. Ferreira3 1 University of Coimbra, COIMBRA, Portugal 2 Hospital of the University of Coimbra - Daniel de Matos Maternity, COIMBRA, Portugal 3 Polytechnic Institute of Viseu - Health School, VISEU, Portugal

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

P321 SOMATOSENSORYAMPLIFICATION AS A PREDICTOR OF MEDICATION SIDE EFFECTS F. Köteles, J. Szécsi, G.y. Bárdos. Eötvös Loránt University, BUDAPEST, Hungary Introduction: A proportion of medication side effects is of psychological origin. Although these nocebo effects are mainly evoked by contextual factors, psychological characteristics of patients may also play a role. In the current pilot study, it was hypothesized that somatosensory amplification (i.e., the tendency to experience a somaticsensation as intense, noxious, and disturbing) would predict the occurrence of side effects among hypertensive patients starting to take medication. Method: In a longitudinal study, 51 patients (64,7% female, mean age = 55.3±13.87 years) with a recent diagnosis of primary hypertension completed the somatosensory amplification scale (SSAS), started to take prescribed medication, and registered side effects on a daily basis for four weeks. Results: After controlling for age, gender, number of drugs taken, and previous personal and family-level experiences with medication side effects in the regression analyses, SSAS-score remained a significant predictor of reported side effects in the first three weeks of the study (β = 0.439, p < 0.01; 0.417, p < 0.05; 0.427, p < 0.05 for the three weeks, respectively). Conclusions: Patients with higher levels of somatosensory amplification are particularly prone to experience medication side effects. To reduce this tendency and to improve compliance, these patients may require special education on the nocebo phenomenon. CORRESPONDING AUTHOR: Dr. F. Köteles, Eötvös Loránt University, Hungary, [email protected] P322 EXAMINATION OF RELATION BETWEEN ACNE AND DEPRESSION IN JAPANESE UNDERGRADUATE STUDENTS H.M. Himachi Tokaigakuen University, NAGOYA, Japan

Background:Pregnancy is rare in women with Chronic Kidney Disease, especially in advanced stages, due to ovarian dysfunction, anovulatory vaginal bleeding and amenorrhea .However, after renal transplantation is possible to keep dreaming about family, but the increased risk of graft loss and the higher susceptibility of pregnancy to complications must be considered.Objective:To assess the risks and identify the variables that influencethe success of pregnancy in women after renal transplantation.Material and Methods:A retrospective, descriptive and analytical study with a sample of 37cases that occurred between1989 and 2010in a Portuguese maternity.Results:The mean age of women was 27.3years.In 32.4%of cases the cause of renal failure was unknown and in21.6% was motivated by Adult Polycystic Kidney Disease. 35.1% of women needed adjustments on the immunosuppressive therapy based on Cyclosporine, Azathioprine and Prednisolone.48.6% of the pregnancies were normal,13.5% had a urinary tract infection and27% developed hypertension.29.7% of pregnancies were full term and 54.1% of deliveries were by cesarean section. In2.7% of the cases a medical termination of pregnancy was proceeded due to transplant rejection,2.7% required dialysis during pregnancy, 8.1%suffered spontaneous abortion and5.4% resulted in stillbirth. Transplant rejection occurred after 12months of delivery in2.7% and also after 36months in2.7% of the cases.32.30% of the infants required neonatal intensive care.Conclusion:In our experience the retrospective analysis of pregnancies in women with renal transplantation shows that a successful pregnancy is possible after transplantation. Monitoring and controlling the most important risks contributes to pregnancies without maternal and fetal severe complications, including kidney graft rejection and the death of the fetus.

Introduction: In Europe and the United States, work is proceeding with some psychological factors and acne excoriee with acne patients (Gupta & Gupta, 1998). Acne is a common skin disorder in Japan. However, the research on acne and acne excoriee is insufficient in our country. The present study examined the characteristics of Japanese undergraduate students' acne and the relation between acne and depression. Method: The participants were 174 undergraduate students (mean age=20.18, SD=1.43). The state of present and past acne, and acne excoriee, were evaluated by using a checklist (Gupta et al., 1996). The participants also completed a Self-rating Depression Scale (SDS: Fukuda & Kobayashi, 1983). Results: 126 people had acne, and 112 people had acne scarring. In addition, 106 participants worried about acne, and 95 about acne scarring. The acne excoriee item was asked to the participants who had acne and acne scar. As a result, a total score of acne students' checklist was 17.65±9.77, and the acne scarred students' was 12.34±10.27. As for the depression score, the participants who had acne or acne scarring were significantly high (t(171)=2.27, p<.05, t(171)=2.91, p<.01). Also the participants who worry about acne and acne scarring were significantly high (t(170)=2.53, p<.05,t(171)=2.84, p<.01). Conclusion: Students with acne or acne scarring were significantly more prone to depression. In future, it will be necessary to examine the acne patients' depression in detail in Japan.

CORRESPONDING AUTHOR: Dr A.J. Andrade, University of Coimbra, Portugal, [email protected]

CORRESPONDING AUTHOR: H.M. Himachi, Tokaigakuen University, Japan, [email protected]

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

P323 CONTRIBUTIONS OF WHITE MATTER LESIONS TO COGNITIVE IMPAIRMENT IN PATIENTS WITH SUBCORTICAL ISCHEMIC VASCULAR DISEASE Y. Dong, L.V. Peiyuan, L.I. Ling Hebei General Hospital, SHIJIAZHUANG, China Objectives:To investigate the nature of white matter lesions(WML)and their correlation to the cognitive function in patients with SIVD. Methods: Extensive neuropsychological tests including MMSE,trail making test (TMT),clock drawing test (CDT), digit span test (DST) and verbal fluency test (VFT) were performed on 53 patients with SIVD and 25 normal elderly controls(NC). The patients were divided into vascular dementia (VaD) and vascular cognitive impairment no dementia (VCIND) group. Age-related white matter change rating scale was used to qualitative measure and locate the WML. Correlation analysis was used to examine the relationship between cognitive function and WML. Results: (1) The overall level of cognitive performance was significant inferior in VaD subjects. Compared with VCIND patients, VaD subjects showed decline on the memory, executive function, visuospatial and attention function (P<0.05). (2) The mean WML rating scores in VaD (12.73±3.35) and VCIND (10.44±3.23) were higher than that of NC(2.00 ±2.18), (P<0.05). The scores of frontal, parieto-occipital, temporal and basal ganglia area in VaD patients increased remarkedly (P<0.05). Patients with VCIND had increased scores in frontal and parieto-occipital area (p<0.05).(3)The score of total WML was negatively correlated with MMSE, CDT, DST and VFT scores significantly (P<0.01). Positive correlations were observed between WML and word recall, recognition, TMT respectively (P<0.05). Conclusions: (1) SIVD is related to comprehensive cognitive impairment, specifically executive, visuospatial and attention function. (2) WML located at frontal and parieto-occipital area was serious. The cognitive function impairment in SIVD could be associated with the degree of WML. CORRESPONDING AUTHOR: Y. Dong, Hebei General Hospital, China, [email protected] P324 INTERLEUKIN-6 AND HIGH SENSITIVITY C-REACTIVE PROTEIN IN PATIENTS WITH SUBCORTICAL ISCHEMIC VASCULAR DISEASE Y. Dong, L.I. Ling, L.V. Peiyuan Hebei General Hospital, SHIJIAZHUANG, China Objective:To evaluate the levels of interleukin-6 (IL-6)and high sensitivity c-reactive protein (hsCRP) in the serum of patients with subcortical ischemic vascular disease (SIVD) and the correlation with cognition and white matter lesions (WML). Methods:19 patients with vascular dementia (VaD), 20 patients with cognitive impairment no dementia (VCIND) and 15 gender and age matched normal controls were recruited. All subjects were underwent clinical examination, neuropsychological assessment and conventional MRI. The serum levels of IL-6, hsCRP were detected by enzyme linked immunosorbent assay and fixedtime nephelometry methods in all subjects, respectively. In addition, the relationship was described among these inflammatory factors and WML, cognitive impairment in patients with SIVD. Results:(1) Compared with controls, patients with SIVD had elevated IL6 serum levels, 97.73±26.56 pg/ml in VaD patients and 83.77±21.29 pg/ ml in VCIND (P<0.05). There was no statistical difference in the VaD and VCIND patients (P>0.05). (2) The serum hsCRP expression was increased in VaD patients (4.75±3.52 mg/L) compared with the other groups(P<0.05). (3) There was no statistical correlation between the serum levels of IL-6, hsCRP and MMSE, Montreal cognitive Assessment (MoCA), WML scores. (4) Slight positive correlation between IL-6, hsCRP was observed(r=0.332,P=0.021).

S99

Conclusions:(1) Increased serum levels of IL-6, hsCRP in patients with SIVD might indicate that inflammatory processes may be activated in SIVD. (2) Enhanced serum levels of IL-6, hsCRP could not imply the association with the severity of cognitive impairment and WML in SIVD patients. CORRESPONDING AUTHOR: Y. Dong, Hebei General Hospital, China, [email protected] P325 PATIENTS WITH ADRENAL INSUFFICIENCY HATE THEIR MEDICATION: THE RELATION BETWEEN MEDICATION BELIEFS AND ILLNESS PERCEPTIONS J. Tiemensma1, C.D. Andela2, A.M. Pereira2, J.A. Romijn2, N.R. Biermasz2, A.A. Kaptein2 1 University of California, Merced, MERCED, United States of America 2 Leiden University Medical Center, LEIDEN, Nederland Introduction: Patients with adrenal insufficiency (AI) require a daily and lifelong intake of hydrocortisone (HC). Previous studies demonstrated that medication beliefs are associated with illness perceptions. The spectrum of medication beliefs in patients with AI is currently unknown. Method: Cross-sectional evaluation of illness perceptions and medication beliefs in 107 patients with primary (n=49) or secondary AI following Cushing’s syndrome (n=29) or non-functioning pituitary adenoma (n=29). The Illness Perception Questionnaire-Revised and the Beliefs about MedicinesQuestionnaire were used. Results: Stronger beliefs about the necessity of HC and stronger concerns about the adverse effects of HC were related to attributing more symptoms to AI, to perceiving AI being more cyclical, to perceiving more negative consequences of AI, and to having stronger emotional representations (all P<0.05). Furthermore, stronger beliefs about the necessity of hydrocortisone were associated with feelings of less personal control over AI (P<0.05). Stronger concerns about the adverse effects of hydrocortisone were associated with lower perceived treatment control and lower illness coherence (both P<0.05). In addition, patients with Cushing’s syndrome reported stronger beliefs regarding the necessity of taking hydrocortisone, compared with patients with Addison’s disease (P=0.039) or NFA (P<0.001). Conclusion: Specific beliefs about the necessity of hydrocortisone and concerns about its adverse effects were strongly associated with more negative illness perceptions. Patients with adrenal insufficiency due to different etiologies demonstrate differences in beliefs about the necessity of HC intake. These results need to be taken into account when developing psychosocial education / self-management programs aiming at improving QoL. CORRESPONDING AUTHOR: J. Tiemensma, University of California, Merced, United States of America, [email protected] P326 SHORT-TERM EFFICACY OF ACUPUNCTURE IN TREATING URINARY INCONTINENCE IN WOMEN: A RANDOMIZED CONTROLLED TRIAL S.J. Engberg, S.M. Sereika, S. Cohen University of Pittsburgh, PITTSBURGH, United States of America Introduction: This study examined the short-term efficacy of acupuncture in treating urinary incontinence (UI) in women. Methods: This was a double-blind, randomized, controlled trial. The convenience sample of women (N=127) was age >25 years with urge and/or stress UI > 2 times/week for >3 months. Subjects were randomization to a true (TA) or sham (SA) acupuncture group. The 6-week intervention protocol for both groups was identical except for the use of a blunted telescopic needle in the SA group. UI severity (measured using e-diaries) and health-related quality of life (HRQOL) (measured using SF-36) were assessed at baseline and 1- and 4-weeks post-intervention. An intention-to-treat approach was used to analyze outcomes.

S100

Results: There were no significant group (TA: n=64; SA: n=63) differences at baseline (socio-demographics, depressive or anxiety symptomatology, body mass index, HRQOL and UI characteristics). The median UI duration = 6.0 years and the mean UI episodes = 2.4/ day (SD=1.9). Most subjects (88.2%) had both urge and stress UI. TA subjects had a significantly greater median reduction in UI (41.7%) at 1week post-intervention than SA subjects (22.8%, p=.008). At 4-weeks post-intervention, the differences were not significant (TA median=49.4; SA median=49.5, p=.38). There were significant group differences in SF36 Mental Component change scores at 1-week (p=.03), but not 4-weeks post-intervention. There were no significant differences in SF-36 Physical Component scores. No serious adverse events were reported. Conclusions: While some patients may benefit from acupuncture, additional research is needed before acupuncture can be recommended as a UI treatment. CORRESPONDING AUTHOR: Professor J. Engberg, University of Pittsburgh, United States of America, [email protected] P327 ANXIETY AND TIME PERCEPTION IN ANAESTHETIZED PATINETS W. Plotek1, M. Sobol-Kwapinska2, M. Cybulski3, A. Kluzik4, M. Grzeskowiak4, L. Drobnik4 1 Poznan University of Medical Sciences, Department of Teaching Anaesthesiology, POZNAN, Poland 2 The John Paul II Catholic University of Lublin, Department of Personality Psycho, LUBLIN, Poland 3 Poznan University of Medical Sciences, Department of Clinical Psychology, POZNAN, Poland 4 Poznan University of Medical Sciences, Department of Anaesthesiology, Intensive, POZNAN, Poland Introduction: The purpose of the study was to evaluate the influence of anxiety on the perception of 1-minute intervals produced before and after short general anaesthesia for colonoscopy. Methods: Depression and serious cognitive impairment were screened and excluded with Sense of Coherence Meaningfulness Subscale, and Mini-Mental State Examination. The patients were premedicated with the calculated dose of oral midazolam (0.1-0.15 mg/kg), but the effect of this dose was various on different participants. The preoperative anxiety was described as the need of additional doses of intravenous 1-2 mg iv midazolam in order to achieve similar levels of preoperative sedation (Ramsay 3 pts). The production (TP) of 1-minute time intervals were tested in three trials during the following sessions: initial, 1.5, 3, and 6 hours after anesthesia (AG, 16 women, 8 men) and were compared to the non-anaesthetized patients (CG, 19 women, 6 men). Results: There were significant differences in the time misestimations between the group of patients who had received midazolam immediately prior to colonoscopy (N=11) and the group of patients who had not received this substance (N=13). The patients who had been administered midazolam were less accurate in generating the 1-minute intervals in comparison with the patients who had not received it. Conclusion: Anxiety and administration of additional doses of midazolam exert a strong influence on the inner clock. CORRESPONDING AUTHOR: W. Plotek, Poznan University of Medical Sciences, Department of Teaching Anaesthesiology, Poland, [email protected] P328 EFFECTIVENESS OF RELAPSE PREVENTION THERAPY FOR PREVENTING RELAPSES IN URINARY INCONTINENCE IN OLDER COMMUNITY-DWELLING ADULTS S.M. Sereika, Sandra J. Engberg University of Pittsburgh, PITTSBURGH, United States of America

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

Introduction: The effectiveness of behavioral therapies for urinary incontinence (UI) among cognitively intact older adults has been demonstrated; however, relapse rates are high (23% to 95%), largely due to patient nonadherence. In a randomized controlled trial we tested the effectiveness of a relapse prevention intervention (RBT) in enhancing adherence and sustaining continence levels following therapy in older (>60 years) adults. Methods: Subjects (N=279) were randomized to either behavioral therapy intervention only (SBT) (biofeedback taught pelvic floor muscle exercises (PFME) with stress and/or urge strategies) (n=138) or SBT followed by RBT (gradual intervention withdrawal combined with self-efficacy enhancing and relapse prevention strategies) (n=141). Follow-up was quarterly for one year following assigned treatment. Diaries assessed continence levels and adherence to strategies and PFME. Relapse was defined as a ≥10% increase in incontinent episodes relative to levels at treatment conclusion. Results: Subjects were mostly women (82.5%), white (96.1%), and aged on average 76.4+8.1 years. At entry, subjects reported on average 2.7+2.5 incontinent episodes/day and being incontinent on average 7.4+9.2 years. During follow-up 64.2% of SBT subjects relapsed, while 58.8% of RBT subjects relapsed (p=.245). A greater proportion of RBT (0.52) than SBT (0.41) subjects were adherent to urge strategies (p=.023). Similarly, a greater proportion of RBT (0.36) than SBT (0.26) subjects were adherent to stress strategies (p=.037). The proportion of prescribed PFM exercises completed declined similarly over time for both groups (p=.005). Conclusion: In this sample, the RBT intervention was no more effective than SBT in maintaining post-treatment continence levels. CORRESPONDING AUTHOR: Professor M. Sereika, University of Pittsburgh, United States of America, [email protected] P329 RECEIVING A KIDNEY FROM A LIVING DONOR: AN INTERPRETATIVE PHENOMENOLOGICAL ANALYSIS J. Spiers, J. Smith Birkbeck University of London, LONDON, United Kingdom Much previous qualitative health research into living kidney donation has focused on the decision making and experiences of the donor, rather than the recipient (eg Brown, 2008; Challenor, 2013). This is despite evidence suggesting that this is a more psychologically challenging time for recipients due to issues around reciprocation and guilt (Gill and Lowes, 2008; de Groot, 2012). More longitudinal, qualitative work in this area has been called for (eg Decker, 2008). The current longitudinal study explored the experiences of four recipients of kidneys from living donors. Four participants receiving kidneys from different family members (sister, husband, son-in-law and daughter) were interviewed at three time points before and after transplant about how they experienced and coped with their illness. Transcripts were analysed using Interpretative Phenomenological Analysis (IPA, Smith, Flowers and Larkin, 2009), a qualitative approach ideally suited to health psychology (Brocki and Wearden, 2005) due to its an idiographic stance, which explicates ‘the patient’s view’ (Armstrong, 1984). Three themes arose from the data. This presentation will communicate material from the first theme, ‘the relational journey of receiving a kidney’. Donor/recipient relationships developed in different ways, each presenting unique challenges. As predicted, this presentation will show that the process of receiving a kidney from a living donor is a complex one that can lead to feelings of guilt and ambiguity for the recipient, but also a strengthening of the donor/recipient relationship. It is hoped that these findings will enable healthcare professionals to support recipients more fully in their illness journeys. CORRESPONDING AUTHOR: Miss J.S. Spiers, Birkbeck University of London, United Kingdom, [email protected]

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

P330 DO INDIVIDUALS WITH AND WITHOUT DEPRESSION VALUE DEPRESSION DIFFERENTLY? AND IF SO, WHY? K. Papageorgiou1, K.M. Vermeulen1, M.J. Schroevers1, A.M. Stiggelbout2, E. Buskens1, P.F.M. Krabbe1, E. R. van den Heuvel1, A.V. Ranchor1 1 University Medical Center Groningen, GRONINGEN, Nederland 2 Leiden University Medical Center, LEIDEN, Nederland Health state valuations are used to assess the effectiveness of health care interventions. Whose valuations should be used is currently debated. The general population seems to value somatic conditions more negatively than patients. Little is known about valuations of psychological conditions. We studied whether individuals with and without depression differ in their valuations of depression; and if so, whether perceptions concerning depression (e.g. stigma) and individual characteristics (e.g. mastery) determine such discrepancies. Methods.Participants performed an online-administered Time-Trade-Off task to value 30 vignettes (four each) describing depression states of different severity (mild, moderate or severe). The PHQ-9 was used to assign participants to “with”and “without depression”groups. A generalized linear mixed model was used to answer the research questions. Results. The sample (N=1268) was representative of the Dutch population on age, gender and residence. We found that individuals with depression (N=200) valued depression more negatively than did individual without depression (n=1068), but only for mild depression (p=.007). The depression groups differed on their perceptions of depression and individual characteristics. Yet, for none of the related variables their association with valuations was different for the two groups. Conclusions. Since the general population values depression less negatively, using their perspective might result in less favorable treatment effects for mild depression interventions. This is the opposite of what previously reported for somatic conditions. Biases related to how individuals with and without depression perceive depression or to individual characteristics did not seem to explain the observed discrepancies in their valuations of depression. CORRESPONDING AUTHOR: K. Papageorgiou, University Medical Center Groningen, Nederland, [email protected] P331 PSYCHOLOGICAL HEALTH PROBLEMS AFTER LIVER TRAN SPLANTATION AT DIFFERENT TIME FRAMES J.H. Annema, P.F. Roodbol, R.J. Porte, A.V. Ranchor University Medical Center Groningen, GRONINGEN, Nederland Introduction: After liver transplantation psychological health problems are common, and psychological functioning does not restore to the level in the general population. However, most studies report on psychological health problems in the first years after transplantation, less is known about the prevalence in the long run. In this study prevalence rates of symptoms of anxiety, depression and posttraumatic stress (PTSS) were examined at different time frames after transplantation, and compared to prevalence rates in the general population. Method: Cross-sectional survey among 373 liver transplant recipients transplanted between 1979 and 2009. Based on time since transplantation clinical relevant time frames were identified: 0.5-2 years, 2-5 years, 5-15 years, and > 15 years after transplantation. Results: Response arte was 75%, Overall, 33% experienced clinical relevant symptom levels of anxiety (29%), depression (17%), or PTSS (10%), either one or in combination. However, prevalence rates differed significantly between time frames. All psychological health problems were most prevalent at 0.5-2 years after transplantation. Symptoms of anxiety and depression decreased after two years but increased >15 years after transplantation. PTSS were most prevalent in the first five years after transplantation. Compared to the general population symptoms of anxiety

S101

were more prevalent at all times, depressive symptoms were more prevalent shortly after and in the long run, and PTSS were more prevalent in the first 5 years. Conclusion: Not only shortly after, but also in the long run a significant subset of transplant recipients encounter psychological health problems. Therefore routine screening of psychological health problems remains important. CORRESPONDING AUTHOR: J.H. Annema, University Medical Center Groningen, Nederland, [email protected] P332 CLUSTER ANALYSIS: A RELEVANT STATISTICAL TOOL TO IDENTIFY SOCIAL VULNERABILITY TRAJECTORIES IN A YOUNG PARENT'S J.L.F. Lachance Fiola1, S. Gendron2, G. Dupuis1, J. Moreau2 1 University of Quebec in Montreal, MONTREAL, Canada 2 University of Montreal, MONTREAL, Canada Objective: To establish the relevance of cluster analysis as a statistical tool to identify different social vulnerability trajectories among pregnant teens participating in a young parent’s program evaluation. Method: 451 pregnant teens aged 20 years or less were recruited from 19 community health and social service centers in Quebec, Canada. Theoretical and statistical criteria were applied to select 22 categorical and continuous variables that were submitted to a TwoStep cluster analysis. Results: Three distinct social vulnerability trajectories were identified (R1, R2, R3). R1 predominantly includes young mothers who are pregnant for the first time and who live with their partner in a household reporting employment as the main source of income. Mothers of R2 are younger than the others, on average, and all live with their parents, or those of their partner, during their pregnancy. Young women from R3 tend to report a less favorable perception of their health and higher levels of psychological distress and childhood trauma than other study participants. The majority report a previous pregnancy and nearly nine out of 10 have dropped out of school. Conclusion: Careful variable selection resulted in the identification of different social vulnerability trajectories in a cohort of pregnant teens participating in a young parent’s program evaluation, while ensuring the desired statistical power for subsequent analyses. Given the different realities and needs of these families, the derived clusters are essential indicators to discriminate the effects of an early intervention program. CORRESPONDING AUTHOR: Dr J.L.F. Lachance Fiola, University of Quebec in Montreal, Canada, [email protected] P333 RELIABILITY AND VALIDITY OF THE DUTCH IPAQ IN PATIENTS AFTER TOTAL HIP AND KNEE ARTHROPLASTY M. Stevens, T. Blikman, S.K. Bulstra, I. Van denAkker-Scheek, I.H.F. Reininga University Medical Center Groningen, GRONINGEN, Nederland Introduction. Despite recognized benefits of regular physical activity, little is known about the physical activity level of patients after THA/ TKA. None of the currently used questionnaires is internationally accepted. The International Physical Activity Questionnaire (IPAQ) tries to address this problem, but its validity and reliability among a THA/TKA population is unknown. Methods. 44 patients completed the IPAQ (short and long form) twice. Test-retest reliability was assessed by Spearman’s correlation coefficients (r) and Intraclass coefficients (ICC). Additionally Standard Error of Measurement (SEM) and Minimally Detectable Change (MDC) were calculated. Concurrent validity was determined by an accelerometer. Spearman’s correlation coefficients were calculated between IPAQ scores and accelerometer data. Bland & Altman analyses were performed for both reliability and validity. Results. Fair-to-good correlation coefficients were found for test-retest reliability of the total and activity scores (r:0.49 - 0.81, ICC:0.27 - 0.71).

S102

SEM and MDC were large. For relative validity, weak to moderate correlation coefficients were found for total and activity scores (r: -0.07 0.54). Systematic bias was found between the IPAQ and accelerometer data, with higher scores on the IPAQ. Conclusion. Overall, the IPAQ showed fair reliability and weak relative validity. These results are in line with previous studies into the reliability and validity of the IPAQ. Due to systematic bias and large SEM and MDC the IPAQ seems only suitable for inter-group comparisons. CORRESPONDING AUTHOR: Dr M. Stevens, University Medical Center Groningen, Nederland, [email protected] P334 THE DEVELOPMENT OF A MEASURE FOR ASEXUALITY: THE ASEXUALITY QUESTIONNAIRE N.K.R. Robbins1, Kathryn Low2 1 Yale University, NEW HAVEN, United States of America 2 Bates College, LEWISTON, United States of America Introduction: Asexuality is commonly defined as a lack of sexual attraction. However, a number of conceptual and methodological obstacles have prevented researchers from drawing conclusions regarding the definition, prevalence, and correlates of asexuality. Additionally, the specific dimensions of asexuality remain unclear as no researcher has published a well-validated objective measure of the construct. The current study aims to: 1) develop a psychometrically sound measure of asexuality; and 2) explore the relationship between asexuality, sexual aversion, and sexual desire. Method: Participants were recruited from an undergraduate institution (n=166) and from three online asexual communities (n=248). Data were collected using a multi-dimensional asexuality measure that drew upon existing scales of sexuality and the available data on asexuality. Results: Based on factor analyses, sensitivity, specificity, and the psychometric properties of the instrument, the Asexuality Questionnaire (AQ) could reliably distinguish between asexual and non-asexual individuals. Correlational analyses revealed a large significant, negative relationship between total score on the Sexual Desire Inventory-2 and the AQ, r (323) = -.93, p< .001, and a positive, but non-significant relationship between scores on the Sexual Aversion Scale and the AQ, r(276) = .092, p > .05. Conclusion: Asexuality appeared to be a multidimensional construct related to sexual cognition, romantic attraction, and sexual desire. Statistical analyses determined that the developed asexuality measure had high internal consistency and acceptable specificity and sensitivity. Results suggest that sexual aversion is not related to the construct of asexuality and that participants who experience more sexual desire are less likely to be asexual. CORRESPONDING AUTHOR: N.K.R. Robbins, Yale University, United States of America, [email protected] P335 MEASURING HEALTH OUTCOMES WITH VALID WHO-ICF CONTENT IN DIVERSE POPULATIONS ILLUSTRATION OF METHODS WITH SF-36 M. Johnston1, B. Pollard1, D. Dixon2 1 University of Aberdeen, ABERDEEN, United Kingdom 2 Strathclyde University, GLASGOW, United Kingdom Introduction. Health outcome measures vary in content and may not be equally valid for all populations. The WHO-ICF provides a theoretical framework for classifying content as Impairment (I), Activity Limitations (AL) and/or Participation Restrictions (PR). Recent developments in the quantitative assessment of Discriminant Content Validity (DCV) and Differential Item Functioning (DIF) allow assessment of content of existing measures against the ICF and identification of items invalid for specific populations. This paper illustrates the application of DCV and DIF to the SF-36.

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

Method. DCV: 10 judges rated each SF-36 item to indicate confidence that the item measured only I, AL or PR or a mixture. Single sample ttests assessed which construct(s) each item measured. DIF: Data from 763 community residents with osteoarthritis were analysed using regression DIF methods to examine whether items loaded equally on full scores for groups differing on age, gender and social deprivation. Results. DCV showed that 16 items were significantly pure I, 7 pure AL and 2 PR; others were mixed. DIF was found for 6 items for age, 2 for gender and 1 for social deprivation (adjusting for age). Conclusions. It is important that health outcome measures are valid for the constructs being measured and for the population assessed. DCV and DIF methods are valuable in assessing the validity of measures. The results suggest that the SF-36 a) cannot be used to measure ICF constructs without selection of valid items, and b) has items which may bias results for some populations. CORRESPONDING AUTHOR: Professor M. Johnston, University of Aberdeen, United Kingdom, [email protected] P336 ILLNESS AMONG SCHOOL-CHILDREN: COMPARISON OF SMS DATA FROM PARENTS AND QUESTINAIRE DATA FROM CHILDREN A.M. Denbaek, A. Johansen, C.T. Bonnesen, A. Andersen, B.E. Holstein, B. Laursen, P. Due National Institute of Public Health, University of Southern Denmark, COPENHAGEN K, Denmark Introduction We collected data about illness among 6-15-year olds by means of SMS communication with parents and questionnaires answered by the children. The aim was to examine 1) participation and adherence in a 22-week SMS-study, 2) potential selection bias due to low parental participation, and 3) differences in illness reported by children and parents. Methods We invited parents to a 22-week SMS-study through their child’s school. Each Sunday, parents received a text message asking how many days the child had been home due to illness past week. The children answered a questionnaire about socio-demographic background, health and absence from school due to illness past week. We used questionnaire data to examine what characterized children whose parent signed up for the SMS-study. Results 3015 parents (35.7%) signed up for the SMS-study of which 90% participated at least 18 weeks. The participation rate among the children was 89.2%, n=7525. Children in 5th-8th grade whose parents signed up for the SMS-study were more likely to live with both parents (p=0,0108), be ethnic Danish (p<0.0001) and have parents from higher socioeconomic strata (p=0.0012). Parents reported on average 9% (0th-4th-grade) and 10% (5th-8th-grade) illness-related absenteeism per week. Eighteen percent of 0th-4th-graders and 13% of 5th-8th-graders, with parental participation in the SMS-study, reported illness-related absenteeism within the past week. Conclusion It is difficult to get parents to participate in a 22-week SMSstudy, but after signing up adherence is good. Children of participating parents differed from the background population. School children reported more illness-related absenteeism than parents. CORRESPONDING AUTHOR: A.M. Denbaek, National Institute of Public Health, University of Southern Denmark, Denmark [email protected] P337 RELIABILITY AND VALIDITY OF A QUESTIONNAIRE ASSESSING FEAR-AVOIDANCE IN PHYSIOHERAPISTS TREATING PATIENTS WITH MUSCULOSKELETAL PAIN I. Demmelmaier1, K. Bivstedt2 1 Uppsala University, UPPSALA, Sweden 2 Västerorts Physiotherapy, STOCKHOLM, Sweden

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

Introduction: Health care providers’ own pain-related fear-avoidance (FA) beliefs and clinical behaviors may influence patients’ understanding of their musculoskeletal pain. Brief and feasible instruments are needed to assess such beliefs and behaviors. This study aimed to investigate the reliability and validity of the Fear Avoidance Scale - Physioherapist (FAS-PT). Method: A total of 37 physiotherapists in primary care in the south of Stockholm responded twice to a questionnaire, consisting of 10 items on FA beliefs and 7 items on self-reported clinical behavior. Test-retest reliability over 14 days was calculated for the total questionnaire. Overall internal consistency and item-to-total correlations were estimated for the items on FA beliefs. Concurrent validity was assessed by analyzing correlations between FA beliefs and each self-reported clinical behavior. Results: The analyses demonstrated good test-retest reliability with ICC = 0.88 (95% CI 0.78 - 0.94). Overall internal consistency was questionable with Cronbach's alpha = 0.68, while item-to-total correlations ranged between 0.19 and 0.60. Concurrent validity analyses resulted in negative correlations between FA attitudes and a) perceived ability to predict chronicity (Pearson’s r = -0.45, p = 0.006) and b) giving advice to work in spite of pain (Pearsons r = -0.56, p = 0.001). Conclusions: Further development of the FAS-PT is needed. In-congruent items should be removed and FA beliefs may be validated against direct observations of clinical behavior. CORRESPONDING AUTHOR: PhD I. Demmelmaier Uppsala University, Sweden, [email protected] P338 VALIDITY OF (ULTRA-)SHORT HEART RATE VARIABILITY RECORDINGS L. Muñoz-Venegas1, A. Van Roon1, H. Riese1, E. Oostenbroek1, I. Westrik1, E.J.C. De Geus2, R. Gansevoort1, J. Lefrandt1, I.M. Nolte1, H. Snieder1 1 University Medical Center Groningen, GRONINGEN, Nederland 2 VU University Amsterdam and Institute for Health and Care Research (EMGO+), AMSTERDAM, Nederland Introduction: In order to investigate the applicability of routine 10s electrocardiogram (ECG) recordings for time-domain heart rate variability (HRV) calculation we explored to what extent these (ultra-)short recordings capture the “true”HRV. Methods: The standard deviation of normal-to-normal intervals (SDNN) and the root mean square of successive differences (RMSSD) were measured in 3,428 adults. SDNN and RMSSD were assessed from (ultra)short recordings of 10s(3x), 30s, and 120s and compared to 240s300s (gold standard) measurements. Pearson’s correlation coefficients (r), Bland-Altman 95% limits of agreement (LoA) and Cohen’s d statistics were used as agreement analysis techniques. Results: Agreement between the separate 10s recordings and the 240s300s recording was already substantial (r=0.757-0.764/Mean bias=0.3990.415/d=0.737-0.776 for SDNN; r=0.853-0.862/Mean bias=0.080-0.096/ d=0.140-0.170 for RMSSD), and improved further when three 10s periods were averaged (r=0.868/Mean bias=0.366/d=0.730 for SDNN; r=0.944/Mean bias=0.058/d=0.108 for RMSSD). Agreement increased with recording length and reached almost perfect agreement at 120s (r=0.956/Mean bias=0.064/d=0.134 for SDNN; r=0.986/Mean bias=0.014/d=0.026 for RMSSD). For all recording lengths and agreement measures, RMSSD outperformed SDNN. Conclusion: Our results confirm that it is unnecessary to use recordings longer than 120s to obtain accurate measures of RMSSD and SDNN in the time domain. Even a single 10s (standard ECG) recording yields a valid RMSSD measurement, although an average over multiple 10s ECGs is preferable. For SDNN we would recommend either 30s or multiple 10s ECGs. CORRESPONDING AUTHOR: L. Muñoz-Venegas, University Medical Center Groningen, Nederland, [email protected]

S103

P339 ADAPTATION OF A DISTRESS SCREENING TOOL TO LUNG CANCER FAMILY CAREGIVERS S. Simard1, M. Aubin2, L. Vezina1, B. Fournier3, L. Tremblay4, R. Verreault5, S. Dumont6, J-F. Desbiens7, L. Fillion8 1 Laval University, QUÉBEC (QC), Canada 2 Family Medicine, Laval University, QUEBEC, Canada 3 Institut universitaire de cardiologie et de pneumologie de Québec, QUEBEC, Canada 4 Medecine, Laval University, QUÉBEC (QC), Canada 5 Social and Preventive Medicine, Laval University, QUEBEC, Canada 6 Social Service, Laval University, QUEBEC, Canada 7 Nursing, Laval Universtiy, QUEBEC, Canada 8 Nursing, Laval University, QUEBEC, Canada Introduction: Lung cancer is associated with higher distress in both patients and their family caregivers (FC). Recently, several Canadian cancer centres have implemented systematic distress screening programs during the cancer trajectory, but they target only patients. Objective: To adapt and preliminary validate for FCs the distress screening tool (DSTFC). Methods: The DST-FC includes the Distress Thermometer, an adaptation of the Canadian Problem Checklist and the Edmonton Symptom Assessment System. Twenty nine FCs of patients at different phases of their lung cancer were recruited at the ambulatory oncology clinic and interviewed to assess their perception of this tool relevance and acceptability. The Hospital Anxiety and Depression Scale (HADS) was used to assess the convergent validity. Results: Qualitative FC acceptance of the DST-FC is good. The mean DST score was 3.9 (range 0-10) and 41% of FCs expressed a clinical level of distress (score ≥ 5). The mean number of problem identified was 7.1 (range 0-21) and was strongly related to distress (r = 0.64). Worries, sadness, coping with disease, understanding illness/treatment, fatigue, sleep and daily tasks were the most frequent problems reported. Using the cut-off score of HADS, patients identified with a clinical distress reported significant higher DTS-FC score (5.3 vs. 2.7; p=0.001) than those with no distress. Conclusion: DST-FC developed in this study has a high acceptance, provides reliable and representative data to identify distress in FCs and it appears to be a valid and useful short screening-tool for FCs. CORRESPONDING AUTHOR: Dr. S. Simard, Laval University, Canada, [email protected] P340 INTERNAL RELATIONS OF COMMON PSYCHOMETRIC MEASURES IN BEHAVIORAL MEDICINE - A METHODOLOGICAL STUDY S. Suominen 1 , L. Sillanmäki 2 , M. Tuloisela-Rutanen 3 , M.J. Koskenvuo4 1 Nordic School of Public Health, GOTHENBURG, Sweden 2 University of Turku, TURKU, Finland 3 The Lahti Deacony Foundation, LAHTI, Finland 4 University of Helsinki, HELSINKI, Finland In empirical research related to behavioral medicine various psychometric measures are used. Many can on good grounds be assumed to describe closely related, opposite or even over-lapping phenomena. By now, their internal relations have not been sufficiently studied. Health and Social Support (HeSSup) is an on-going prospective mail survey initiated in 1998 with follow-up surveys in 2003 and 2012. The original population sample of the study comprised of 64,797 Finns representative of the age groups 20-24, 30-34, 40-44 and 50-54 years of the concurrent population. The first survey yielded 25,898 acceptable responses (response rate 40 %). The respondents of this study completed in 2003 instruments measuring sense of coherence (Antonovsky’s 13 item scale), depression (Beck’s 31 item inventory), life satisfaction (Allardt’s 4 item scale), hostility

S104

(Koskenvuo’s 3 item scale), cynical distrust (8 items derived from the Cook - Medleyn scale), optimism and pessimism (Scheier’s and Carver’s 10 item scale) and social support (Sarason’s 6 item scale). The analyses were restricted to individuals who participated both at baseline and 2003. All observations with missing values had to be excluded. A maximum likelihood factor analysis with Promax rotation was carried out for all the items separately for women (N=10,254) and men (N=6,432). The statistical analysis was carried out with SAS 9.3 software. Eight factors with eigenvalues > 1 were retained for both genders. The factor structure was similar for men and women with a fairly good correspondence to the original measures. The results support the validity and independent domain of each measure. CORRESPONDING AUTHOR: Professor S.S. Suominen, Nordic School of Public Health, Sweden, [email protected] P341 EVALUATION OF THE GERMAN HEADACHE MANAGEMENT SELF-EFFICACY SCALE J. E. G. Graef1, W. Rief1, Y. Nestoriuc2 1 Philipps-University of Marburg, MARBURG, Germany 2 University of Hamburg, HAMBURG, Germany Introduction: We translated the Headache Management Self-efficacy Scale (HMSE, French et al., 2000) into German, and investigated its psychometric criteria. Further, we investigated headache-specific selfefficacy as a predictor of pain disability in German headache sufferers. Methods: HMSE was translated forward and backward using guidelines for translation of instruments in foreign languages by Schmitt and Eid (2007). Validity and reliability were investigated cross-sectionally (N = 304). Relations between headache-specific self-efficacy and general selfefficacy (GSE), coping (PCQ) and anxiety/depression (PHQ) were calculated for construct validity. The impact of headache-specific selfefficacy on pain disability was calculated by hierarchical multiple regression analysis. Responsiveness and stability were assessed in a longitudinal sample (N = 32). Different primary headaches proved ecological validity of the instrument. Results: The Headache Management Scale-German (HMSE-G) is a valid, reliable (Cronbach's alpha = .89), internally responsive (SRM =.33) and stable (rtt = .46) assessment tool. Patients with high headache-specific self-efficacy were less impaired by their headache. The predictors age, headache intensity and headache-specific self-efficacy each explained incremental variance on pain disability (R2 = .28, F (11,234) = 8.24, p < .001). Headache-specific self-efficacy explained the second most variance in this model (betaage = .31, betaself-efficacy=-.22, betaintensity = .20). Conclusion: The HMSE-G assesses headache-specific self-efficacy in a reliable, valid and stable way. The unique predictive value of headachespecific self-efficacy with regard to pain-related disability was demonstrated. Hence, the questionnaire can be recommended as an evaluation tool for behavioral headache treatments. CORRESPONDING AUTHOR: J. E. G. Graef, Philipps-University of Marburg, Germany, [email protected] P342 DISTRESS IN PREMENSTRUAL SYNDROME (PMS): VALIDATION OF A NEW PMS-IMPAIRMENT MEASURE J. N. Kues, C. Janda, S. Wittine, M. Kleinstäuber, C. Weise Philipps University Marburg, MARBURG, Germany Introduction: With 75 % among women of reproductive age, premenstrual symptoms are very common. Symptoms range from emotional and cognitive to physical symptoms. Severe premenstrual syndrome (PMS) can have a strong impact on everyday functioning and quality of life. Impairment can be as serious as that of dysthymic disorders. A measure to assess this impairment would be necessary in the diagnostics of PMS,

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

however, has not yet been developed. In this pilot study we evaluated the validity and reliability of a new PMS-Impairment Measure (PMS-B). Method: A sample of 70 women reporting medium to severe premenstrual distress were assessed with the 22-item PMS-B during the premenstrual (luteal) phase. An exploratory factor analysis with direct oblimin rotation was conducted. Reliability and validity were tested using Cronbach’s alpha and correlational analysis. Results: The factor analysis revealed a two-factor solution accounting for 49% of the total variance. The resulting factors were labelled 'Mental Impairment' and 'Functional Impairment' and contained seven items each. Cronbach’s alpha of .93 for 'Mental Impairment', and .90 for 'Functional Impairment' indicated good reliability. Convergent construct validity was demonstrated by moderate to high correlations with the Pain Disability Index. Low correlation between the PMS-B-subscales and the subscales of the Big Five Inventory-10 indicated good divergent validity. Conclusions: The PMS-B was shown to be a valid, reliable and economic measure to assess impairment of PMS. In future research, confirmatory factor analyses as well as cross validations in different samples should be conducted to confirm our preliminary findings. CORRESPONDING AUTHOR: Dipl. Psych. J. N. Kues Philipps, University Marburg, Germany, [email protected] P343 PSYCHOMETRIC ANALYSIS OF A CHINESE-VERSION FATCOD AND NURSING CARE BEHAVIOR SCALE (NCBS) Y.H.C. Cheng, J.J.L. Liaw National Defense Medical Center, TAIPEI, Taiwan Purpose: The study was to determine the internal consistency, content, and construct validity of the Chinese-version Frommelt Attitudes Toward Care of the Dying (FATCOD) scale and NCBS for the end-of-life (EOL) patients and their families. Methods: The study adopted cross-sectional design. Convenience sampling was used to recruit nurses who met the study criteria. There are 318 nurses from different work units at four hospitals in the north areas of Taiwan. Factor analysis was used to examine the construct validity of the Chinese-version FATCOD and the NCBS. Cronbach alpha was used to examine the internal consistency of the two scales. Results: The results suggest that internal consistency was 0.83 and 0.95 for FATCOD scale and the NCBS respectively. There are four subscales of the FATCOD scale, and the KMO is 0.89, and the Bartlett’s test of sphericity is 4608.25 (p < 0.001). There are two subscales of the NCBS, the KMO is 0.94, and the Bartlett’s test of sphericity is 4397.89 (p < 0.001). The factor matrix of the FATCOD scale suggest the factor loading is greater than 0.46. The factor matrix of the NCBS suggests the factor loading is greater than 0.51. Conclusions: Both the Chinese-version FATCOD scale and NCBS are valid and reliable scales. Nurses can use the FATCOD scale to measure their attitude toward care of the dying patients, and the NCBS to measure their nursing care behaviors for EOL patients and their families. Nurses also can understand their care quality of the EOL patient and families. CORRESPONDING AUTHOR: Y.H.C. Cheng, National Defense Medical Center, Taiwan, [email protected] P344 QUANTIFICATION OF POSTURAL CONTROL IN ELDERLY DURING BALANCE TRAINING USING AN EXERGAME M. van Diest1, J. Stegenga1, H.J. Wörtche1, K. Postema2, G.J. Verkerke2, C.J.C. Lamoth2 1 INCAS3, ASSEN, Nederland 2 University Medical Center Groningen, GRONINGEN, Nederland Introduction. Fall injuries are responsible for significant disability among elderly. Exergames provide an opportunity for training balance ability, but

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

sensors and algorithms for assessment of postural control during gameplay are scarce[1]. Our objective is to identify age-related differences in postural control during an exergaming task. Methods. Nine elderly (age 72.9 ±4.9) and five young adults (age 21.8 ±2.0) performed 1-min task-embedded mediolateral weight-shift exergame trials while standing on force plates. Five conditions were applied in random order: neutral, lifting one leg and increased game speed, sway frequency (SF) and sway amplitude (SA). From lateral Center of Pressure (COP) displacements, SA, SF and root mean square (RMS) values were calculated. Smoothness and predictability of sways were quantified using index of harmonicity (IH) and sample entropy (SEn)[2]. Results. RM ANOVAs showed age effects on SA, RMS and SEn (p<0.05) and trends towards lower SF and IH in elderly. Condition effects were observed for SA, RMS and SF (p<0.05). Elderly showed lower values for all outcome measures indicating a smaller, slower, and more irregular sway movement compared to young adults. Conclusion. This study showed that during exergame tasks, age- and task-related differences in postural control can be quantified by analyzing the time-varying structure of sway patterns. These findings can be used to adapt exergames to the individual user. 1. van Diest M, et al. J. Neuroeng. Rehabil. 2013, 10:101. 2. Lamoth, et al. Gait & Posture 2011, 35:489-93. CORRESPONDING AUTHOR: M. van Diest, INCAS3, Nederland, [email protected] P345 AN EXAMINATION OF THE SELF-COMPASSION SCALE'S PSYCHOMETRIC PROPERTIES: RECONSIDER THE USE OF THE TOTAL SCORE A. López Angarita, R. Sanderman, E. van Sonderen, A. Ranchor, M.J. Schroevers University Medical Center Groningen, GRONINGEN, Nederland Introduction: Self-compassion has been defined as the action of taking care and giving kindness to the self in instances of suffering. The Self-compassion Scale (SCS; Neff, 2003) is currently the only available self-report questionnaire to measure self-compassion. The SCS is widely used, particularly in studies on mindfulnessbased and more recently, compassion-based psychological interventions. However, evidence for the scale’s psychometric properties is limited, as the scale has been mostly validated in college students. The aim of this study was to examine the factor structure, reliability, and construct validity of the SCS in a large, representative sample from the community. Methods: Analyses were made with a sample of 1643 Dutch adults who completed a series of self-reported questionnaires. Results: The SCS’s proposed six subscales could not be adequately identified. Instead, a two-factor solution was found, with each factor formed by the positively and negatively worded items (i.e. positive factor and negative factor), respectively. Internal consistency was good for the positive and negative factors, as well as for the total score. The negative factor correlated strongly, positively to measures of depressive symptoms, stress, negative affect, rumination and neuroticism, whereas the positive factor correlated positively with positive affect and negatively with depressive symptoms, stress and neuroticism. Conclusion: Taken into account the results regarding the factor structure as well as the strong association of the negative factor with rumination and neuroticism, our results question the common use of the total SCS score and propose the use of a score based on the positively items only. CORRESPONDING AUTHOR: A. López Angarita, University Medical Center Groningen, Nederland, [email protected]

S105

P346 ‘PARTICIPATION' AS A BEHAVIOURAL HEALTH OUTCOME: USING GOAL HIERARCHY TO ADVANCE MEASUREMENT AND INTERVENTIONS M. Kolehmainen1, M. Johnston2 1 Newcastle University, NEWCASTLE, United Kingdom 2 University of Aberdeen, ABERDEEN, United Kingdom Introduction: Participation (‘involvement in life situations’, WHO) is a key behavioural health outcome across conditions and age groups. There is no consensus over how participation should be operationalized for measurement and interventions. The most commonly discussed dimensions of participation are: observable behaviour (e.g. do sports) and subjective sense of ‘being’ (e.g. be involved in sports, be a sports person). A method is needed to integrate these two dimensions. Method: We propose to apply an existing framework, ‘Goal hierarchy’ (from Control Theory), as a method. This involves: (1) eliciting and mapping out the observable participation behaviours (‘actions’) and goals (‘behavioural goals’, or ‘desired states’ of being), and the relationships between them (goal-action ‘networks’); (2) using the resulting map to identify the ‘node’ behaviours/goals that connect the different parts of the network; and (3) specifying these node behaviours/goals as outcomes by articulating their Target, Action, Context and Time (TACT). We piloted this in research about children’s physical play. Results: (1) A range of measurable behaviours (e.g. ‘do things with others’, ‘ride a bike’) and goals (‘play with peers’, ‘be confident’) across the two participation dimensions were identified. (2) Three nodes emerged: ‘play’, ‘interact/do things with others’ and ‘develop’. (3) Data from the programme were further used to specify these, e.g., as ‘the child (T) to play (A) with others (C) daily (T)’. Conclusion: Goal hierarchy has potential to advance the identification and specification of participation outcomes, and development of multidimensional participation measures. This is compatible with personalised outcomes in clinical practice. CORRESPONDING AUTHOR: M. Kolehmainen, Newcastle University, United Kingdom, [email protected] P347 ARE SUICIDAL AND NON-SUICIDAL SELF-HARM CATEGORICAL OR DIMENSIONAL? A LATENT CLASS ANALYSIS R.C. O'Connor1, E. Ferguson2 1 Institute of Health & Wellbeing, GLASGOW, United Kingdom 2 University of Nottingham, NOTTINGHAM, United Kingdom Introduction. There has been intense debate concerning the nomenclature used to describe non-fatal self-injurious behaviour. In the USA, for the most part self-injurious behaviour is dichotomised into attempted suicide (AS) and non-suicidal self-injury (NSSI). By contrast, the majority of researchers in the UK adopt a dimensional approach, tending to use the term self-harm to describe all self-injurious behaviour and then defining the dimensions of self-harm. To date, no study has rigorously tested whether a dichotomy exists or not. Methods. This is secondary data analysis of the British Adult Psychiatric Morbidity Survey (APMS). The APMS assessed the prevalence of mental health disorders, self-harm, stressful life events as well as social support among 7403 adults in England. This study is based on a sub-sample, comprising those adults who reported self-harm with and without suicide intent (n=512). We used key marker variables (e.g., mental disorder) to investigate whether non-suicidal and suicidal self-harm were distinct groups. We conducted latent class modelling to determine whether selfharm sub-groups emerged from the analyses. Results. Two class and four class solutions were examined in terms of the probability of class membership and the distribution of key marker variables. None of the classes was differentiated by suicidal intent.

S106

However, severity of psychiatric morbidity and social functioning indices distinguished self-harm sub-types. Conclusions. This study supports the view that non-suicidal and suicidal self-harm are not categorically different phenomena. Established markers of self-injurious behaviour, not suicidal intent, best described the selfharm sub-types. The conceptual and clinical implications for the field are discussed. CORRESPONDING AUTHOR: Professor R.C. O'Connor, Institute of Health & Wellbeing, United Kingdom, [email protected] P348 VALIDATION OF THE PATIENT HEALTH QUESTIONNAIRE-9 FOR DEPRESSIVE SYMPTOMS IN COLOMBIAN RHEUMATOID ARTHRITIS PATIENTS H. Rogers1, D. Rivera1, H. Brotherton1, S. Olivera Plaza2, A.F. Córdoba Patiño2, P.B. Perrin3, M.L. Peña Altamar4 1 University of Deusto, BILBAO, Spain 2 Surcolombian University, NEIVA, Colombia 3 Virginia Commonwealth University, RICHMOND, VA, United States of America 4 Clínica Saludcoop Neiva, NEIVA, Colombia Introduction: The Zung Depression Scale is the most popular depression scale administered in Colombia, with 49 as a validated cut-off for identification of clinically meaningful depressive symptoms (CMDS). The Patient Health Questionnaire-9 (PHQ-9) is commonly used to identify CMDS in Colombian research studies, but its construct validity is currently unknown. Methods: 103 individuals with Rheumatoid Arthritis (RA) from Neiva, Colombia, were administered the Zung and PHQ-9. Results: The patients were 53.8 years old (SD=12.7) and 85.4% female (n=88) with an average of 8.4 years of education (SD=4.3). The mean PHQ-9 score was 7.5 (SD=4.9, range=0-26) with a Cronbach’s alpha of 0.77. Continuous PHQ-9 and Zung scores were highly correlated (rho=0.57, p<0.01). Using established cut-offs, 27% (n=28) had Zungidentified depression (≥49) and 31% (n=32) had PHQ=9 identifieddepression (≥10). Sensitivity=57.1%(47.6%-66.7%), specificity=78.7%(70.8%-86.6%), positive predictive value=50.0%(32.7%67.3%), and negative predictive value=83.1%(75.9%-90.3%). Likelihood ratio for a positive test result was 2.6 and 0.54 for a negative test result. Area under the Receiver Operating Characteristic (ROC) curve was 0.72 (0.61-0.82, p<0.001). The ROC curve suggests that the cut-off point that best classified patients with CMDS was a PHQ-9 of ≥7, with sensitivity=75.0%(66.6%-83.4%), specificity=56.0% (46.4%-65.6%), positive predictive value=38.9%(25.9%51.9%), negative predictive value=85.7% (79.0%-92.5%), and likelihood ratio for a positive test result, 1.70, and a negative test result, 0.45. Conclusions: The PHQ-9 appears to be a useful tool to identify CMDS in Colombian RA patients using a cut-off point of 7 or greater. The instrument has high internal consistency in this population and is quick and straightforward to administer. CORRESPONDING AUTHOR: H. Rogers, University of Deusto, Spain, [email protected] P349 MEASUREMENT INVARIANCE ACROSS ENGLISH AND SPANISH LANGUAGE MEASURES IN U.S. HISPANIC/LATINO ADULTS: RESULTS FROM HCHS/SOL C. Brintz1, L.C. Gallo2, P. Gonzalez2, M.M. Llabre1, V.L. Malcarne2, A. Nuñez2, F.J. Penedo3, S.C. Roesch2 1 University of Miami, CORAL GABLES, United States of America 2 San Diego State University, SAN DIEGO, United States of America 3 Northwestern University, CHICAGO, United States of America

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

Introduction: When making inferences about a construct from scores obtained on a measurement instrument administered in multiple languages, a concern is whether the instrument has the same structure across languages, so that scores can be interpreted similarly across groups. Literature provides little guidance with respect to minimum acceptable levels of invariance. The aim was to decide upon psychometric guidelines and assess measurement invariance across English and Spanish language versions of instruments in a study of health correlates in U.S. Hispanic/ Latino adults. Method: The sample consisted of 5,313 U.S. Hispanic/Latino adults who participated in the Hispanic Community Health Study/Study of Latinos(HCHS/SOL), Sociocultural Ancillary, a study of sociocultural/ psychological risk factors for chronic diseases. Measures were administered in English or Spanish. The authors decided upon a set of rules for declaring invariance using two fit indices. After establishing the factorial structure of each measure, multiple-group confirmatory factor analyses were compared to assess configural, metric, and scalar invariance across language groups. Initially, loadings and intercepts were estimated freely, after which loadings and intercepts were sequentially constrained equal across groups. Results: Out of 15 instruments, a measure of Familism did not meet minimum invariance requirements. Modifications to the instrument were made by removing variant items. The authors recommended that instruments meet requirements for configural language invariance before combining English and Spanish responses in analyses. Conclusion: Within cross-cultural research, it is important to assess measurement invariance before combining scores on instruments administered in multiple languages. This paper aimed to establish such guidelines for the HCHS/SOL. CORRESPONDING AUTHOR: C. Brintz, University of Miami, United States of America, [email protected] P350 COPING, PSYCHOPATHOLOGICAL SYMPTOMS, AND PAIN IN COLOMBIAN PATIENTS WITH FIBROMYALGIA H. Brotherton1, H. Rogers1, A. De Luis1, S. Olivera-Plaza2, M. PeñaAltamar3, I.L. Gomez-Gutiérrez2 1 Universidad de Deusto, BILBAO, Spain 2 Universidad Surcolombiana, NEIVA, Colombia 3 Clinica Saludcup, NEIVA, Colombia Introduction: Fibromyalgia (FM) is a neurosensory disorder characterized by widespread pain, joint stiffness, and fatigue. There is a bi-directional relationship between somatic symptoms and psychosocial factors. Little is known about this relationship in Latin American FM patients. Methods: Twenty-seven patients diagnosed with FM were recruited from ambulatory centers in Neiva, Colombia. All were women and almost all were from the two lowest socio-economic strata. Patients were administered the Visual Analog Scale (VAS) to evaluate pain, Brief COPE scale, and Symptom Checklist-90 Revised (SCL-90R). Recruitment will continue for the next six months. Results: Preliminary results indicate that neither socio-demographic characteristics nor medications used was associated with VAS pain rating. Higher VAS was associated with more use of COPE self-distraction (r=0.38, p<0.05) and COPE positive reframing (r=0.41, p<0.05), and the SCL-90R Global Severity Index, (r=0.55, p<0.01), Positive Symptom Total (r=0.45, p<0.05), Positive Symptom Distress Index (r=0.40, p<0.05), Somatization (r=.0.59, p<0.01), and Depression (r=0.51, p<0.01). A linear regression model indicated that SCL-90R Somatization was independently associated with VAS pain scores with marginal significance (B=0.15, p=0.07), controlling for SCL-90R Depression. Conclusions: Those with increased pain perception used two specific coping strategies - self-distraction and positive reframing - either because they engage in more coping overall or these strategies are not effective. Overall psychopathological symptoms and their severities are associated

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

with increased pain perception. Somatization, above and beyond depression, tended to be associated with increased pain. These findings may be able to be generalized to FM patients in low income areas in Latin America. CORRESPONDING AUTHOR: H. Brotherton, Universidad de Deusto, Spain, [email protected] P351 PAIN, NEGATIVE MOOD, AND DISCREPANCY IN EMOTIONAL SUPPORT IN CHRONIC PAIN PATIENTS: A PRELIMINARY STUDY K.S.O. Kim, S.C. Cho, Sungkun Chungnam National university, DEAJEON, South-Korea Introduction. Numerous studies have demonstrated that perceived emotional support has positive influence on pain and negative mood in chronic pain patients. However, a recent study has suggested that measuring discrepancy in preferred vs. perceived emotional support rather than perceived emotional support needs to be considered to predict pain and negative mood in chronic pain patients. Prior studies have relied on retrospective assessment, which may be insensitive to changes and effects of perceived emotional support or discrepancy in emotional support over the shorter term. Thus, the aim of this preliminary study was to examine the temporal patterning of pain, negative mood, perceived emotional support, and discrepancy in emotional support in chronic pain patients, by using a daily diary method. Methods.A total of 10 chronic pain patients seeking treatment in pain management centers located in Korea participated in this preliminary study for 15 days. Results. Multilevel random effects analyses indicated that (a) increases in discrepancy in emotional support were significantly associated with increases in same-day pain and negative mood, but perceived emotional support were not; (b) discrepancy in emotional support significantly predicted increases in pain and negative mood on the following day, but perceived emotional support did not. Conclusion. The results of this preliminary study suggest that discrepancy in emotional may lead to pain and negative mood in chronic pain patients. Also, measuring discrepancy in emotional support may be more relevant than perceived emotional support, to predict pain and negative mood. CORRESPONDING AUTHOR: K.S.O. Kim, Chungnam National university, South-Korea, [email protected] P352 PSYCHOLOGICAL DISTRESS IN PATIENTS WITH RHEUMATOID ARTHRITIS: DOES A MINDFUL ATTITUDE HELP ACROSS TIME? I. Nyklicek1, F. Hoogwegt2, T. Westgeest2 1 Tilburg University, TILBURG, Nederland 2 Máxima Medical Center, EINDHOVEN, Nederland Introduction: Mindfulness may diminish effects of adversities on psychological well-being in medical patients, but studies are scarce, especially in patients with rheumatoid arthritis (RA). Therefore, the aim was to examine the prospective moderating effect of mindfulness regarding psychological distress associated with disease activity and disability in patients with RA. Methods: Two-hundred-and-one patients with RA who started treatment with methotrexate or a TNF-alpha antagonist (mean age 57.4, SD 11.7, 55% women) completed questionnaires at baseline and at six and twelve months follow-up. Disease activity score was assessed by the rheumatologist. Results:Across time, disability, but not disease activity, was associated with psychological distress. Controlled for sex, age, education, marital status, disease activity, disability, and medication, mixed linear models analyses showed a strong association of higher baseline mindfulness with

S107

lower psychological distress across time up to the one year follow-up assessment (Beta coefficient = 0.036, p < .001). Baseline mindfulness also significantly moderated (i.e., diminished) the association between disability and psychological distress (p=.022). Conclusion: Mindfulness showed beneficial prospective associations with psychological distress in RA patients, both as a simple direct association and moderating the effect of disability. Thus, mindfulness may protect against psychological distress associated with disability in RA. Potential benefits of mindfulness-based interventions for prevention should be examined in this population. CORRESPONDING AUTHOR: dr. I. Nyklicek, Tilburg University, Nederland, [email protected] P352 HEALTHCARE PROFESSIONAL REACTIONS TO PATIENT PAIN: IMPACT OF KNOWLEDGE ABOUT MEDICAL EVIDENCE AND PSYCHOSOCIAL INFLUENCES L.D.R. De Ruddere1, L. Goubert2, M.A.L. Stevens2, M. Deveugele2, K.D. Graig2, G Crombez2 1 Universiteit Gent, GENT, Belgium 2 Ghent University, GHENT, Belgium Introduction. This study examined the impact of evidence concerning the presence of: a) a biomedical basis for pain, and b) psychosocial influences, on practitioner appraisals of patient pain experiences. Further, the potential moderating role of patient pain behaviour was examined. Methods. In an online study, 52 general practitioners (GPs) and 46 physiotherapists (PTs) viewed video sequences of 4 patients manifesting pain, with accompanying vignettes describing presence or absence of medical evidence and psychosocial influences. Participants estimated pain intensity, daily interference, felt sympathy, effectiveness of pain medication, self-efficacy, their likability and suspicions of deception. Results. Primary findings indicated lower perceived pain and daily interference, less sympathy, lower expectations of medication impact, and less self-efficacy when medical evidence was absent. The same results were found when psychosocial influences were present, but only when the patient displayed higher levels of pain behavior. Further, absence of medical evidence was related to less positive evaluations of the patients and to higher beliefs in deception in both professions. The presence of psychosocial influences was related to less positive evaluations and higher beliefs in deception in both professions. Conclusion. In sum, a range of contextual factors influence healthcare practitioner responses to patient pain. Implications for caregiving behavior are discussed. CORRESPONDING AUTHOR: L.D.R. De Ruddere, Universiteit Gent, Belgium, [email protected] P353 PAIN ASSOCIATED WITH WOUND CARE TREATMENT IN BURULI ULCER IN GHANA AND BENIN M. Alferink1, J. de Zeeuw1, C. Agossadou2, Y.T. Barogui2, G.E. Sopoh2, T.S. van der Werf1, Y. Stienstra1, A.V. Ranchor1 1 University Medical Center Groningen, GRONINGEN, Nederland 2 Programme de Lutte Contre la Lèpre et l’Ulcère de Buruli, Ministries of Health, COTONOU, Benin (Dahomey) Introduction: Buruli ulcer (BU) is a skin disease affecting people in remote areas with tropical climates such as Sub-Saharan Africa. Wound care includes changing dressings and cleaning the wounds, which is important for the healing of the wound. Wound care is performed in BU treatment centers in Benin and Ghana twice a week by a nurse. The patient is accompanied by a relative. Clinical observations reveal that there is fear prior to and pain during wound care, however, this has never been studied. Clinical pain assessment is usually by self-reported and observational methods, of which self-reported is the gold standard. The

S108

aim was to estimate the prevalence of fear prior to and pain during and after wound care in BU patients. Methods.50 patients and their relative, treated in one of the treatment centers in Ghana or Benin, were included between December 2012 and June 2014. Patients were above 5 years of age, and regularly attending wound care. Measures included observational (Face Legs Activity Cry Consolability) and self-reported pain (Faces Pain Scale-Revised /WongBaker Pain scale), and self-reported fear, reported by the patient as well as the relative during three consecutive weeks. Stable and modifiable factors influencing pain perception were measured, such as previous pain experiences, environment, invasiveness and duration of the procedure, use of pain medication and background pain. Results/Conclusion: The mean and range of pain ratings were on the lower edge, with large variability between patients. Pain scores declined over time after treatment. CORRESPONDING AUTHOR: Msc M. Alferink, University Medical Center Groningen, Nederland, [email protected] P354 ASSOCIATIONS OF JOB-DEMAND-CONTROL, EFFORTREWARD IMBALANCE, AND ORGANIZATIONAL JUSTICE WITH PAIN IN AN OCCUPATIONAL CROSS-SECTIONAL STUDY R.M. Herr1, B. Schmidt1, J.A. Bosch2, A. Loerbroks3, M.N. Jarczok1, J.E. Fischer1 1 Heidelberg University, MANNHEIM, Germany 2 University of Amsterdam, AMSTERDAM, The Netherlands 3 University of Düsseldorf, DÜSSELDORF, Germany Introduction: Work stress is associated with increased pain, one of the most prominent causes of long-term work-absenteeism. However, it is not yet clear which aspect of the work environment represents the most pertinent determinant of pain. This study is the first to simultaneously consider three established conceptualizations of work stress as determinants of pain; i.e., the Job-Demand-Control model (JDC), the Effort-Reward-Imbalance model (ERI), and Organizational Justice (OJ). Moreover, as both pain and work stress are typically measured by self-reports, common method variance (CMV) may result in spurious associations. Thus, another aim was to estimate associations between work stress and pain while controlling for CMV. Methods: In a cross-sectional study among employees in Germany (n=1,634, male=87%), validated questionnaires were used to assess JDC, ERI, and OJ. Participants rated the frequency of perceived pain for nine anatomical sites experienced over the previous year, which was summed to an overall pain score (OPS). We controlled for CMV using structural equation modelling. Results: In multivariate analyses, adjusted for physical-workload, demographics, socioeconomic-status, and lifestyle-factors, all three work stress measures were associated with OPS. After additional mutual adjustment for all work stress models, ERI showed the strongest (βERI=.26, p<.001), OJ a weak (βOJ=-.09, p=.004), and JDC no association with OPS (βJDC=.04, p>.1). After accounting for CMV, associations of ERI with OPS persisted. Conclusion: ERI and, to a lesser extend, OJ appear to be independent determinants of self-reported pain. CMV did not explain these associations. Identification of these ‘toxic’ elements may inform the development of workplace interventions. CORRESPONDING AUTHOR: R.M. Herr, Heidelberg University, Germany, [email protected] P355 WHERE DOES PAIN FIT IN? PERSPECTIVES OF PAIN AND MANAGEMENT IN PEOPLE WITH MULTIPLE SCLEROSIS A. Harrison1, A. Bogosian1, E. Silber2, L.M. McCracken1, R. Moss-Morris1 1 Institute of Psychiatry, LONDON, United Kingdom 2 King’s College Hospital NHS Foundation Trust, LONDON, United Kingdom

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

Introduction: Pain is a common and unpredictable symptom affecting around 63% of people with Multiple Sclerosis (pwMS). Studies have shown pain in pwMS is associated with poorer outcome. However, little qualitative research has explored people’s reactions to pain and pain management and views on service provision. Our aim was to understand how pwMS perceive, differentiate and respond to their pain, and identify views on pain management and service provision. Method: Twenty-five in-depth, semi-structured telephone interviews with pwMS with a range of demographic, illness and pain characteristics (neuropathic and non neuropathic) were conducted. Interviews were audiotaped, transcribed and analysed using inductive thematic analysis with elements of grounded theory. Results: Anger rather than depression or anxiety was the most frequent emotional response to pain. Descriptions of pain were often vivid but paradoxical in nature. Common and idiosyncratic pain-related causal beliefs were identified, and pain was perceived as unpredictable and progressive. A split between pain reduction and control versus management or acceptance of pain was observed. A focus on pain reduction or control appeared to coincide with an on going struggle and unhelpful cycles of symptoms and distress. While pwMS believed clinician’s efforts to treat pain was inadequate, well-organized and communicative services were attributed to improved management. Conclusion: This research uniquely portrays the ubiquitous and disruptive influence of pain in the wider MS context. It provides novel insight into specific pain-related beliefs, emotional reactions and disparate painmanagement attitudes. The results will be discussed in relation to pain interventions and service provision for pwMS. CORRESPONDING AUTHOR: Mr A. Harrison, Institute of Psychiatry, United Kingdom, [email protected] P356 WILLINGNESS TO UNDERGO UNSEDATED COLONOSCOPY WITH VIRTUAL REALITY AS NON-PHARMACOLOGICAL ANALGESIC S.J. Blokzijl1, K.F. Lamberts1, L.A. Van der Waaij1, J.M. Spikman2 1 Martini Hospital Groningen, GRONINGEN, Nederland 2 University of Groningen, GRONINGEN, Nederland Unsedated colonoscopy has advantages for patients and health care settings. Literature shows that 17-56 percent of the patients are willing to undergo colonoscopy unsedated. This percentage could be increased by offering non-pharmacological analgesics, like Virtual Reality (VR). VR is an active distraction technique, based on the theory of limited attention capacity and with a proven analgesic effect in burn care. This study investigated the willingness of patients to undergo unsedated colonoscopy with VR and identified patient-related factors associated with their choice. A questionnaire was sent to all adult patients referred for colonoscopy by our outpatient department between April and November 2013. The questionnaire was returned by 51 percent of 443 patients and of these, 23.7 percent was willing to undergo colonoscopy with VR. The main arguments pro were: receiving as little medication as possible, resuming daily activities sooner and participating in traffic independently afterwards. The main arguments con were: previous positive experience with colonoscopy, lack of affinity for VR and anticipated pain. The choice for VR was associated with lower age, higher educational level and the absence of procedure-related worries. Unexpectedly, no association was found with gender, abdominal pain as presenting symptom, anticipated pain, previous colonoscopy, anxiety as a trait, need for control, coping style and somatisation. Conclusion: 23.7 percent of the patients are willing to undergo unsedated colonoscopy with VR; this is comparable to studies with unsedated colonoscopy. The choice for VR was associated with lower age, higher educational level and absence of procedure-related worries. CORRESPONDING AUTHOR: MSc S.J. Blokzijl, Martini Hospital Groningen, Nederland, [email protected]

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

S109

P357 PREDICTING NON-IMPROVED SUBJECTIVE HEALTH IN RHEUMATOID ARTHRITIS PATIENTS WITH A CLINICAL RESPONSE TO BIOLOGIC TREATMENT E. Taal, M.A.H. Oude Voshaar, P.M. Ten Klooster, H.E. Vonkeman, M.A.F.J. Van de Laar University of Twente, ENSCHEDE, Nederland

patients experience elevated levels of fatigue already at an early stage of OA. This finding asks for more attention regarding fatigue in OA research, and warrants tailored psychosocial interventions for patients with elevated levels of fatigue.

Background: Modern pharmacological treatment is frequently successful in reducing disease activity in rheumatoid arthritis (RA). However, changes in clinical measures may not always correspond to a patient’s own perception of improvement. The objective of this study was to identify subjective predictors of self-reported health changes in patients with a good clinical response after one year of treatment in a tight control study. Methods: Patients of the Dutch Rheumatoid Arthritis Monitoring (DREAM) registry who completed baseline and 12 months assessments and achieved a good clinical response to 12 months of tight-control treatment were selected for analysis. Patients were categorized according to self-reported change in health status and binary logistic regression analysis was performed to identify predictors of patient reported health changes. Results: A good clinical response was achieved by 57% of 282 patients. Of These, 40 patients (24.7%) did not consider their health to have improved since starting treatment. Baseline clinical and patient reported measures did not significantly differ between improvers and nonimprovers. Twelve month change scores in bodily pain (OR = 1.03) and vitality (OR = 1.03) were significant (p <0.05) multivariate predictors of self-reported health improvement. Discussion: Most patients with a good response also considered their health to have improved. However, a substantial minority did not, despite significant improvements according to clinical measures. These results suggest that clinical improvements do not necessarily equate to improved subjective health. Pain and fatigue are patient reported outcomes that might be inadequately represented in current clinical outcome measures in RA research.

P359 THE EFFECT OF EMOTIONAL DIFFERENTIATION ON ACUTE PAIN AND NEGATIVE MOOD: A PRELIMINARY STUDY P.Y.S. Park, S.C. Cho, Sungkun Chungnam National University, DAEJEON, South-Korea

CORRESPONDING AUTHOR: Dr. E. Taal, University of Twente, Nederland, [email protected] P358 EXPLORING FATIGUE PATTERNS OVER TIME IN EARLY OSTEOARTHRITIS: RESULTS FROM THE CHECK STUDY C. Bode1, J. Botterman1, L. Siemons1, M. Van de Laar1, J. Dekker2 1 University of Twente, ENSCHEDE, Nederland 2 VU Medical Centre Amsterdam, AMSTERDAM, Nederland Objective - The goal of this study was to examine whether different groups of fatigue trajectories can be identified among middle-age patients with early symptomatic osteoarthritis (OA), to establish the level of fatigue severity within each of these fatigue groups, and to assess the role of age, gender and OA severity in relation to group classification. Methods - Data on fatigue and patient characteristics were collected from the CHECK cohort. Growth mixture modeling (GMM) was applied in order to identify distinct fatigue groups as well as to take into account the influence of radiographic OA severity (Kellgren-Lawrence grading), age and gender on group classification. Results - A stable, a U-shape, and an inverted U-shape trajectory were identified. The fatigue levels of the latter two groups (i.e. approximately two-thirds of the sample) were comparable to the Dutch cancer population. Females and younger OA patients experienced higher levels of fatigue within each group and females were more likely than males to exhibit the inverted U-shape fatigue pattern. The covariates were not related to the interindividual differences in fatigue change over time within each fatigue group. Conclusion - Three different fatigue patterns were found among a sample of early OA patients. These patterns indicate that a large number of OA

CORRESPONDING AUTHOR: Dr C. Bode, University of Twente, Nederland, [email protected]

Introduction. Emotional differentiation is the ability to report emotional experiences as separate and distinct from each other and has been known to make intense negative mood more manageable. It has been well established that greater negative mood is associated with higher pain perception. Thus, the aim of this preliminary study was to examine differences in pain perception and negative mood generated by the cold pressor task, depending on the level of emotional differentiation. Methods. This preliminary study consisted of two groups of college students in Daejeon, Korea; high emotional differentiation group (N = 14; Mean age = 22.36 years, SD = 2.79) and low emotional differentiation group (N = 14; Mean age = 21.36 years, SD = 2.50), all of whom performed the cold pressor task. Results. The results show that the high emotional differentiation group reported significantly lower affective pain (F(1, 27) = 4.63, p < .05) and marginally significantly lower sensory pain (F(1, 27) = 3.79, p = .06) than the lower emotional differentiation group. However, there were no significant differences between the groups in time of immersion (F(1, 27) = 1.24, p = n.s.) and negative mood during the task (F(1, 27) = 1.09, p = n.s.). Discussion. The results of this preliminary study suggest that emotional differentiation may be an important factor to have an influence on the experience of pain. Also, it may need to consider emotional differentiation to deal with the experience of pain in clinical settings. CORRESPONDING AUTHOR: P.Y.S. Park, Chungnam National University, South-Korea, [email protected] P360 DOES SELF-REGULATORY EFFICACY HELP UNIVERSITY STUDENTS PERSIST IN OVERCOMING BARRIERS TO WALKING FOR ACTIVE TRANSPORATION? M. Secora1, C. Gyurcsik1, M.A. Cary1, D.R. Brittain2 1 University of Saskatchewan, SASKATOON, Canada 2 University of Northern Colorado, GREELEY, United States of America Introduction: Although walking for active transportation (W-AT) produces health benefits, challenging (frequent; limiting) barriers may impact adherence. Individual differences in self-regulatory efficacy (SRE) may help people persist in overcoming such barriers. The purpose of this self-efficacy theory-based study was to examine whether people with higher or lower SRE differed in persistence to W-AT. Method: The experimental study design was a 2 (higher vs lower SRE) x 2 (higher vs. lower barriers challenge) mixed factorial, with 3 persistence measures as the dependent variables (time spent reporting barrier coping solutions, number of reported solutions, anticipatory persistence to WAT). Participants were 45 students (31 females) who W-AT to/from their university campus. Higher (n=22) and lower (n=23) SRE groups were identified via a median split. Each participant read higher and lower challenge vignettes in a lab setting, followed by completion of the persistence measures after reading each vignette. Results: 2 x 2 ANOVAs illustrated a significant interaction between SRE groups and challenge vignettes with the persistence measure of time spent

S110

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

reporting coping solutions, (p=.037). As expected, simple main effect results showed that the higher SRE group spent more time reporting solutions than the lower SRE group under the higher challenge condition (M higher=5 min. and 35 seconds; M lower=3 min. and 49 seconds). No other significant interactions were found. Conclusion: Findings provide first-ever evidence of the importance of higher SRE levels in helping students persistent in overcoming barriers to W-AT. Combining environmental and social cognitive approaches may best promote W-AT among university students.

p < .001. AE was a significant predictor of PA group, p < .001, and the interaction between pain intensity and AE was significant, p < .01. Approximately 73% of individuals with arthritis were correctly classified as active or inactive. Nagelkerke R square indicated that 24% of the variance in predicting PA group was accounted for by pain-related variables. Conclusions: Perceptions of pain intensity vary as a function of pursuing valued life activities despite pain (AE) to differentiate active and inactive individuals with arthritis. Enhancing AE among inactive individuals with high pain intensity may be useful in promoting PA.

CORRESPONDING AUTHOR: M. Secora, University of Saskatchewan, Canada, [email protected]

CORRESPONDING AUTHOR: Dr. P. Flora, University of Saskatchewan, Canada, [email protected]

P361 DO PAIN INTENSITY AND PAIN ACCEPTANCE PREDICT PHYICAL ACTIVITY AMONG ACTIVE INDIVIDUALS WITH ARTHRITIS? P. Flora, M.A. Cary, J.D. Sessford, L.R. Brawley, N.C. Gyurcsik University of Saskatchewan, SASKATOON, Canada

P363 PAIN ACCEPTANCE MODERATES THE PAIN ANXIETYADAPTIVE SELF-REGULATORY RESPONSES RELATIONSHIP AMONG EXERCISING ADULTS WITH ARTHRITIS M.A. Cary, L.R. Brawley, N.C. Gyurcsik University of Saskatchewan, SASKATOON, Canada

Introduction: Moderate to vigorous physical activity (MVPA) is recommended for disease self-management by arthritis agencies. Many individuals with arthritis are inactive and arthritis pain is a frequently reported barrier. Recent studies identify pain acceptance as a factor related to activity participation despite pain. We examined (1) pain intensity and (2) dimensions of pain acceptance (activities engagement [AE]; pain willingness [PW]) regarding their utility in predicting volume of MVPA among active individuals with arthritis. We hypothesized that pain acceptance dimensions would be related to MVPA, after controlling for pain intensity. Method: Participants from across North America (N = 134; 84% female; Mage = 49) with medically-diagnosed arthritis prospectively completed online measures of pain intensity and pain acceptance at baseline and MVPA two weeks later. Results: Hierarchical multiple regression indicated that pain intensity did not significantly predict MVPA, p > .05. After controlling for pain intensity, pain acceptance dimensions predicted MVPA, p<.001; R square change = .11, with AE making the greatest contribution. Conclusions: Despite pain being a frequently reported barrier to activity, pain intensity levels of 4.7 + 1.9 (out of 9) in this sample seemed to not impede MVPA. Findings suggest that individual differences in pain acceptance, particularly the pursuit of life activities regardless of pain (AE), relate to weekly MVPA.

INTRODUCTION. Exercise is internationally recognized as an effective self-management strategy for arthritis. Individuals frequently experience arthritis pain anxiety. However, no research has examined psychosocial processes that help individuals adapt to their anxiety and regularly exercise. According to the Fear Avoidance model, people may deal with pain anxiety through using adaptive self-regulatory (SR) responses (e.g., alter exercise time/type in response to anxiety). Pain acceptance, involving willingness to engage in valued activities despite pain, may moderate the anxiety-adaptive SR relationship. The present study examined whether acceptance moderated the anxiety-adaptive SR response relationship. METHODS. Participants were 136 exercising adults (M age=49.75 ±13.88 years) with medically-diagnosed arthritis who completed online measures of pain anxiety, pain acceptance, and adaptive SR responses. RESULTS. A hierarchical multiple regression analysis illustrated that the pain anxiety X pain acceptance interaction significantly predicted adaptive SR responses (p<.05). Follow-up analyses illustrated that when anxiety was lower, those with higher pain acceptance used adaptive SR responses less frequently than participants with lower acceptance. When anxiety was higher, both higher and lower acceptance was associated with the more frequent use of SR responses. CONCLUSIONS. Individuals with lower pain acceptance may use adaptive SR responses, regardless of their pain anxiety, causing a potential self-regulatory drain. However, higher pain acceptance may help individuals to respond to anxiety only when it is higher and needed, freeing up self-regulatory resources to regularly exercise. Future research should examine whether adaptive SR responses predict exercise. Findings would contribute to identifying psychosocial variables that should be targeted in tailored arthritis-exercise interventions.

CORRESPONDING AUTHOR: Dr. P. Flora, University of Saskatchewan, Canada, [email protected] P362 PAIN ACCEPTANCE AND BEING PHYSICALLY ACTIVE IN INDIVIDUALS WITH ARTHRITIS P. Flora, M.A. Cary, J.D. Sessford, L.R. Brawley, N.C. Gyurcsik University of Saskatchewan, SASKATOON, Canada Introduction: Physical activity (PA) is important for arthritis management. PA involvement may vary as a function of individual differences in pain intensity and pain acceptance. Pain acceptance is an acknowledged correlate of PA in chronic disease, but is seldom examined relative to pain intensity. Pain acceptance and pain intensity have not been considered together in differentiating active and inactive individuals. We examined whether pain intensity and pain acceptance predicted active/inactive groups. We hypothesized that (1) pain intensity and (2) dimensions of pain acceptance (activity engagement [AE], pain willingness [PW]) would predict PA group (active, inactive), and (3) a significant interaction would exist between pain intensity and pain acceptance. Method: Participants (N = 275; 87% female; Mage = 50) with medically-diagnosed arthritis completed online measures of PA, pain intensity, and pain acceptance. Results: Logistic regression indicated that the overall model predicted individuals to their activity groups, chi-square = 52.18, df = 5,

CORRESPONDING AUTHOR: M.A. Cary, University of Saskatchewan, Canada, [email protected] P364 VALIDITY AND RELIABILITY OF A SINGLE-ITEM MEASURE FOR SCREENING ADOLESCENTS' PHYSICAL ACTIVITY LEVELS J. Scott, P.J. Morgan, R.C. Plotnikoff, D.L. Lubans The University of Newcastle, NEWCASTLE, Australia Introduction: The aim of this study was to examine the test-retest reliability and concurrent validity of an adolescent single-item physical activity measure by comparing it to the existing Oxford Physical Activity Questionnaire (OPAQ) and accelerometer output. Methods: Participants were 123 adolescents (14.7 ± 0.5 years) from three secondary schools in NSW, Australia. To determine reliability, participants completed both questionnaires on two occasions separated by two

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

weeks. To assess validity, participants wore Actigraph GT3X+ accelerometers for a 7-day monitoring period and completed both physical activity questionnaires. Bivariate correlations between self-reported moderate-to-vigorous physical activity (MVPA) and accelerometer MVPA minutes/day were calculated. Results: The single-item [Intraclass correlation coefficient (ICC) =0.75, p <0.001] and the OPAQ (ICC =0.79, p <0.001) were both found to have moderate-to-strong reliability. Correlations between self-reported and objectively measured MVPA were similar for the single-item measure (r =0.44, p =0.01) and the OPAQ (r =0.48, p <0.001). Conclusion: These findings suggest the single-item measure can provide a reliable and valid assessment of youth physical activity. The single-item measure may have utility for screening adolescents and identifying youth for targeted physical activity interventions. CORRESPONDING AUTHOR: Mr J. Scott, The University of Newcastle, Australia, [email protected] P365 PERCEIVED STEEPNESS; AN ENVIRONMENTAL BARRIER TO INCREASED LIEFSTYLE PHYSICAL ACTIVITY F. Eves University of Birmingham, BIRMINGHAM, United Kingdom Introduction Increased stair climbing during daily living, a current public health goal, is avoided more by women, the old and the overweight than their comparators. Perception of potential climbs is related to physiological resources for climbing; individuals with reduced resources report potential climbs as steeper. Three studies tested perceived steepness as the cue that promotes avoidance. Methods Participants judged the steepness of stairs in a train station (n=269), a shopping centre (n=249) and a public square (n=734). Sex, age, height and weight were recorded. In the shopping centre and square, samples of pedestrians were recruited from those who chose the stairs and those who avoided them, stratified by sex, age and weight status. Results In the station, stairs appeared steeper for women, older participants and the overweight. Pedestrians who avoided stair climbing reported the stairs as steeper both after they had chosen the escalator (shopping centre) and before the choice was made (square), even when demographic differences were controlled for by stratified sampling and subsequent statistical analyses. Discussion Perceived steepness is a contextual, environmental cue that pedestrians use to avoid resource depletion by stair climbing. A perceptual signal that deters energy expenditure runs counter to public health efforts to encourage it. CORRESPONDING AUTHOR: Dr F. Eves, University of Birmingham, United Kingdom, [email protected] P366 IRANIAN WOMEN'S PHYSICAL ACTIVITY NETWORKS: INCREASING PHYSICAL ACTIVITY IN TRADITIONAL CULTURES H. Allahverdipour, Zeinab Javadivala, Ahmad Kush, Mohammad Asgharijafarabadi Tabriz university of Medical Sciences, TABRIZ, Iran Introduction: Social networks in traditional communities have been found to be a useful tool for enhancing womens' participation in preventative health behavior. Compared with men, women are more prone to sedentary habits and loss of fitness, so this study examined the effect of creating neighborhood networks for increasing physical activity among women.

S111

Method: Community health volounteers were used to identify social networks in their neighborhoods that included mosques and all activates in mosques. Other people (outside the network) were identified and invited, through both the primary community health volunteers of the networks, as well as active people within the networks. In addition, several physical activity neighborhood networks were created in the area covered by the health center for supporting local physical activity promotion clubs. Results: 20 physical activity neighborhood networks were established based on collaboration of community health volunteers and local health centers. Number of persons per net ranged from 8 to 12 people, consequently the number of people participating in local clubs grew up from 20 to 200. Discussion: Womens' physical activity networks can be the main persuasive factor for increasing physical activity. The leadership of community health volunteers as well as group relationship between women has an important role in physical activity promotion, especially in the traditional communities. CORRESPONDING AUTHOR: Prof. Allahverdipour Tabriz, University of Medical Sciences, Iran, [email protected] P367 DEVELOPMENT OF THE ENVIRONMENTAL ASSESSMENT TOOL (EAT) JAPANESE VERSION: A PILOT STUDY K.W. Watanabe, Y. ph.d Otsuka Hiroshima University, HIGASHI-HIROSHIMA, Japan Introduction: Physical activities are crucial for improving physical and psychological health. Although health promotion environments at worksites are one of the most important factors in increasing physical activity among employees, there is, as of yet, no Japanese scale that assesses them quantitatively. In this study, we developed a Japanese version of the Environmental Assessment Tool (EAT) and compared these EAT scores between Japanese and US samples. Method: We obtained the approval for Japanese translation of the EAT from the original research team in the US. We back-translated this measure to English and modified some items following the Japanese translation to improve clarity. The total EAT score ranged from 0 to 100 and consisted of three subordinate components (“Physical Activity Assessment,””Nutrition and Weight Management,”and “Organizational Characteristics and Support”). We administered the Japanese EAT at ten worksites that agreed to cooperate. T-tests were conducted to determine the differences in the total, and the subordinate, Japanese and US EAT scores. Results: The mean Japanese total EAT score was 20.4 (SD = 8.4), and this was significantly lower than the EAT score that has previously been obtained in US studies (t (20) = -3.94, p < .01)). The means of three subordinate scores were also lower than their US counterparts. Conclusions: Health promotion environments may be generally worse in Japanese worksites than they are in US worksites. CORRESPONDING AUTHOR: Mr. K.W. Watanabe, Hiroshima University, Japan, [email protected] P368 GO GREEN? COMPARING THE EFFECTS OF OUTDOOR AND INDOOR EXERCISE ON WELL-BEING AND PERCEIVED STRESS S. Klaperski, Anja Schempp University Freiburg, FREIBURG, Germany Introduction. There is profound evidence suggesting that physical exercise as well as exposure to natural environments is beneficial for health. In line with these findings recent studies found that physical exercise in natural environments (“green exercise”) has more positive effects on health than physical exercise in indoor settings. In the present study, we

S112

for the first time compare the effects of green exercise and indoor exercise on well-being and perceived stress in a German sample. Methods. Forty-seven participants completed questionnaires before and after exercising in a natural environment (running, n = 25) or in an indoor setting (aerobics, n = 22). We assessed well-being (Multidimensional Mood State Questionnaire), perceived stress (Perceived Stress Scale), as well as “greenness”and “calmness”of the environment. Results. Participation in physical exercise in general led to a significant reduction in bad mood (p < .001, ηp2 = .29), agitation (p < .001, ηp2 = .31) and perceived stress (p < .001, ηp2 = .54). No clear interaction effects emerged. However, multiple regression analyses revealed a significant linear relationship between the calmness value of the environment and reductions in perceived stress (β = -0.32, p = .01) and agitation (β = -0.46, p = .003); we found no significant relationships with regard to greenness. Conclusion. Our results support the known positive effects of physical exercise on mental health; however, effects did not differ between the two exercise groups. Yet, our findings suggest that beneficial effects might be related to the perceived calmness of the environment. CORRESPONDING AUTHOR: Dr. S.K. Klaperski. University Freiburg, Germany, [email protected] P369 PHYSICAL ACTIVITY BEHAVIOR AND NUTRITIONAL STATUS AMONG ADULTS FROM AN URBAN AREA I. Chirila1, E. Paulik2, F.D. Petrariu3, A. Albu3, V.L. Drug3 1 National Institute of Public Health - RCoPH, Iasi, Romania, IASI, Romania 2 Faculty of Medicine, University of Szeged, SZEGED, Hungary 3 University of Medicine and Pharmacy, Grigore T. Popa, IASI, Romania Introduction. More evidence haS accumulated from large, long term, epidemiological studies, supporting a strong association (inverse and independent) between physical activity, health, and cardiovascular and overall mortality in apparently healthy individuals. Physical activity levels strongly predicted mortality, independent of body size. Method: In this cross-sectional study performed in an urban area in Romania we used the long form of self-administered International Physical Activity Questionnaire and we measured body mass index, waist and body fat percentage. Results: Physical activity performed in the sample (n=207) was on average 3835 ± 3494.7 MET-minutes / week, with higher values for males (p<0.001). 10% of the subjects were sedentary and 40% had high levels of physical activity with differences between the sexes (p <0.01). Distribution of physical activity was balanced on areas of activity: leisure (20.5%), active transport (24.2%), housework and gardening (28.4%) and place of work (26.8%). Walking was the most common activity (43.8%), followed by moderate (41.6%) and high intensity activity (14.6%). Using different criteria, the prevalence of obesity was 12% (body mass index), 19.3% (body fat percent), 33.3% (waist), with significant differences between the two sexes. There was no significant correlation between the type or level of physical activity and nutritional status (regardless of diagnostic criteria), neither in men, nor in women (r <0.2); but no underweight subject was physically inactive. Conclusions: Our findings suggest assessing not just the nutritional status but also the physical activity level as independent factor for health. CORRESPONDING AUTHOR: I. Chirila, National Institute of Public Health - RCoPH, Iasi, Romania, Romania [email protected]

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

P370 THE ASSOCIATION OF VO2MAX WITH SERUM LIPIDS, REGULAR EXERCISE AND PHYSICAL ACTIVITY AMONG HEALTHY WORKERS Y. Miyazaki1, A. Uehara2, I. Mizuta2, K. Odagiri3, M. Yamamoto2, A. Inoue1, H. HIro1 1 University of Occupational and Environmental Health, Japan, KITAKYUSHU, Japan 2 Yamaha Health Care Center, HAMAMATSU, Japan 3 Hamamatsu University School of Medicine, HAMAMATSU, Japan Objective: In this study, we investigated 1) the relationship between VO2max and serum lipids, 2) the association of VO2max with physical activity and regular exercise and 3) the difference in VO2max according to occupational physical activity (i.e., sedentary and standing at work). Methods: A total of 3,084 healthy male workers, who worked at a company in 2011, enrolled in the study. Based on the Japanese guideline, the participants were divided into two groups according to VO2max. Regular exercise and physical activity were assessed by a self-administered questionnaire. Physical activity included occupational, leisure time and household physical activity. Occupational physical activity was divided into sedentary and standing at work. Mann-Whitney U test, t-test, logistic regression analysis and one-way analysis of variance were conducted. Results: 1) Participants with high VO2max had a significantly higher level of HDL cholesterol than those with low VO2max. Also, participants with high VO2max had lower levels of triglyceride and LDL cholesterol than those with low VO2max. The odds ratio of dyslipidemia in participants with high VO2max was 0.82 after adjustment for age, body mass index, smoking status, alcohol drinking, physical activity, regular exercise and fasting plasma glucose. 2) and 3) Both regular exercise and physical activity were positively and significantly associated with VO2max, while no significant association of occupational physical activity with VO2max was observed. Conclusions: Among healthy male workers, the higher VO2max had a significant association with serum lipids. Regular exercise and physical activity, but not occupational physical activity, were positively and significantly associated with VO2max. CORRESPONDING AUTHOR: Dr. Y. Miyazaki, University of Occupational and Environmental Health, Japan, [email protected]

P371 INSOMNIA AND MORTALITY IN A MIDDLE-AGE POPULATION B. Sivertsen1, S. Pallesen2, N. Glozier3, B. Bjorvatn2, P. Salo4, G.S. Tell2, R. Ursin2, S. Øverland1 1 Norwegian Institute of Public Health, BØNES, Norway 2 University of Bergen, BERGEN, Norway 3 The University of Sydney, SYDNEY, Australia 4 Finnish Institute of Occupational Health, HELSINKI, Finland Objective: Previous research suggests a possible link between insomnia and mortality, but findings are mixed and well-controlled studies are lacking. The aim of the current study was to examine the effect of insomnia on all-cause mortality. Methods: Using a cohort design with 14 years follow-up, official registry data on mortality were linked to health information obtained during 1997-99, as part of the communitybased Hordaland Health Study (HUSK), in Western Norway.6,236 participants aged 40-45 completed baseline information including selfreported information on insomnia (defined according to the DSM-V), sociodemographic factors, health behaviors, shift/night-work, obstructive sleep apnea symptoms, sleep duration, sleep medication use, anxiety, depression, as well as a range of somatic diagnoses and symptoms. Height, weight and blood pressure were measured. Information on mortality was obtained from the Norwegian Cause of Death Registry. Results: Insomnia was found to be a significant predictor of all-cause-mortality (hazard ratio [HR]=2.74 [95% CI:1.75-4.30]). Adjusting for several

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

possible confounders did not attenuate the effect (HR=3.29 [95% CI:1.63-6.62]). Stratifying on sex, the effect was significant in men (adjusted HR=4.33 [95% CI:1.40-13.38]); but not in women (adjusted HR=2.14 [95% CI:0.83-5.49]). The mortality risk among participants with both insomnia and short sleep duration (<6.5hours) was particularly high (adjusted HR=5.99 [95% CI:1.60-23.38]). Conclusion: Insomnia was associated with a four-fold risk of mortality in men during 14 years follow-up. The risk was especially high when insomnia was combined with short sleep duration. Establishing prevention strategies and lowthreshold interventions should consequently be a prioritized task for public health policy. CORRESPONDING AUTHOR: Prof. B.S. Sivertsen, Norwegian Institute of Public Health, Norway, [email protected] P372 SLEEP AND SCHOOL ATTENDANCE IN ADOLESCENCE: RESULTS FROM A LARGE POPULATION-BASED STUDY M. Hysing, S.H. Haugland, K.M. Stormark, T. Bøe, B. Sivertsen UniHealth, BERGEN, Norway Objective: To examine the association between sleep and school attendance in a population-based study of adolescents aged 17 to 19 years. Methods: A large population based study in the Hordaland county in Norway conducted in 2012, the ung@hordaland study, surveyed 10,220 adolescents aged 16-19 years (54% girls). Self-reported sleep measures included bedtime, rise time, sleep duration, sleep efficiency, sleep onset latency (SOL), wake after sleep onset (WASO), insomnia symptoms, and tiredness and sleepiness. School attendance was obtained from official registries. Results: Most sleep parameters were associated with increased risk of school non-attendance. After adjusting for gender and socioeconomic status, short sleep duration and sleep deficiency were the sleep measures with the highest odds of non-attendance ((OR=4.61, CI 95% 3.29-6.46) and (OR=3.26, CI 95% 2.67-3.99), respectively). Also, large bedtime discrepancies in weekend versus weekdays were associated with nonattendance (OR=2.43, CI 95% 1.93-2.02), as well as insomnia (OR=2.25, CI % (1.89-2.67) and daytime tiredness (OR=2.09, CI 95%1.70-2.57). The associations were somewhat reduced after additional adjustment for depression, but remained significant in the fully adjusted model. Conclusion: The demonstrated relationship between sleep problems and school absence suggests that careful assessment of sleep is warranted when adolescents present with extensive school absence. Future studies on how the sleep-school absence relationship in adolescence may impact later work affiliation in adulthood are needed. CORRESPONDING AUTHOR: Dr. M. Hysing, UniHealth, Norway, [email protected] P373 THE EFFECT OF AUSTRALIA'S BETTER ACCESS INITIATIVE ON INJECTING DRUG USERS' MENTAL HEALTH B.J.R.S. Stewart1, N. Sindicich2, D. Turnbull1, J.M. Andrews3, A.A. Mikocka-Walus4 1 University of Adelaide, ADELAIDE, Australia 2 University of New South Wales, SYDNEY, Australia 3 Royal Adelaide Hospital, ADELAIDE, Australia 4 University of York, YORK, United Kingdom Introduction: Injecting drug users are at significantly higher risk of experiencing psychiatric morbidity and thus represent a sentinel population for mental health policy and practice. This paper aimed to assess the influence of Australia’s publically funded Better Access initiative on regular injecting drug users.Method: Data were sourced from Illicit Drug Reporting System national surveys with 914 regular injecting drug users in 2006, preceding the Better Access reform, and 883 injecting drug users

S113

in 2012, following its implementation. Changes in rates of self-reported mental health problems and service use were assessed. K10 scores were available for a subsample of 2006 respondents and all 2012 participants, which were used to assess the validity of the self-report measures.Results: Rates of self-reported mental health problems increased from 38.3% in 2006 to 43.7% in 2012 - mainly due to increases in reported anxiety problems from 15.3% to 25.6%. Conversely, there was a decrease in overall mental health service use from 70.2% in 2006 to 58.4% in 2012, while at the same time, there was a proportional increase in the use of psychologists. These trends remained even after controlling for sociodemographic and medical differences between the 2006 and 2012 samples. K10 scores were significantly associated with self-reported mental health problems.Conclusions: Methodological limitations notwithstanding, the available data indicate that the Better Access national initiative may have resulted in better quality, although not better quantity, of access in regular injecting drug users. CORRESPONDING AUTHOR: Mr. B.J.R.S. Stewart, University of Adelaide, Australia, [email protected] P374 DEPRESSION AND ITS DETERMINANTS AMONG FLOODAFFECTED INTERNALLY DISPLACED PERSONS IN PAKISTAN K. Shafique1, Z.I. Arain2, F. Tareen3 1 Dow University of Health Sciences, KARACHI, Pakistan 2 Isra Medical University, HYDERABAD, Pakistan 3 Department of Health,, GOVERNMENT OF PAKISTAN, Pakistan Introduction. Flooding accounts for 40% of all natural disasters and causes considerable morbidity and mortality. Flood affected internally displaced persons (IDPs) also face financial and mental constraints. However, there is limited evidence available on mental health status of IDPs in developing countries, therefore, the aim of this study was to examine the prevalence of depression and its determinants among flood affected IDPs. Methods. This cross-sectional study was conducted in Hyderabad, Pakistan. Camps of flood affected IDPs were identified and a sample of 305 individuals was recruited through a simple random sampling using the official registration lists in the year 2009 and 2010. We used Hamilton Rating Scale for Depression (HRSD) to assess the depression. Logistic regression models were used to examine the determinants of depression. Results. Of 305 individuals, 154 (50.5%) were males and 151 females (49.5%), with a mean age of 38.6±12.5 and a mean depression score of 15.9±10.3. The prevalence was 20.6%, 8.5% and 42.3% for mild, mild to moderate and moderate to severe depression, respectively. Loss of house (OR 2.85, 95% CI 1.38-5.82) and a family member (2.58, 95% CI 1.203.98) were significantly associated with severe depression while no statistically significant association of age, sex and area of residence were observed with depression. Conclusion. This study suggests a high prevalence of moderate and severe depression among IDPs particularly among those who had lost their house or a family member due to flood disaster. Screening of high-risk persons is indicated to prevent the health consequences of depression. CORRESPONDING AUTHOR: Dr. K. Shafique, Dow University of Health Sciences, Pakistan, [email protected] P375 SLEEP DURATION AND GLUCOSE METABOLISM AMONG JAPANESE NON-DIABETIC PEOPLE A. Hino1, A. Inoue1, K. Mafune1, S. Yamamoto2, T. Nakagawa2, T. Hayashi2, H. Hiro1 1 University of Occupational and Environmental Health, Japan, KITAKYUSHU, Japan 2 Hitachi Health Care Center, Hitachi Ltd., HITACHI, Japan

S114

Introduction: The previous studies reported that short sleep duration is related to the lifestyle-related disease. We investigated the association between sleep duration and glucose metabolism among Japanese nondiabetic people. Methods: We used data from a health examination of 5,878 people without diabetes or psychiatric disease in FY 2011. Regarding sleep duration, participants were classified into four groups: less than 5 hours, 5-6 hours, 6-7 hours and 7 hours or more. Odds ratios (ORs) of insulin resistance, defined as HOMA-IR (homeostasis model assessment-insulin resistance) ≥2.0, were calculated by multiple logistic regression analyses adjusted for age, gender, quantity of drinking, smoking history, activity level, stress degree, insomniac presence and body mass index (BMI). Results: The ORs (95% confidence interval) of insulin resistance were significant for those who had 5-6 hours of sleep (OR=1.20 [1.02-1.40]) and less than 5 hours of sleep (OR=1.42 [1.08-1.87]) compared with those who had 6-7 hours of sleep. After additionally adjusted for BMI, ORs for short sleep duration groups were no longer significant, although significant OR for 7 or more hours of sleep group was observed (OR=1.34 [1.00-1.78]). Conclusions: Short sleep duration can cause insulin resistance. However, since these significant associations disappeared after additionally adjusting for BMI, the association of short sleep duration with insulin resistance may be explained by BMI. Too much sleep can also cause insulin resistance. Since this significant association was observed after adjusting for BMI, obesity appears to have little influence on insulin resistance among those who sleep longer hours. CORRESPONDING AUTHOR: Dr. A. Hino, University of Occupational and Environmental Health, Japan, [email protected] P376 R O L E O F E X P E C TAT I O N S A N D I N D I V I D U A L C H A R ACTERISTICS IN PLACEBO AND NOCEBO EFFECTS ON ITCH D.J.P. Bartels1, A.I.M. van Laarhoven1, E.A. Haverkamp2, O.H. WilderSmith2, R.T. Donders2, H. Middendorp1, P.C.M. Kerkhof2, A.W.M. Evers1 1 Leiden University, LEIDEN, Nederland 2 Radboud University Nijmegen Medical Center, NIJMEGEN, Nederland Background: Placebo and nocebo effects are treatment effects, unrelated to the treatment mechanism, which are induced by patients’ expectations of improvement or worsening. These effects have frequently been investigated with regard to pain, but have hardly been investigated with regard to itch, which is associated with high burden for patients when symptoms are chronic. This study investigated different methods of inducing placebo and nocebo effects on itch. Methods: Expectations regarding itch stimuli (electrical QST) were induced in healthy subjects by different expectation induction methods of verbal suggestion, conditioning or a combination of both, and compared to a control procedure. Individual characteristics (e.g., optimism and neuroticism) linked to placebo responding were exploratively investigated. Findings: Conditioning combined with verbal suggestion resulted in significant placebo and nocebo effects in comparison with the control group, in contrast to either procedure alone. Individual characteristics were especially associated with the nocebo effect. Discussion: This study showed that placebo and nocebo effects can be induced on itch through conditioning combined with verbal suggestion. The finding that individual characteristics may play a more prominent role in nocebo rather than in placebo effects on itch, is in line with previous studies showing that particularly negative affect and cognitions can enhance itch. CORRESPONDING AUTHOR: D.J.P. Bartels, Leiden University, Nederland, [email protected]

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

P377 THE HEART RATE VERIABILITY ON EMOTIONAL STIMULI OF THE LEFT- AND RIGHT-HANDED CHILDREN E.I. Nikolaeva Petersburg State Transport University, SAINT-PETERSBURG, Russian Federation Introduction: The hypothesis was offered that left-handed people have other mechanisms of central heart regulation than this of left-handed (Nikolaeva, Leutin, 2011). Methods: We have tried to understand the specificity of vegetative regulation of heart rate in different situations of 7-8 old children. Handedness of children was evaluated with a set of probes (Nikolaeva, Borisenkova, 2009). To analyze the heart rate variability of a child, an electrocardiogram was recorded. In a separate inquiry, we made three consecutive recordings of 300 R-R intervals. The first recording was an at rest one. At the two following recordings, we asked the child questions about punishment and encouragement in their families. Results: Regression analysis showed that the more left-sided characteristics a child had, the higher his (her) heart rate variability was in all situations. In any emotional situations, left-handed children demonstrated more expressive heart rate variability parameters. Our data are consistent with the hypothesis. CORRESPONDING AUTHOR: Prof E.I. Nikolaeva, Petersburg State Transport University, Russian, Federation, [email protected] P378 IS NEGATIVE MOOD COGNITIVE DEVIATION BIGGER OF MAJOR DEPRESSIVE PATIENTS INDUCED BY STRESSFUL LIFE EVENT G.Y. Li1, M. Yao1, H. Sheng2, B. Bai1, Z. Yang1, Z. Zhang1, H. Ma1 1 Jining Medical University, JINING, China 2 Affiliated Hospital of Jining Medical University, JINING, China Objective: To compare the difference of areas of brain activity stimulated by picture of negative mood, in female patients with first episode major depression induced by stressful life events. Methods: Using cross visual stimulation with pictures of neutral and negative mood, for female patients with first episode of major depression induced by stressful life events (SLE), and for those without SLE. Functional magnetic resonance imaging was taken at the same time, statistical analysis was performed by SPM8 to compare the difference in the activation area. Results: Compared with the group not induced by SLE, the patients group induced by SLE brain area significantly increased the activation of the bilateral superior temporal gyrus, left middle temporal gyrus, left Middle occipital gyrus, left medial frontal gyrus, right gyrusfrontalis inferior, bilateral anterior central gyrus, bilateral posterior central gyrus,bilateral middle frontal gyrus, right precuneus, left paracentral lobule, bilateral thalamus, bilateral hippocampal, and left cerebellum, but no brain area activity was significantly reduced. Conclusions: Activation of multiple brain regions in depressant patients induced by SLE abnormal in the loop of the mood processing, indicating that these patients may have more negative cognitive biases to the perception of stimulation of negative mood. CORRESPONDING AUTHOR: G.Y. Li, Jining Medical University, China, [email protected] P379 ASSOCIATIONS BETWEEN COPING STRATEGIES AND WELLBEING IN WOMEN DIAGNOSED WITH BREAST CANCER: A META-ANALYSIS P. Kvillemo, Richard Bränström Karolinska Institute, STOCKHOLM, Sweden

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

INTRODUCTION: The aim of this meta-analysis was to determine the strength of the evidence for the effectiveness of different coping strategies among women diagnosed with breast cancer. Examination of which types of coping that are related to physical and mental health, and well-being, as well as exploration of the potential moderating influence of situational and measurement factors on the association between coping and outcomes, were carried out. METHOD: Studies from Europe and North America published before April 2013 in peer-reviewed scientific journals, with a study sample of n ≥30 breast cancer patients (18 years or older), with quantitative measures of coping and psychological or physical health outcomes, were selected. Correlation coefficients between different types of coping styles and outcomes were gathered from each study. Random effects models were used to calculate effect sizes and these are presented as correlations. A total of 77 articles reporting data from 11 462 individuals were included in the final analysis. RESULT: Acceptance and Positive Reappraisal were related to more favorable outcomes in people coping with breast cancer. Further, use of avoidant or disengagement coping strategies were associated with less favorable outcomes. In a numb er of cases situational and measurement factors influenced the association between coping and outcome. Over all, the effect sizes were small to medium. CONCLUSION: The meta-analysis indicates that efforts to facilitate adaptation to stress, such as Accep tance and Positive Reappraisal, are particularly beneficial in coping with stressors related to breast cancer. CORRESPONDING AUTHOR: PhD student P. Kvillemo, Karolinska Institute, Sweden, [email protected] P380 UPRIGHT POSTURE IMPROVES EMOTIONS AND STRESS RESPONSE IN PEOPLE WITH MILD TO MODERATE DEPRESSION E. Broadbent, C. Wilkes, R. Kydd, M. Sagar The University of Auckland, AUCKLAND, New Zealand Introduction. Slumped posture has been shown to be associated with depression in cross-sectional research. This study aimed to investigate the effects of an intervention to correct slumped posture in people with mild to moderate depression using a stress paradigm. It was hypothesised that upright posture would buffer stress responses and protect against negative emotions. Method. Sixty-one participants with mild-moderate depression were randomized to usual posture (control) or upright posture (intervention). Both groups were fitted with physiotherapy tape to maintain this posture, with a cover story to reduce expectation effects. Participants completed the Trier Social Stress Test, a writing task and questionnaires. Linguistic Analysis and Word Count was performed on speech and writing. Blood pressure was measured, and posture assessed by video tracking technology. Results. At baseline, the total sample had significantly more slumped posture compared to a non-depressed normative sample. During the tasks, the intervention group maintained a significantly more upright posture, had significantly lower fatigue and more positive mood than the control group. They wrote and spoke a significantly greater total number of words, used fewer personal pronouns and more sadness words. Upright shoulder angle was associated with lower negative emotions, anxiety and blood pressure, and more ingestion words. Conclusions. An intervention to correct slumped posture in people with mild to moderate depression can increase positive affect and rate of speaking, and reduce fatigue and self-focus, during a social stress test. This novel research has potential implications for therapeutic interventions for stress and depression. CORRESPONDING AUTHOR: Dr E. Broadbent, The University of Auckland, New Zealand, [email protected]

S115

P381 SERUM DEHYDROEPIANDROSTERONE-SULFATE RELATED TO CARDIOVASCULAR STRESS REACTIVITY IN JAPANESE MEN AND WOMEN K. Hirokawa1, M. Nagayoshi2, T. Ohira3, M. Kajiura4, A. Kitamura4, M. Kiyama4, T. Okada4, H. Iso2 1 Baika Women's University, IBARAKI, Japan 2 Graduate School of Medicine, Osaka University, SUITA, Japan 3 Radiation Medical Science Center, Fukushima Medical University, FUKUSHIMA, Japan 4 Osaka Center for Cancer and Cardiovascular Disease Prevention, OSAKA, Japan Introduction: Cardiovascular stress reactivity is a predictor of atherosclerosis and cardiac events. Dehydroepiandrosterone (DHEA) protects against cardiovascular diseases, but results are inconsistent among previous studies. We aimed to investigate the association between dehydroepiandrosterone-sulfate (DHEA-s) and cardiovascular stress reactivity in Japanese men and women. Method: Among 979 healthy Japanese people (338 men and 641 women), serum DHEA-s, systolic and diastolic blood pressure (SBP and DBP), heart rate, heart rate variability, and peripheral blood flow were measured under rest and two different tasks. Mean differences in measured variables during tasks and a post-task period were calculated as a decline in stress reactivity. Stratification by menopause status for women and by age group based on the median split for men was conducted as a subanalysis. Variables of stress reactivity were adjusted for multiple potential confounding factors. Results: In men, there was no significant association between DHEA-s and a decline in stress reactivity. In women, DHEA-s showed positive associations with a decline in SBP and DBP (standardized-beta=0.12, p=0.020; 0.17, p=0.002, respectively). When stratified by menopause status for women, those significant positive associations remained in pre-menopausal women (0.14, p=0.037; 0.18, p=0.005). For men, when stratified by age group, no significant association was found, except for a decline in heart rate, showing a tendency toward significance in the older group ('0.17, p=0.059). Conclusion: DHEA-s may function to adjust blood pressure reactivity to stress, especially for pre-menopausal women. CORRESPONDING AUTHOR: Ms. K. Hirokawa, Baika Women's University, Japan, [email protected] P382 ASSOCIATIONS AMONG COGNITIVE DISTORTION, DEPRESSION, AND QOL IN PERSONS WHO WORK AT THE WELFARE SERVICE M. Ota, S. Takeda Tottori University, SHIMANE, Japan Objective: The purpose of this study was to examine whether cognitive distortion is associated with quality of life (QOL) and depression in persons who work in the welfare service. Methods: Subjects were 219 persons who work in the welfare service. The questionnaires were Workers' Cognitive Distortion Scale; WCDS, including cognitive distortion of self-contained/ cognitive distortion of environment-dependent, Self-rating Depression Scale; SDS, MOS 36-item Short-Form Health Survey; SF-36, including role-physical/ vitality/ social functioning/ role-emotional. Results: The result indicated that cognitive distortion is an influence on vitality of QOL through depression, using covariance structure analysis. As for influence of depression on cognitive distortion, it shows that self-contained cognitive distortion was bigger than cognitive distortion that is environment-dependent, and depression and vitality of depression were associated with each other, which means vitality decreases as depression increases, and depression increases as vitality increases.

S116

Conclusion: These results suggest that cognitive distortion influences not only depression but also daily life, especially vitality. From this aspect, it suggested using cognitive distortion to improve depression, or attaining a higher QOL. CORRESPONDING AUTHOR: M. Ota, Tottori University, Japan, [email protected] P383 SELF-ESTEEM AND STRESS AS ASSOCIATED WITH DIURNAL PROFILES OF SALIVARY ALPHA-AMYLASE AND CORTISOL IN MID-ADOLESCENTS L. Folkesson Stockholm University, STOCKHOLM, Sweden Salivary cortisol and alpha-amylase (sAA) that reflect hypothalamopituitary-adrenal axis (HPA-axis) activity and sympathetic activity within the autonomic nervous system (ANS) respectively, are biomarkers with pronounced diurnal rhythms. While research on salivary cortisol is increasing, little is known about the diurnal rhythm of salivary alphaamylase, particularly in adolescents. Also, the linkages between individual factors and self-reports of stress as related to HPA-axis activity and autonomic/sympathetic functioning remain to be investigated. This study set out to investigate diurnal rhythms of salivary cortisol and alphaamylase in 14-16 year-old girls and boys. Moreover, the study investigated whether stress and self-esteem are related to aggregate salivary cortisol and alpha-amylase measures. Besides self-reports in questionnaires, self-administered salivary samples were collected from 47 girls and 23 boys during a school day. Results showed girls had higher levels of morning cortisol than did boys, while there were no differences in morning or diurnal sAA. Additionally, self-esteem and self-reported stress were associated with different measures of cortisol and sAA but for girls only. Taken together, the findings suggest that both self-reported stress and self-esteem are linked to various aspects of sympathetic ANS activity and HPA-axis activity, particularly among mid-adolescent girls. CORRESPONDING AUTHOR: L. Folkesson, Stockholm University, Sweden, [email protected] P384 CHANGES IN HEART RATE VARIABILITY AFTER PHYSICAL EXERCISE IN PATIENTS WITH IRRITABLE BOWEL SYNDROME T. Hamaguchi1, J. Tayama2, T. Saigou2, T. Tomiie3, M. Kanazawa4, K. Kayaba1, S. Fukudo4 1 Saitama Prefectural University, School of Health and Social Services, Saitama, Japan 2 Center for Health and Community Medicine, Nagasaki University, NAGASAKI, Japan 3 School of Psychological Science, Health Sciences University of Hokkaido, SAPPORO, Japan 4 Tohoku University, Graduate School of Medicine, Miyagi, Japan Introduction: We hypothesized that exercise, with breathing method, would have a beneficial effect on sympathetic nervous system in irritable bowel syndrome (IBS). Methods: Ten healthy controls and 20 untreated IBS participants underwent experimental muscle stretching with breathing method for 4-weeks. Participants were instructed to breathe at two different breathing rates (6 bpm and naturalness). All participants were underwent muscle stretching, the 20 IBS were randomly divided into two groups: control group (n=10), with spontaneous breathing rate during the experiment; intervention group (n=10), with breathing 6 bpm for 15 min every day for 28-day. We detected heart rate variability (HRV) of participants durring mental-arthmetic task. Changes in the Gastrointestinal Symptoms Rating Scale (GSRS), State Trait Anxiety Inventory and HRV during psychological stress were compared between controls and IBS subjects before and after intervention. A three-

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

factor analysis of variance with period (before vs. after) as the withinsubject factor and group (IBS vs. Control), and breathing method (6 bpm vs. naturalness) as the between-subject factors was carried out on HRV. Results: GSRS score were significantly decreased after intervention in patients with IBS (p<0.01). Low frequency (LF) power band in HRV showed significant interactions between period and groups (p<0.001). At the baseline, no significantly change in LF was found in exercise with 6 bpm group compered with breathe controls. After the intervention, LF was significantly decreased in 6 bpm group during stress. Conclusion: Our results suggest the possibility of improving IBS pathophysiology by breath control during muscle stretching as indicated by parasympathetic nervous responses. CORRESPONDING AUTHOR: Prof. T. Hamaguchi, Saitama Prefectural University, School of Health and Social Services, Japan, [email protected] P385 DEPRESSED BUT HYPERACTIVATED: THE ROLE OF PSYCHOPHYSIOLOGICAL ASSESSMENT IN SUBCLINICAL HYPOTHYROIDISM C.A. Pruneti1, C. Cosentino1, F. Monzani2, A. Innocenti1, D. Sgromo1 1 University of Parma, PARMA, Italy 2 University of Pisa, PISA, Italy Introduction: Subclinical hypothyroidism is characterized by elevated TSH levels in the face of quite normal thyroid hormone concentrations. Several studies report that these subjects show some neuropsychological and behavioral alterations and features, but little, or nothing, has been studied about the autonomic characteristics of these people (Monzani et al., 1993). Methods: Since 1998, fifty patients diagnosed with a form of pre-clinical hypothyroidism were consecutively recruited. In addition to medical routine subjects underwent an investigation of the reported symptoms by the Crown & Crisp Experiential Index and a Psychophysiological Assessment. A control group of fifty subjects were given the same evaluation. Results: Data shows a significant elevation of some of CCEI subscales: Anxiety, Depression, Somatic Complaints. At the autonomic level there is a general pattern of excessive arousal. Conclusion: A possible interpretation of this data, is that in subclinical hypothyroidism, the body tries to support and integrate the general lack of energy caused by the hypothyroidism, with an acceleration of the autonomic activity. If this data will be confirmed, could represent a very important discovery with a useful fallout into the clinical practice. One of the most important clinical consequences of hypothyroidism, is the presence of depressive symptoms, often treated by antidepressant. In case of hyper autonomic activation, the antidepressant could cause several problems in the ménage of the case with a lot of negative consequences. A brief psychophysiological assessment would seem to be a good method to be used to better understand the clinical picture. CORRESPONDING AUTHOR: Prof. C.A. Pruneti, University of Parma, Italy, [email protected] P386 WORK STRESS AND HAIR CORTISOL LEVELS AMONG WORKERS IN A BANGLADESHI READY-MADE GARMENT FACTORY M. Steinisch1, R. Yusuf2, J. Li3, T. Stalder4, J.A. Bosch4, O. Rahman5, J.E. Fischer1, A. Loerbroks3 1 Mannheim Institute of Public Health, MANNHEIM, Germany 2 School of Life Sciences, Independent University, Bangladesh, DHAKA, Bangladesh 3 Institute of Occupational and Social Medicine, DÜSSELDORF, Germany 4 Department of Psychology, DRESDEN, Germany 5 School of Public Health, Independent University, Bangladesh, DHAKA, Bangladesh

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

Background. Evidence on the association of work stress with cortisol levels is inconsistent and mostly stems from Western countries, with limited generalizability to other regions of the world. This study therefore set out to explore associations of work stress components with long-term integrated cortisol levels in hair among 175 workers of a ready-made garment (RMG) factory in Dhaka, Bangladesh.”“ Methods. Work-related demands (WD), interpersonal resources (IR) and work-related values (WV) were assessed using a psychometrically evaluated questionnaire consisting of a 7-item version of the Effort-RewardImbalance model, which was combined with five setting-specific items derived from ethnographic research. Hair cortisol concentrations (HCC) were analyzed by liquid chromatography-mass spectrometry. Multivariate linear regression was used to estimate associations of HCC with the three work stress components. Results. The mean HCC equaled 3.31 (SD 2.62) pg/mg. HCC were found to be significantly associated with WV (beta= 0.193; p= 0.026), but not with WD and IR. Separate post-hoc analyses of the three WV items revealed that this association was largely driven by one particular item on “promotion prospects”implying that the perception of good promotion prospects was associated with higher HCC. Conclusion. The finding of elevated HCC with good promotion prospects may seem counter-intuitive in light of the evidence from Western work settings. However, being promoted in the Bangladesh RMG industry may represent a stressful experience: job promotions are rare in this setting and are associated with the need to meet exceptional demands. Thus, this finding may represent a culture-specific association. CORRESPONDING AUTHOR: M. Steinisch, Mannheim Institute of Public Health, Germany, [email protected] P387 LOWER HEART RATE VARIABILITY OVER THE WORKING DAY IS ASSOCIATED WITH ELEVATED HAIR CORTISOL LEVELS B. Serwinski, A. Steptoe University College London, LONDON, United Kingdom Introduction. There is ample evidence linking work-related stressors to elevated cortisol levels and cardiovascular activity (e.g. increased systolic blood pressure) which are subsequently related to negative health outcomes. Heart rate variability (HRV) is a promising marker of cardiac autonomic function and reduced HRV has been associated with other adverse biomarker outcomes and with higher cardiac morbidity and mortality. Conventional cortisol assessments employing salivary or plasma specimens are limited as they provide momentary cortisol exposure. Hair cortisol reflects long-term cortisol exposure and might therefore further elucidate the relationship between psychological stressors and chronic illness. The aim of the present study was to investigate biological factors in the work environment in a longitudinal design. Method. Working women (n=42, aged 24-55) completed 24-hour ambulatory HRV measurements over a work and a leisure day. Four years later, cortisol concentration from scalp hair (3cm) was measured. Results. HRV was analysed using frequency-based measures. We found an inverse relationship on the work day between high frequency HRV (HF-HRV) and hair cortisol levels independently of age, socioeconomic status, BMI and hair treatment, B = -0.297, 95% C.I. -0.60 - -0.004, p < 0.05. HRV over the leisure day was not associated with cortisol. Conclusions. HF-HRV reflects parasympathetic control of cardiac activity, so these results indicate that women with less parasympathetic control have greater hair cortisol concentrations several years later. These findings not only highlight the impact of work stress on the nervous and endocrine system, but as well further validate hair cortisol analysis as a stress-related biomarker. CORRESPONDING AUTHOR: B. Serwinski, University College London, United Kingdom, [email protected]

S117

P388 CONSISTENT ASSOCIATIONS BETWEEN MEASURES OF PSYCHOLOGICAL STRESS AND CMV REACTIVATION IN A LARGE OCCUPATIONAL SAMPLE J. Rector1, J.B. Dowd2, A. Loerbroks3, V.E. Burns1, P. Moss4, M.N. Jarczok5, T. Stalder6, K. Hoffman5, J.E. Fischer5, J.A. Bosch7 1 University of Birmingham, BIRMINGHAM, United Kingdom 2 CUNY School of Public Health and CUNY Institute for Demographic Research (CIDR), NEW YORK, United States of America 3 Institute of Occupational and Social Medicine, Centre for Health and Society, Fa, DÜSSELDORF, Germany 4 Cancer Research UK Centre, University of Birmingham, BIRMINGHAM, United Kingdom 5 Mannheim Institute of Public Health, Social and Preventive Medicine (MIPH), Mann, MANNHEIM, Germany 6 Department of Psychology, Technische Universität Dresden, DRESDEN, Germany 7 Department of Psychology, University of Amsterdam, AMSTERDAM, The Netherlands Introduction: Cytomegalovirus (CMV) has been implicated in biological aging and impaired health. Preliminary evidence, largely accrued from small-scale studies, suggest that stress may promote reactivation of this virus in infected hosts. However, absent is evidence from larger and more representative samples that allow better statistical adjustment for confounding and analyses of mediating factors. Methods: Participants were factory workers (n=887; Mean age=44; 88% male) participating in a periodical health check. Validated questionnaires assessed anxiety, depression, vital exhaustion, and perceived mental health, as well as demographic, socioeconomic (SES) and lifestyle variables. Plasma was analyzed for the presence and levels of CMV-specific IgG antibodies (CMV-IgG), indicating infection and viral reactivation, respectively. Results: SES (education and job status) was negatively associated with CMV infection prevalence (p<.05). CMV-IgG was significantly associated with increased anxiety (β=.14), depression (β=.11), vital exhaustion (β=.14), and decreased subjective mental health (β=-.14), while adjusting for age, gender, marital status, smoking, and SES. Exploratory analyses showed that these associations were stronger in low SES individuals. Analyses did not support proposed pathways linking stress to viral reactivation, such as inflammation (assessed as CRP) and HPA activity (diurnal cortisol and awakening response). Conclusion: In the largest study to date, we observed consistent associations between CMV-IgG levels, signifying viral reactivation, and multiple indicators of psychological stress. These results demonstrate the robustness of prior findings, and extend these to a general working population. We propose that stress-induced reactivation of CMV warrants further research as a psychobiological mechanism linking stress, aging and health. CORRESPONDING AUTHOR: Mr J. Rector, University of Birmingham, United Kingdom, [email protected] P389 RELATIONSHIP BETWEEN POST-TRAUMATIC STRESS DISORDER AND RESILIENCE IN CHILDREN AFTER DEBRIS FLOW DISASTERS (ZHOU QU) L. Zhang, J. Zhu, X. Zhu Second Hospital of Lanzhou University, LANZHOU, China Objective: Research the incidence in children of the post-traumatic stress disorder after the debris flow disaster of Zhou Qu. Analyzing the influential factors, estimate the condition of the children’s ego resilience, discussing the relationship between the post-traumatic

S118

stress disorder and children's ego resilience. To provide scientific basis for predicting the incidence of the post-traumatic stress disorder and taking the effective measures of psychological intervention. Methods: 596 students are selected with the method of random cluster sampling to be investigated with general condition survey scale, children ego resilience scale, post-traumatic stress disorder, screening of questionnaire. Obtained 592 valid questionnaires, with SPSS17.0 data analysis. Statistical methods used: descriptive statistics, independent sample t-test, chi-square test, correlation analysis, linear regression analysis(etc.) Result: 45 students were positive for PTSD, accounting for 7.60% of the total, 15 male students, accounting for 2.53% of the total, and 30 female students accounting for 5.07% of the total. The females in the PCL-C scale scored higher than the average of male, in negative group of PTSD, the females average scores are higher than average males. There are deferences between male and female in PCL-C scale score significantly (P<0.05). TThere are no significant differences between Male and female in resilience scale score. PTSD positive group psychological screening resilience total cent and factors points were lower than negative groups (P<0.05), there are significant differences between the total score and mental flexibility, PTSD total score is a significant negative correlation. Conclusion: Resilience can be used as predictors of incidences of PTSD. CORRESPONDING AUTHOR: L. Zhang, Second Hospital of Lanzhou University, China, [email protected] P390 EFFECTS OF THROWING LIVE GRENADE ON SOLDIERS' PSYCHOLOGY AND PHYSIOLOGY AT DIFFERENT MILITARY AGES W. Lu, Z.J. Zhang, H.B. Fu The 256th Hospital of PLA, Zhengding, HEBEI, China Objective: To investigate effects of throwing live grenade on soldiers’ psychology and physiology at different military ages. Methods: Two hundred and four male soldiers living in the same camp were divided into three groups according to military ages: recruits(≤3 months), soldiers (12-24 months), Non-comissioned officers(NCOs≥24 months). Twenty items of Hamilton Self-rating Anxiety Scale (SAS) and pulse rates measurements were respectively carried out among the subjects four weeks and half an hour before throwing live grenade. Results: The results of SAS and pulse rates of all subjects showed significant differences (P<0.05) between four weeks and half an hour before throwing live grenade. Four weeks before throwing grenade, SAS and pulse rates among recruits, soldiers (12-24 months) and NCOs had significant differences(P<0.05), while had no differences between soldiers (12-24 months) and NCOs (P>0.05). Half an hour before throwing grenade, SAS and pulse rates between recruits and NCOs had significant differences(P<0.05), while had no differences between recruits and soldiers (12-24 months) (P>0.05); SAS between soldiers (12-24 months) and NCOs had significant differences(P<0.05), while pulse rates had no significant differences (P>0.05). Conclusions:Throwing live grenade, whether recruits or experienced NCOs have psychological and physiological stress changes in different degrees. This may increase training accidents. Therefore, in order to reduce the possible training accidents in normal time, it is necessary to take various psychological interventions on soldiers especially the recruits to improve the soldiers’anti-stress ability. CORRESPONDING AUTHOR: W. Lu, The 256th Hospital of PLA, Zhengding, China, [email protected]

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

Feasibility and clinical outcomes of mindfulness-based interventions for chronic ill patients S391 FEASIBILITY AND CLINICAL OUTCOMES OF MINDFULNESSBASED INTERVENTIONS FOR CHRONIC ILL PATIENTS M.J. Schroevers1, M. Schellekens2, K.A. Tovote3, I. Nyklicek4, M. Davis5 1 University of Groningen, University Medical Center Groningen, GRONINGEN, Nederland 2 Radboud University, Nijmegen Medical Center, Psychiatry, NIJMEGEN, Nederland 3 University of Groningen, University Medical Center, Dep of Health Psychology, GRONINGEN, Nederland 4 Tilburg University, Department of Medical and Clinical Psychology, TILBURG, Nederland 5 Arizona State University, Department of Psychology, TEMPE, United States of America Mindfulness-based group interventions such as Mindfulness-Based Stress Reduction (MBSR) and Mindfulness-Based Cognitive Therapy (MBCT) are increasingly being used and found to be effective in both healthy and clinical populations. Key elements of these types of interventions include doing meditation and yoga exercises and group sharing. The aim of this symposium is to present data of recent and ongoing research on the feasibility and effectiveness of mindfulness-based interventions and predictors of clinical improvement. First, results regarding the feasibility of MBSR in patients with lung cancer and their partners are presented. Second, results of a large RCT in depressed diabetes patients are presented, in which the effectiveness of a novel type of intervention based on individual MBCT was compared to individual Cognitive Behavioral Therapy (CBT) and waitlist control group. Third, results of a large trial on group MBCT in diabetes patients are described, particularly focusing on moderators of effectiveness, thus examining for whom mindfulness-based interventions are most likely to work. Finally, results of a RCT in patients with chronic pain due to fibromyalgia are presented, comparing the efficacy of a mindfulness-based treatment for emotion regulation, cognitivebehavioral treatment targeting pain management, and pain education, including a description of potential moderators of treatment effects. CORRESPONDING AUTHOR: Dr. M.J. Schroevers, University of Groningen, University Medical Center Groningen, Nederland, [email protected] S392 FEASIBILITY AND EFFECTIVENESS OF MINDFULNESS-BASED STRESS REDUCTION IN LUNG CANCER PATIENTS AND PARTNERS: MIXED-METHOD PILOT M.P.J. Schellekens, D.G.M. Van den Hurk, J. Molema, J.B. Prins, A.E.M. Speckens, M.A. Van der Drift Radboud University Nijmegen Medical Centre, NIJMEGEN, Nederland Background.In the last ten years MBSR has proven to be effective in reducing anxiety and depressive symptoms in cancer patients. However, the generalization of these results is limited because most participants of these studies were female patients with breast cancer. By means of this mixed-method pilot study, we examined whether MBSR might be a feasible and effective intervention for lung cancer patients and their partners. Method. Nineteen lung cancer patients and 16 partners participated in the MBSR training. Assessments took place at baseline, post intervention and at three-month follow-up. Measurements included questionnaires on psychological distress, quality of life and caregiver burden. After the training six patients and five partners were interviewed about how they experienced the training.

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

Results. Sixteen patients and fourteen partners completed the training. Participants who dropped out did not differ from completers. Interviews showed participants felt positive about the intervention and enjoyed participating with their partner and in a peer group. Among patients and partners, changes in psychological distress after MBSR were in the direction of improvement, but did not achieve statistical significance. Among partners, caregiver burden decreased significantly after following MBSR at post and follow-up measurement. Interviews showed MBSR seems to instigate a process of change in patients and partners. Conclusions. The MBSR training seems to be a feasible intervention for lung cancer patients and their partners. However, it remains unclear whether MBSR is also effective in reducing their psychological distress. Therefore a randomized controlled trial is needed to examine the effectiveness of the intervention. CORRESPONDING AUTHOR: M.P.J. Schellekens, Radboud University Nijmegen Medical Centre, Nederland, melanie.schellekens@ radboudumc.nl S393 INDIVIDUAL MINDFULNESS-BASED COGNITIVE THERAPY AND COGNITIVE BEHAVIOR THERAPY FOR DEPRESSION IN PATIENTS WITH DIABETES K.A. Tovote1, J. Fleer1, E. Snippe1, T.P. Links1, P.M.G. Emmelkamp2, R. Sanderman1, M.J. Schroevers1 1 University of Groningen, University Medical Center Groningen, GRONINGEN, Nederland 2 University of Amsterdam, AMSTERDAM, Nederland INTRODUCTION: Depression is a common comorbidity of diabetes, negatively influencing patients’ physical and mental functioning. Mindfulness-Based Cognitive Therapy (MBCT) and Cognitive Behavior Therapy (CBT) have been shown to improve psychological outcomes in various populations. The primary objective of this randomized controlled trial was to compare individual MBCT and individual CBT to a waiting list control condition in terms of their effectiveness in reducing the severity of depressive symptoms. The secondary objective was to examine associations between patients’ outcome expectations, homework compliance, and depressive symptom reduction. METHODS: Ninety-four diabetes patients with depressive symptoms (i.e., BDI-II ≥14) were randomized to an 8-week I-MBCT course (n=31), 8-week I-CBT course (n=32), or waiting list (n=31). Primary outcome measure was severity of depressive symptoms (BDI-II). Patients’ outcome expectations and weekly homework practice were also assessed. RESULTS: Participants receiving MBCT and CBT reported significantly greater reductions in depressive symptoms compared to patients in the waiting list condition (respectively F (1, 60)= 9.7, p = .004 and F (1, 61)= 15.6, p < .001). In both interventions, patients’ expectations were found to predict greater depressive symptom reduction, whereas homework compliance did not. CONCLUSION: Both individual MBCT and CBT were found to be effective interventions in reducing depressive symptoms. The outcomes of CBT and MBCT may be even more optimized by increasing patients’ outcome expectations. CORRESPONDING AUTHOR: K.A. Tovote, University of Groningen, University Medical Center Groningen, Nederland, [email protected] S394 WHO BENEFITS MOST FROM MINDFULNESS-BASED COGNITIVE THERAPY FOR PEOPLE WITH DIABETES? THE DIAMIND TRIAL S.J. van Son1, I. Nyklícek1, V.J.M. Pop1, B. Bravenboer2, A.G. Lieverse3, F. Pouwer1 1 Tilburg University, TILBURG, The Netherlands 2 Sint-Maria Regional Hospital, HALLE, Belgium 3 Máxima Medical Center, EINDHOVEN, Nederland

S119

Introduction: Research has shown the effectiveness of mindfulness-based interventions in various populations. However, little is known regarding potential moderators predicting differential treatment efficacy. Therefore, the main question of the current study was: which characteristics (demographic, personality, and baseline level of mindfulness) moderate the effectiveness of Mindfulness-Based Cognitive Therapy (MBCT)? Methods: The sample consisted ofoutpatients with diabetes (type 1 or type 2) and co-morbid emotional distress, who participated in Diabetes and Mindfulness (DiaMind) randomized controlled trial and completed the group MBCT intervention (N=67). Primary outcomes were symptoms of anxiety and depressive symptoms. Potential moderators included demographic (i.e., sex, age, education), personality (i.e., negative affectivity, social inhibition), and mindfulness characteristics. Results: Mixed models moderator analyses showed that negative affectivity appeared to be a significant moderator of the effect of MBCT on levels of anxiety (p=0.002, Cohen’s d=0.88-1.27) and depressive symptoms (p<0.001, Cohen’s d=0.55-1.27), reflecting that MBCT was more effective in reducing symptoms in individuals with a high level of negative affectivity than individuals with a low level of negative affectivity. For depressive symptoms, this effect was independent of baseline symptom level. There were no moderator effects of demographic factors, social inhibition, and baseline mindfulness. Conclusion: The results regarding negative affectivity are in the opposite direction to some previous results on efficacy of cognitive behavioral therapy. A direct comparison with cognitive behavioral therapy is needed to determine if level of negative affectivity mayhelp clinical decision-making as to which therapy may be most suitable for which patient. CORRESPONDING AUTHOR: S.J. van Son, Tilburg University, The Netherlands, [email protected] S395 MINDFULNESS, COGNITIVE-BEHAVIORAL, AND EDUCATION INTERVENTIONS FOR CHRONIC FIBROMYALGIA PAIN: RESULTS OF A RANDOMIZED CLINICAL TRIAL M. Davis1, A.J. Zautra1, E.W. Yeung1, L.D. Wolf1, S. Stark1, K. Thummala1, P. Nicassio2 1 Arizona State University, TEMPE, United States of America 2 University of California, LOS ANGELES, United States of America Introduction: This study compared the impact of seven sessions of groupdelivered cognitive-behavioral therapy for pain (CBT-P), mindfulacceptance (M) treatment, and an arthritis education condition (E) on symptoms and functional health among adults with fibromyalgia (FM). Method: Adult FM patients (N=170) who were mid-aged (M age = 52 years, SD = 10 years) and primarily Caucasian (85%) were randomized to CBT-P, M, or E. CBT-P targeted pain coping skills; M targeted awareness and acceptance of current experience to enhance coping with a range of aversive experiences; and E provided health information. At pre- and post-treatment, and 6- and 12-month follow-up, participants completed questionnaires regarding coping, physical symptoms, and daily physical and social functioning. Results: Multilevel models compared groups in the magnitude of change in outcomes across the 4 time points. Both CBT-P and M produced greater changes in pain coping efficacy, physical symptoms and functioning, and social functioning than did the E group. For physical functioning outcomes, CBT-P produced larger effects than did M. All effects of the interventions were largely sustained at 12-month follow-up. Conclusions: For individuals with FM, CBT-P and M yield broad and lasting improvements in pain coping, symptoms, and physical and social functioning relative to E. CBT-P may be especially effective at targeting physical functioning. CORRESPONDING AUTHOR: Professor M. Davis, Arizona State University, United States of America, [email protected]

S120

A capability perspective on health, work and health and sustainable employability S396 A CAPABILITY PERSPECTIVE ON HEALTH, WORK AND HEALTH AND SUSTAINABLE EMPLOYABILITY J.J.L. van der Klink1, S. Venkatapuram2, P.W. Saleeby3, F.I. Abma1 1 UMCG, GRONINGEN, Nederland 2 King's College, LONDON, United Kingdom 3 Southern Illinois University, CARBONDALE, United States of America The capability concept, as formulated by Nobel Prize Laureate Amartya Sen, brings the individual’s capacity or ability (what a person can do or is able to) in relation to the possibilities that the environment offers (to which an individual is enabled). The CA emphasises the real opportunities we have to do what we want to do and to live the life that we want to live. This can be regarded as very relevant for health, disability management and sustainable employability. At present employability is an important and dominant concept in the field of work and health in which it is the responsibility of the employee to be employable. In the capability approach (CA) there is also an environmental (societal or organisational) responsibility to be inclusive for people with a disability The CA has been applied to and is operationalised in many fields a.o. the field of health and disability. In the symposium we display the applicability of the capability approach with respect to health, work and health and sustainable employability in four contributions. 1) a capability perspective on health (S. Venkatapuram); 2) a capability perspective on disability (P. Saleeby) 3) a capability perspective on sustainable employability (J. van der Klink) and 4) Development of a capability set for sustainable employability (F. Abma) pability set for sustainable employability. CORRESPONDING AUTHOR: Prof. dr. J.J.L. van der Klink, UMCG, Nederland, [email protected] S397 THE CAPABILITIES APPROACH AND HEALTH S. Sridhar Venkatapuram King's College London, LONDON, United Kingdom The capabilities approach is a conceptual framework and ethical theory developed by the philosopher and economist Amartya Sen. The approach is significantly influencing a variety of academic disciplines and professions such as economics, social statistics, public policy, development programmes, and increasingly health related professions. This paper will present some of the central ideas of the capabilities approach, and how it can be applied to human health. In particular, an argument will be presented for how health should be conceptualized as an inter-related cluster of basic capabilities to be and do certain things that constitute a life with equal human dignity. Such a view of health is contrasted with purely biological approaches or utility related approaches such as QALYs in identifying the aims, implementation, and evaluation of health and social care programmes. CORRESPONDING AUTHOR: Dr S. Sridhar Venkatapuram, King's College London, United Kingdom, [email protected]

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

to better understand the impact of impairment on individual functioning. Rather than focusing solely on individual capacity measured in clinical settings, capabilities represent real opportunities - thec ombination of individual ability, environment, and life situation. Method: This study operationalized the Welch Saleeby schematic representation of the capability approach using the WHO’s International Classification of Functioning, Disability and Health. The ICF provides a comprehensive classification of body structures and functions, activities and participation, and the environment. Students and clinicians across disciplines used the ICF to code various case studies that addressed functioning in major life areas (e.g. education andemployment) both pre and post intervention. Result: Cases provided information articulating individual goals (rehabilitation, treatment) indicating that coders were able to determine both potential (capabilities) and actualized (functioning) outcomes. Conclusion: Findings indicate the utility of the ICF for addressing interventions and outcomes as well as emphasizing the importance of moving beyond functioning to address capabilities as a more effective measure of disability. CORRESPONDING AUTHOR: Professor P.W. Welch Saleeby, Southern Illinois University Carbondale, United States of America, [email protected] S399 A C A PA B I L I T Y P E R S P E C T I V E O N S U S TA I N A B L E EMPLOYABILITY J.J.L. van der Klink1, W.B. Schaufeli2, M. Cuijpers3, U. Bültmann1, S. Brouwer1, G.J. van der Wilt4, F.I. Abma1, I. Arends1 1 UMCG, GRONINGEN, Nederland 2 University of Utrect, UTRECHT, Nederland 3 University of Maastricht, MAASTRICHT, Nederland 4 Radboud University, NIJMEGEN, Nederland Introduction: Due to demographic and social developments sustainable employability of workers is an important issue. Work and the valuation of work changed substantially in recent decades. Our conceptualisation of health changed too. The capability concept of Amartya Sen is helpful to understand the complexity of sustainable employability in relation with the new concepts of work and health. Methods: Based on literature and expert opinions, a definition and a model of sustainable employability were developed. Based on the model a questionnaire was constructed that was completed by more than 1000 employees. The hypothesised centrality of the capability items was tested with a.o. regression analyses. Results: The capability items are associated with antecedents and consequents in the model and add unique variance in explaining outcomes when controlling for work function (a behavioral measure) and engagement(an affective-cognitive measure). Conclusions: The hypothesised centrality and unique contribution of the capability concept was confirmed in the analyses. CORRESPONDING AUTHOR: Prof. dr. J.J.L. van der Klink, UMCG, Nederland, [email protected]

S398 A CAPABILITY PERSPECTIVE ON DISABILITY P.W. Welch Saleeby Southern Illinois University Carbondale, CARBONDALE, United States of America

S400 DEVELOPMENT OF A CAPABILITY SET FOR SUSTAINABLE EMPLOYBILITY F.I. Abma1, U. Bültmann1, S. Brouwer1, I. Arends1, G.J. van der Wilt2, J.J.L. van der Klink1 1 University Medical Center Groningen, GRONINGEN, Nederland 2 Radboud University, NIJMEGEN, Nederland

Introduction: Disability has become increasingly recognized as the dynamic interaction of the individual and his/her environment - specifically, the demands of an impairment, society’s interpretation of it, and the broader societal context of disability. Sen's capability approach enables

Introduction. Studies show that individual capabilities, work-related factors, and healthstatus profoundly impact workers’ abilities to be actively engaged in paid employment and to prolong their meaningful contribution to a productive society. Instruments are needed to identify an

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

individuals’ capability set that allows him/her to achieve valuable work functionings. Therefore, the aim of this study is to develop a capability-set for sustainable employability. Methods. Various methods were used to develop the items. First a literature search was conducted to identify existing capability instruments or items in other areas, proving insight in different ways to operationalize capability items. A second literature search was conducted to identify valued aspects of work. In addition, interviews were conducted to explore what is needed for workers to value their work and what is needed to continue working. Results. The input from the literature and the interviews, together with a group discussion in the consortium led to the content of 7 capability items important for sustainable employability. For all items it is asked if A) people think this aspect is important , B) their work offers them sufficient opportunities to do it, and C) they are able to succeed in realising it. Analyses showed significant relations between the capabilities and the three outcome measures work ability, work productivity and future work participation (i.e. how many years do participants expect to continue working). Different capabilities were more relevant for different outcome measures. Conclusions. Seven capability items were developed to asses a workers capability set for sustainable employability. CORRESPONDING AUTHOR: F.I. Abma, University Medical Center Groningen, Nederland, [email protected] Are findings in behavioral medicine believable? S401 ARE RESEARCH FINDINGS IN BEHAVIORAL MEDICINE BELIEVABLE? J.C. Coyne UMCG, PHILADELPHIA, United States of America Overall symposium abstract ‘Are findings in behavioral medicine believable? (ID 735) John Ioannidis documented that many positive findings in the biomedical literature do not replicate and many apparent “discoveries”turn out to be exaggerated or simply false. Many apparent discoveries are created by a combination of confirmatory bias, flexible rules of design, data analysis and reporting and significance chasing. Behavioral medicine is particularly susceptible because many of the psychological variables and outcomes measured and analyzed are often complex and highly correlated. There is wide acceptance of flexibility in definitions, use of cut-offs, modeling, and statistical handling of the data that can turn initially null findings into publishable, positive findings. Other areas of psychology have accepted the pervasiveness of confirmatory bias and are adopting a greater emphasis on replication and publication of null findings, but these reforms are not yet noticeable in behavioral medicine. This symposium will consist of co-authored presentations combining the behavioral medicine expertise of James Coyne and Gozde Ozakinci with basic statistical and methodological skills of co-authors Nick Brown, Mortize Heene and Daniel Lakens. Brown and Coyne will demonstrate a failure of reanalyses to replicate some high profile findings linking positive psychology to physical health. Ozakinci and Heen will show how misplaced is the typical confidence in failsafe N to assure the lack of publication bias and the generalizability of meta analyses of psychosocial interventions. Lakens and Coyne will apply the concept of p-hacking and show the likelihood that many established findings in behavior medicine will fail in independent attempts to replicate. Coyne will offer an integrative discussion. Discussant: Coyne (University of Groningen, Groningen, The Netherlands) CORRESPONDING AUTHOR: Professor C. Coyne, UMCG. United States of America, [email protected]

S121

S402 PHYSICAL HEALTH OUTCOMES IN POSITIVE PSYCHOLOGY: WEAK EVIDENCE FOR STRONG CLAIMS N.J.L. Brown1, J.C. Coyne2 1 New School of Psychotherapy and Counselling, STRASBOURG, France 2 University of Pennsylvania, PHILADELPHIA, United States of America We critically examined some recent claims from positive psychology research suggesting that positive attitudes and behaviors can have a profound impact, not just on psychological well-being, but also on physical health outcomes. Although not yet replicated, these claims have received considerable attention in the media and popular self-help books. We focused first on claims that the expression of genes associated with immune response are differentially affected by adopting an orientation of eudaimonic versus hedonic striving, such that a preponderance of eudaimonia leads to better health outcomes. We performed factor and regression analyses on the original data from the study, first replicating the authors’ original analyses and then exploring alternative explanations for their findings; we established that the authors’ factor analysis is seriously flawed, and their 'innovative' regression method produces essentially the same results even with randomly-generated data. We next examined claims concerning practicing loving-kindness meditation on physical health, with high-frequency heart rate variability as a surrogate for cardiac vagal tone. Although some of the deficiencies of this study (such as the absence of a statistically significant main effect) were already evident in the published article, we also obtained the original dataset from the authors and discovered further problems, including the use of assignment to the control or treatment group as the (only) dependent variable. We conclude that the authors’ have not established any credible link between positive attitudes and behaviors and physical health outcomes, and their claims in this regard are premature and exaggerated. CORRESPONDING AUTHOR: N.J.L. Brown, New School of Psychotherapy and Counselling, France, [email protected] S403 THE UNSAFE DEPENDENCE OF META-ANALYSES IN BEHAVIORAL MEDICINE ON FAILSAFE N G. Ozakinci1, M. Heene2, J. Coyne3 1 University of St Andrews, ST ANDREWS, SCOTLAND, United Kingdom 2 Ludwig Maximilian University, MUNICH, Germany 3 University of Groningen, GRONINGEN, The Netherlands Meta-analyses of health-behavior change and psychosocial interventions uniformly conclude that they are effective and ready for dissemination. Concerns about publication bias and the reliance on small, underpowered studies of only low to medium quality are frequently waived off with calculations of failsafe N most associated with Robert Rosenthal. Such calculations estimate the many null studies lying in desk drawers that would be needed to unseat the conclusions of the meta-analyses. They consistently exceed the number of available studies many times over. However, outside of psychology, the failsafe N of Rosenthal and others has fallen into disfavor. The Cochrane collaboration and many epidemiologists recommend against its use. The Rosenthal’s (Rosenthal & Rubin, 1978) Fail-Safe-Number (FSN) aimed to estimate the number of unpublished studies in meta-analyses required to bring the meta-analytic mean effect size down to a statistically insignificant level is widely-used in order to determine the stability of meta-analytic results against potential publication bias threats. However, the FSN is invalid because it treats the file drawer as unbiased and almost always severely overestimates the number of unpublished studies (Elsahoff, 1978; Iyengar & Greenhouse, 1988).

S122

CORRESPONDING AUTHOR: Dr G. Ozakinci, University of St Andrews, United Kingdom, [email protected] S404 HOW LIKELY IS IT AN INTERVENTION STUDY WILL REPLICATE? THE PAIN LITERATURE AS AN EXAMPLE D. Lakens1, J.C. Coyne2 1 Eindhoven University of Technology, Eindhoven, Nederland 2 University of Groningen, University Medical Center, Dep of Health Psychology, GRONINGEN, Nederland Background. The goal of science is to learn what is likely to be true. Incentive structures in academia often work against researchers trying to achieve this goal. Low statistical power, flexibility during data analysis with publication bias have yielded a scientific literature in which, some researchers argue, most research findings are false. As a consequence, concerns have arisen regarding replicability of scientific findings. In this presentation, We will give a demonstration of the problem and provide practical recommendations to deal with these uncertainties when performing and evaluating research. Method. We discuss why researchers cannot blindly interpret significant p-values as support for hypotheses, giving some examples of when a significant p-value is stronger support for the null hypothesis than the alternative hypothesis. Then we introduce the ideas behind p-curve analysis, providing some examples of how this technique can be used to evaluate the evidential value of published research. Using p-curve analysis, we can make better predictions about the likelihood an experiment will work, or whether findings can be expected to replicate. Finally, we demonstrate p-curve analysis with studies previously identified for a meta-analysis of psychosocial interventions for pain and show how we can increase the informational value of the studies that were considered. Results. P-curve analyses reveal that demonstrate weaknesses in this particular literature, and explain how publishing research can be avoided that lacks evidential value and is unlikely to replicate. Conclusion. Analyses of other intervention literatures in behavioral medicine would likely yield similar signs of weaknesses. CORRESPONDING AUTHOR: D. Lakens, Eindhoven University of Technology, Nederland, [email protected] O405 PREDICTORS AND OUTCOME OF AVOIDANCE OF ACTIVITIES I N P E R S O N S W I T H E A R LY S Y M P T O M AT I C K N E E OSTEOARTHRITIS J.F.M. Holla1, M. van der Leeden1, D.L. Knol2, L.D. Roorda1, W.K.H.A. Hilberdink3, W.F. Lems2, M.P.M. Steultjens4, J. Dekker2 1 Reade, AMSTERDAM, Nederland 2 VU University Medical Centre, AMSTERDAM, Nederland 3 Allied Healthcare Centre for Rheumatology and Rehabilitation, GRONINGEN, Nederland 4 Glasgow Caledonian University, GLASGOW, United Kingdom Objective. It has been hypothesized (i) that pain and low vitality lead to an increase in avoidance of activities in persons with early symptomatic knee osteoarthritis (OA), and (ii) that avoidance of activities leads to an increase in activity limitations. The present study aimed to evaluate these hypotheses in a cross-sectional and longitudinal design. Methods. Baseline, 2-year and 5-year follow-up data of 828 participants from the Cohort Hip and Cohort Knee (CHECK) study with early symptomatic knee OA were used. Knee pain, vitality, avoidance of activities and activity limitations were measured with reliable and validated instruments. Regression and autoregressive generalized estimating equations (GEE) models analyzed the cross-sectional and longitudinal associations between pain, vitality, avoidance of activities and activity limitations. The models were adjusted for age, gender, education level, body-mass index, comorbidity, and radiographic severity.

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

Results. (i) In cross-sectional analyses, greater knee pain and lower vitality were associated with higher levels of avoidance of activities. In longitudinal analyses, pain and vitality predicted a subsequent increase in avoidance of activities. (ii) In cross-sectional analyses, a higher level of avoidance of activities was associated with greater activity limitations. In longitudinal analyses, this relationship was marginally significant. Conclusions. Knee pain and low vitality lead to a subsequent increase in avoidance of activities, already at an early stage of knee OA. Avoidance of activities is related to activity limitations: alternative measurement instruments and a longer follow-up may be required to establish the longitudinal relationship. CORRESPONDING AUTHOR: J.F.M. Holla, Reade, Nederland, [email protected] O406 CAREGIVERS' SUPPORTIVE RESPONSES: DO THEY ALWAYS REINFORCE PATIENTS' PAIN BEHAVIORS? S. Mohammadi1, Mohsen Dehghani2, Robbert Sanderman1, Mariet Hagedoorn1 1 University Medical Center Groningen, GRONINGEN, The Netherlands 2 Family Research Institute, Shahid Beheshti University, TEHRAN, Iran Aim: The literature seems to suggest that caregivers’ supportive responses (i.e. solicitous and distracting responses) reinforce patients’ pain behaviors. However, patients with supportive caregivers know that help is available when needed. Therefore, it is hypothesized that these patients express more pain behaviors to obtain support when they experience more pain. Patients whose caregivers are unsupportive have learned that support is not forthcoming when needed, and therefore, their levels of pain behavior are not expected to be related to the intensity of their pain. Subjects and Methods: The sample consisted of 154 chronic pain patients and their family caregivers. Patients completed validated questionnaires about pain intensity, pain behaviors, and their perception of caregivers’ responses. Caregivers indicated their responses to the patients’ pain. Hierarchical regression analyses examined the interaction between pain intensity and caregivers’ responses on pain behavior. Results: Pain intensity was associated with pain behaviors. This association was only significant if patients reported high levels of caregivers’ solicitous (B=.227, p<.001) and distracting responses (B=.231, p<.001), and if caregivers reported high levels of solicitous (B=.231, p<.001) and distracting responses (B=.231, p<.001). Conclusion: Our findings are similar for both patients’ and caregivers’ perceptions of supportive responses. Living with a supportive caregiver does not always result in higher levels of pain behavior. That is, patients with supportive caregivers only show pain behavior when they need support. If they experience relatively low levels of pain, they show little pain behavior (even less than patients who do not have a supportive partner). CORRESPONDING AUTHOR: S. Mohammadi, University Medical Center Groningen, The Netherlands, [email protected] O407 IMPACT OF CLINICAL OSTEOARTHRITIS ON THE SOCIAL PARTICIPATION OF OLDER ADULTS IN EUROPE S. van der Pas1, M.V. Castell2, C. Cooper3, M. Denkinger4, E.M. Dennison3, M.H. Edwards3, F. Herbolsheimer4, S. Maggi5, Á. Otero2, N.L. Pedersen6, R. Peter4, S Zambon5, N.M. Van Schoor1, D.J.H. Deeg1 1 VU University Medical Centre / LASA, AMSTERDAM, Nederland 2 Department of Preventive Medicine and Public Health, Unit of Primary Care and Fa, MADRID, Spain 3 MRC Lifecourse Epidemiology Unit, University of Southampton, Southampton General, SOUTHAMPTON, United Kingdom 4 Bethesda Geriatric Clinic, ULM, Germany 5 Department of Medicine, University of Padova, PADOVA, Italy

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

S123

6 Department of Clinical Science and Education, Karolinska Institutet, STOCKHOLM, Sweden

CORRESPONDING AUTHOR: Dr M. Stevens. University Medical Center Groningen, Nederland, [email protected]

Introduction: Older people participate in social activities within the context of social relationships which in turn may enhance their quality of life. Pain and disability caused by OA may have a negative impact on social functioning. Little is known about the actual social participation of older adults with OA at different sites. The objective of the current study is to examine the relation between clinical OA at different sites and social participation in the general population. Methods: The European Project on OSteoArthritis (EPOSA) is a sixcohort study (Germany, Italy, the Netherlands, Spain, Sweden, and the United Kingdom) on community-dwelling persons aged 65-85 years. Clinical OA was defined according to the ACR criteria. Community and informal social participation was assessed using the Maastricht Social Participation Profile (MSPP) and the Lubben’s Social Network Scale (LSNS). The MSPP measures frequency and diversity of community social participation. The LSNS assesses the frequency of contact with family and friends. Results: Of the 2816 participants, 19% had knee OA, 6% had hip OA and 17% had hand OA. After controlling for age, gender, partner status and chronic conditions, regression analyses revealed that knee and hand OA were associated with less frequent community social participation. Only knee OA was associated with less diverse community social participation. Knee and hand OA were associated with less informal social participation with family. Only hip OA was associated with less informal social participation with friends. Conclusion: Clinical OA at different sites was related to less community and informal social participation.

O409 THE RELATIONSHIP OF GOAL MANAGEMENT WITH FATIGUE OVER ONE YEAR IN PATIENTS WITH POLYARTHRITIS C. Bode, E. Taal, R. Arends, M. Van de Laar University of Twente, ENSCHEDE, Nederland

CORRESPONDING AUTHOR: Dr S. van der Pas, VU University Medical Centre / LASA, Nederland, [email protected] O408 THE EFFECT OF PREOPERATIVE PHYSICAL ACTIVITY ON POSTOPERATIVE RECOVERY IN PATIENTS AFTER TOTAL HIP/ KNEE ARTHROPLASTY M. Stevens, S. Poortinga, S.K. Bulstra, I. Van den Akker-Scheek University Medical Center Groningen, GRONINGEN, Nederland Introduction. There are indications that the level of physical activity preoperatively affects the recovery after total hip and knee arthroplasty (THA/TKA). Based on these indications, it is hypothesized that preoperative exercise therapy (prevalidation) improves recovery. However, literature is inconclusive about the effect of prevalidation. Aim is to analyze the basic assumption of prevalidation: a relationship between preoperative physical activity level and degree of recovery one year after THA/TKA. Method. From 2006-2012 658 patients underwent a primary THA or TKA at University Medical Center Groningen. Preoperatively and one year postoperatively patients completed the SQUASH-questionnaire to measure physical activity level and the WOMAC-questionnaire to measure degree of recovery. Missing data were imputed. Data were analyzed using linear regression analysis. Results. There was no significant relationship between amount of physical activity preoperatively and degree of recovery one year postoperatively (β = 0.040, p = 0.343). There was also no significant relationship between amount of leisure time physical activity preoperatively and degree of recovery one year postoperatively (β = 0.0730, p = 0.110). Also complying with the guidelines for health-enhancing physical activity preoperatively did not lead to a higher degree of recovery (β = 0.003, p = 0.061). The same results were found after adjusting for type of surgery, sex, BMI, ASA-class and age. Conclusions. No relationship between physical activity level preoperatively and the degree of recovery one year after THA/ TKA could be determined. It can be concluded that the results do not support the assumption underlying prevalidation programs.

Background: Many patients with polyarthritis experience severe fatigue. How to cope with severe fatigue in polyarthritis is unclear. For related symptoms as pain, depressive feelings and feelings of anxiety some studies have pointed to the beneficial effects of goal management. Aim of this study was to describe the relationship of four goal management strategies (goal maintenance, goal adjustment, goal disengagement, goal re-engagement) and fatigue (severity, impact and coping) in a longitudinal perspective. Methods: 331 patients (61% women, mean age=61 years) participated in a questionnaire study with measurements at baseline, after six and twelve months. Goal management was measured by TenFlex and GAS, fatigue was assessed by 3 visual analogue scales, asking for the severity and impact of fatigue and coping with fatigue. Findings: The studied aspects of fatigue and the use of goal management strategies were stable over the course of one year. In regression analyses higher use of goal adjustment was significantly related to less severe fatigue, less impact of fatigue and better coping with fatigue. Unexpectedly, higher goal reengagement was related to more severe fatigue. Goal maintenance and goal disengagement did not contribute to the explanation of fatigue. Discussion: The results indicated stability of goal management and levels of fatigue over one year. For patients with elevated levels of fatigue it might be beneficial to change the way they deal with fatigue by adjusting personal goals. The role of goal re-engagement in relation to fatigue should be further explored since previous studies with related symptoms showed opposite relationships. CORRESPONDING AUTHOR: Dr C. Bode, University of Twente, Nederland, [email protected] O410 THE INFLUENCE OF WEATHER CONDITIONS ON JOINT PAIN IN OLDER PEOPLE WITH OSTEOARTHRITIS IN EUROPE E.J. Timmermans VU University Medical Centre / LASA, AMSTERDAM, Nederland Introduction: It was aimed to examine whether daily weather conditions, ambient weather conditions (average weather conditions over 3-day period before a pain report) and changes in weather influence joint pain in older people with osteoarthritis (OA) in six European countries. Methods: Data from the population-based European Project on OSteoArthritis (EPOSA) were used. ACR classification criteria were used to determine OA in older people (65-85 years). After the baseline-interview, at six months, and after the 12-18 months follow-up interview, joint pain was assessed by using two-week pain calendars. Daily values for temperature, precipitation, atmospheric pressure, relative humidity and wind speed were obtained from local weather stations. Multilevel regression modelling was used to examine the associations between the weather parameters and joint pain, adjusted for sex, age, country, medication use and season. Results: The study included 811 participants with OA in knee, hand and/ or hip. Their mean age was 73.6 years (SD=5.5). After adjustment, there were significant associations of joint pain with daily average humidity (B=0.01; p<0.001) and ambient humidity (B=0.01; p<0.001), but not with other daily or ambient conditions. Changes in weather parameters between two consecutive days were not significantly associated with reported joint pain.

S124

Conclusions: Humidity is independently associated with joint pain severity in older people with OA. These results support the thought that humidity affects expansions and contractions of various tissues in joints, which may result in pain. CORRESPONDING AUTHOR: E.J. Timmermans, VU University Medical Centre / LASA, Nederland, [email protected] O411 PAT I E N T S ' P R E F E R E N C E S F O R P O S T- T R E AT M E N T BREAST CANCER FOLLOW-UP IN PRIMARY CARE VERSUS SECONDARY CARE C. Roorda-Lukkien, G.H. de Bock, C. Scholing, K. van der Meer, M.Y. Berger, M. de Fouw, A.J. Berendsen UMCG, GRONINGEN, The Netherlands IntroductionMore qualitative research concerning patients’ preferences in the issue of primary care-based breast cancer follow-up is needed.MethodsA cross-sectional design was employed, involving semi-structured interviews with 70 female patients with a history of early-stage breast cancer. Using descriptive content analysis, interview transcripts were analysed independently and thematically by two researchers.ResultsPatients expressed the strongest preference for annual visits (31/68), a schedule with a decreasing frequency over time (27/68), and follow-up > 10 years, including lifelong follow-up (21/64). The majority (56/61) preferred to receive follow-up care from the same care provider over time, for reasons related to a personal doctor-patient relationship and the physician’s knowledge of the patient’s history. About 75% (43/56) preferred specialist follow-up to other follow-up models. However, primary care-based follow-up would be accepted by 57% (39/ 68) provided that there is good communication between GPs and specialists, and sufficient knowledge among GPs about follow-up. Perceived benefits of primary care-based follow-up referred to the personal nature of the GP-patient relationship and the easy access to primary care. Perceived barriers included: limited oncology knowledge and skills, time available, and motivation among GPs to provide follow-up care, and patients' confidence with present specialist follow-up.ConclusionsMore than half of the patients were open to primary care-based follow-up. Patients’ confidence with this follow-up model may increase by using survivorship care plans to facilitate communication across the primary/secondary interface and with patients. Training GPs to improve their oncology knowledge and skills might also increase patients’ confidence. CORRESPONDING AUTHOR: Ir C. Roorda-Lukkien, UMCG, The Netherlands, [email protected] O412 GOAL ADJUSTMENT STRATEGIES IN CANCER PATIENTS M. Janse1, A.V. Ranchor1, M. A. G. Sprangers2, J. Fleer1 1 University Medical Center Groningen, GRONINGEN, Nederland 2 Academical Medical Center, AMSTERDAM, Nederland Introduction: Cancer can lead people to reconsider their personal goals, either because the illness has made their goals unattainable, or it has led to a changed outlook on life. How patients actually adjust their goals has not yet been investigated, whereas knowing patients’ goal adjustments strategies (GAS) and knowing which strategies are beneficial for well-being, may provide relevant information for the development of interventions to help patients successfully adjust to cancer. Therefore, the current study aims to investigate GAS in colon cancer patients. Methods: Patients listed their goals 1 and 6 months post-diagnosis. All goals were scored on goal characteristics (importance, attainability and effort by patients; life domain and level of abstraction by two authors). A new method was developed to measure 6 theory-based GAS using these characteristics.

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

Results: Patients (n=201, 37.8% female, mean age 66) most often used Give up effort but remain committed (put goals on hold) (54.7%), then Scale back goals in the same life domain (30.8%), Scale up goals in the same life domain (18.4%) and Reprioritize (10.4%). Continue to pursue disturbed goals (1.5%) was hardly used, and Give up goal commitment without turning to a new goal not at all. Twenty-four percent of patients did not use an adjustment strategy, while 28.4% used a combination of more strategies. Conclusion: Most cancer patients adjust goals during the first 6 months after cancer diagnosis, mostly putting their goals on hold. Our newly developed method can next be used to study the influence of GAS on well-being. CORRESPONDING AUTHOR: M. Janse, University Medical Center Groningen, Nederland, [email protected] O413 BODY IMAGE IN CANCER SURVIVORS: A SYSTEMATIC REVIEW AND META-ANALYSIS V. Lehmann, M.A. Tuinman, M. Hagedoorn University Medical Center Groningen, GRONINGEN, Nederland Objective: We reviewed the impact of cancer on body image in cancer survivors. In order to draw conclusions as to whether body image might be altered due to cancer, we included case-control studies only. Methods: Medline, Cinahl, Embase, and PsychInfo were searched for studies concerning body image and cancer (including related terms). After duplicate extraction, 1932 hits were retrieved and screened for eligibility. Studies were included if they were published in peerreviewed journals, written in English, included cancer survivors off treatment, measured body image, and included a healthy control group. Results: Twenty-five studies were identified. The same number of studies reported a more negative body image in survivors versus no differences between survivors and controls (n=9 each). Fifteen studies could be included in meta-analyses. Due to the reporting of subgroups and subscales, we conducted one meta-analysis including the subgroups and subscales with the smallest differences, and one meta-analysis including the largest differences between survivors and controls. This has been done in order to give a ‘lower’ and ‘upper’ estimate of the extent of potential differences between survivors and controls. The overall standardized mean differences were .22 [0.02; 0.42] and .30 [0.02; 0.58], favoring healthy controls. Sub-meta-analyses suggested that the length of survival may be crucial, as short-term survivors differed from healthy controls (0.59 [0.01; 1.17]), while longer term survivors did not. Conclusion: Body image seems less favorable in cancer survivors, but differences are small. It also turned out more impaired during short-term survival, while effects seem to fade over time. CORRESPONDING AUTHOR: V. Lehmann, University Medical Center Groningen, Nederland, [email protected] O414 THE RELATIONSHIP BETWEEN IMPLICIT OPTIMISM AND EXPLICIT 'POSITIVE' PSYCHOLOGICAL FACTORS, AND SURVIVAL IN LUNG CANCER Y.G. Gidron1, F.M. Mols2, I.N. Nyklícek2, A.V. Vingerhoets2, B.B. Biesma2, A.V. Visser2 1 Free University of Brussels (VUB), BRUSSELS, Belgium 2 University if Tilburg, TILBURG, The Netherlands Introduction: Though a meta-analysis of over 150 studies showed that psychosocial factors predict cancer prognosis independent of confounders, the findings are inconsistent and remain controversial. One possible reason for past inconsistent results is the use of self-reported measures, which are prone to multiple biases. Implicit (indirect) measures partly bypass such biases and often predict objective outcomes. This

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

study evaluated the prognostic value of implicit and explicit “positive”psychological factors and depression in relation to 12-month survival in non-small cell lung cancer (NSCLC) patients. Methods: Seventy eight NSCLC patients underwent assessment for mindfulness, benefit finding and depression, assessed by self-reported questionnaires, and optimism, assessed by an implicit word-search test. The latter included optimistic and pessimistic words that patients had to locate in one minute. Confounders (stage, age, etc.) were considered as well. Results: Only implicit optimism significantly predicted 12-month survival, independent of confounders such as tumor stage and time from diagnosis (Hazard Ratio =1.73, 95% confidence interval: 1.04-2.87). Other “positive”psychological factors and depression did not predict survival. Conclusions: Future studies need to test the prognostic role of psychosocial factors using implicit tests and to examine whether increasing optimism by indirect manners may improve prognosis. CORRESPONDING AUTHOR: Prof. Y.G. Gidron, Free University of Brussels (VUB), Belgium, [email protected] O415 TRAJECTORIES OF DEPRESSIVE SYMPTOMS IN COUPLES AFFECTED BY BREAST CANCER: A NATIONWIDE PROSPECTIVE COHORT STUDY N. Rottmann1, D.G. Hansen1, M. Hagedoorn2, P.V. Larsen1, A. Nicolaisen1, C. Johansen3 1 University of Southern Denmark, ODENSE, Denmark 2 University Medical Center Groningen, University of Groningen, GRONINGEN, The Netherlands 3 Danish Cancer Society Research Center, COPENHAGEN, Denmark Introduction: Studies have indicated that women affected by breast cancer have different trajectories of distress over time. The same is likely to be true for their partners. However, to our knowledge no study has examined trajectories for both patients and their partners. The purpose of the present study is (a) to identify distinct trajectories of depressive symptoms for women and their partners during treatment, re-entry and short-term survivorship phases and (b) to explore sociodemographic, diseaserelated and relationship characteristics as predictors of these trajectories. Methods: A nationwide cohort of women diagnosed with breast cancer and their cohabiting male partners was established in Denmark (inclusion: 07/2011-08/2012). The couples received questionnaires on individual and dyadic adjustment at baseline (≤4 months following surgery) and 5 and 12 months later. Depressive symptoms were assessed with the Center for Epidemiologic Studies-Depression Scale (CES-D). Sociodemographic and disease-related characteristics were retrieved through registries. Results: Trajectory analyses were based on 546 women and 508 partners. A total of 84% of women followed a low slightly decreasing trajectory and 16% followed a high stable trajectory. Similarly, 87% of partners followed a low slightly decreasing trajectory, whereas 13% followed a high stable trajectory. Predictor analyses are ongoing. Conclusion: Most women and partners reported little distress, yet a minority experienced high levels of depressive symptoms throughout the study period. Knowledge about predictors of trajectory membership can help us to identify couples at risk for a high stable trajectory of depressive symptoms. CORRESPONDING AUTHOR: N. Rottmann, University of Southern Denmark, Denmark, [email protected] O416 DOES SPIRITUALITY INFLUENCE THE WELL-BEING OF CANCER PATIENTS? A. Visser1, A. Garssen1, A.M. Vingerhoets2, N.F. Uwland1 1 Helen Dowling Institute, BILTHOVEN, Nederland 2 Tilburg University, TILBURG, Nederland

S125

Introduction: It has been argued for several decades that spirituality is an important resource for cancer patients in the adjustment to their illness. However, until recently such statements were based on cross-sectional research that defined spirituality in religious terms. Such results do not provide any information about a possibly causal pathway between spirituality and well-being and cannot be generalized to the Dutch, secularly spiritual, population. Therefore, we have investigated in a longitudinal study whether spirituality, defined in non-religious terms, predicts the well-being of cancer patients. Method: Regression analyses with M-estimators were performed on the data of 383 participants. It was explored whether the level of spirituality at diagnosis and around the end of medical treatment (6 months after diagnosis) predicted the level of well-being and distress around the end of medical treatment and at 6 months follow-up. Spirituality was measured with the Spiritual Attitude and Involvement List. Well-being was measured with the subscale Joy in Life of the Health and Disease Inventories. Distress was determined with the Hospital Anxiety and Depression Scale. Results: Spirituality was not predictive of well-being or distress, controlling for the use of pain medication, anti-depressants and anti-anxiety medications, receiving support from a social worker and receiving support from a psychologist, pain, fatigue, experienced life threat of the cancer, and the experience of life events other than cancer during the study period. Conclusion: Although research has shown that spirituality is related to psychological well-being, the present study indicates that this relationship might not be causal in nature. CORRESPONDING AUTHOR: A. Visser, Helen Dowling Institute, Nederland, [email protected] O417 DEVELOPMENT OF A SCALE MEASURING UNSUPPORTIVE SOCIAL INTERACTIONS AMONG DIABETES PATIENTS O. Baron-Epel University of Haifa, HAIFA, Israel Introduction: Social support and its positive impact on health have been extensively studied in the past. However, social negativity may be a source of negative outcomes. In this study a scale measuring unsupportive social interactions was developed. The scale includes two constructs: interference and insensitivity. Methods: A list of 22 items depicting negative aspects of social networks related to diabetes was developed. A telephone survey was administered to 731 Israelis adult diabetics insured by Maccabi healthcare services. In addition to the list of items developed, the questionnaire included questions about diabetes management, social networks, social support, socioeconomic variables and health. The reliability, face validity, content and construct validity of the scale were tested. Correlations between the sub-scales, the social network variables, social support and socioeconomic variables were measured.Results: Explanatory factor analysis yielded a 15 item scale with two factors, consistent with the theoretical factors interference and insensitivity. Cronbach's alpha was 0.84, indicating internal consistency. The scale's scores were associated with education, religiosity, employment and marital status. Gender and income were not correlated with the scale. In addition, the scale was negatively correlated with social support and some aspects of social networks. The scale was correlated also with the patient's evaluation of his/her treatment success and self reported health.Conclusions: The diabetic non-supportive social interactions scale is a valid and reliable measure including two constructs of social negativity- interference and insensitivity. CORRESPONDING AUTHOR: Prof O. Baron-Epel, University of Haifa, Israel, [email protected]

S126

O418 TRENDS IN THE PREVALENCE OF DIABETES AND ASSOCIATED OBESITY IN PENNSYLVANIA ADULTS, 1995 - 2010 O.G.D. Dominic1, E.J.L. Eugene J. Lengerich2, F.C. Fabian Camacho2, N.R.G. Nancy R. Gallant2, L.A.W. Linda A. Wray3, F.A. Frank Ahern3, G.B. Greg Bogdan4, G.W. Gene Weinberg4, J.S.U. Jan S. Ulbrecht3 1 Highmark, Inc.; Penn State, CAMP HILL, United States of America 2 The Pennsylvania State University, College of Medicine, Department of PHS, HERSHEY, United States of America 3 Penn State University, College of Health & Human Development,Dept Biobehavioral, UNIVERSITY PARK, United States of America 4 Pennsylvania Department of Health, Div. of Community Epidemiology, HARRISBURG, United States of America Introduction: This study examined in adult Pennsylvanians from 1995 to 2010 trends in the prevalence and sociodemographic distributions of diabetes and the associations of diabetes with obesity over time. Methods: BRFSS data collected between 1995 and 2010 were used for this study. Diabetes prevalence was assessed by self-report of physician diagnosis. Obesity was assessed by BMI computed from self-report of height and weight. State-level data for diabetes and associated obesity prevalence from 1995 to 2010 were reported each year. The sociodemographic factors (age, sex, race, income, education) and the known disease risk factor, obesity were reported for 1995-2010. Logistic regression modeling was used to examine associations between diabetes, sociodemographic factors and obesity. Results: Diabetes prevalence in Pennsylvania, which increased from 5.6% in 1995 to 10.5% in 2010, followed national trends, but exceeded them each year by approximately 0.6% for 12 out of the 16 years. A pattern of obesity becoming a more dominant risk factor for diabetes emerged over this sixteen year time period. Conclusion: The burden of diabetes and obesity in Pennsylvania is substantial and increasing. Our study demonstrates that the increase in diabetes prevalence is not equal across all sociodemographic groups. Program managers and policy-makers in Pennsylvania should consider these trends when allocating limited resources and programs intended to reduce the morbidity from diabetes. Other states may consider similar studies to monitor the prevalence of diabetes in their own states to determine if the disparities are changing and whether programs and resources should also shift. CORRESPONDING AUTHOR: Conslt;Adj Prof O.G.D. Dominic, H ig h m ar k , I nc . ; P en n S t at e , U n it e d S t at e s o f Am er ic a , [email protected] O419 A LONGITUDINAL STUDY OF CHILDREN'S OBESITY, TELVISION EXPOSURE, PARENTAL RULES ABOUT TV, AND PHYSICAL ACTIVITY L.B.S. Snyder, T.M. tao Ma University of Connecticut, STORRS, United States of America Although television viewing is consistently correlated with children’s obesity, longitudinal studies have mixed results and the causal direction has not been established. The current study tested whether the relationship is that of a reinforcing spiral (Slater, 2007), in which television exposure has an effect on children’s body mass index (BMI), and at the same time BMI influences choices about time spent with television. The study also examined parental mediation in the form of household rules restricting television viewing plays a protective role, thus dampening the relationship between viewing and obesity. The relationships were tested in a large national longitudinal study of children in the U.S. (N = 21,260) followed from kindergarten to 8th grade. Using multilevel and lagged modeling, the results indicated that there were contemporaneous effects between BMI and TV exposure, a lagged effect from TV exposure to BMI, and that childhood trends in BMI and TV exposure were related. Having a larger BMI on average than other children or going through a period in which

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

the child’s BMI was larger than their own average were both related to watching more TV at a given time point. In addition, watching more TV was related to having an increased BMI at the next wave. Physical activity and time-based rules were negatively related to TV viewing, but contentbased rules had mixed effects. The findings support that TV effects on children’s obesity can be conceptualized as a spiral, and that interventions could target time-based rules and physical activity to reduce viewing time. CORRESPONDING AUTHOR: Professor L.B.S. Snyder, University of Connecticut, United States of America, [email protected] O420 CRP, DEPRESSIVE SYMPTOMS AND INCIDENCE OF DIABETES: RESULTS FROM THE ENGLISH LONGITUDINAL STUDY OF AGEING K.J. Smith, B. Au, N. Schmitz McGill University, MONTREAL, Canada Introduction: People who have depression are more likely to develop diabetes. Postulated mechanisms for this association include poorer lifestyle behaviours and increased BMI. However, researchers are increasingly interested in the role that inflammation may have in this relationship. Evidence exists that depression is associated with increased levels of the inflammatory marker CRP, and increased CRP has been linked to risk of diabetes onset. The aim of this study was to assess the association of both CRP and depressive symptomatology with diabetes incidence in a representative sample of English people ≥ 50 years old. Methods: Participants were 5475 community-dwelling men and women without diabetes at baseline from the English Longitudinal Study of Ageing. Wave 2 was used as baseline (first assessment of CRP), with diabetes incidence assessed over 6-years of follow-up. Depressive symptoms were assessed using the 8-item CES-D scale (≥4 high depressive symptoms). High CRP level was defined as > 3 mg/L. Diabetes incidence was indicated by self-reported doctor diagnosis. Association of diabetes incidence with baseline CRP and depressive symptomatology groups was examined using multivariate logistic regression adjusted for sociodemographic, lifestyle, metabolic, and health variables. Results: Participants with both high CRP and elevated depressive symptoms were more likely to develop diabetes over 6 years of follow-up compared to people with high CRP and low depressive symptoms and people with normal CRP and elevated depressive symptoms. Conclusions: People with high CRP and elevated depression symptoms are more likely to develop diabetes. CORRESPONDING AUTHOR: Doctor K.J. Smith, McGill University, Canada, [email protected] O421 MENTAL HEALTH IN ADOLESCENTS WITH TYPE 1 DIABETES. RESULTS FROM A LARGE POPULATION-BASED STUDY B. Sivertsen1, K.J. Petrie2, A. Langeland-Wilhelmsen3, M. Hysing4 1 Norwegian Institute of Public Health, BØNES, Norway 2 University of Auckland, AUCKLAND, New Zealand 3 University of Bergen, BERGEN, Norway 4 Uni Research, Uni Health, BERGEN, Norway Objective/Introduction: To compare symptoms of mental health problems, sleep and eating disturbances in adolescents with and without type 1 diabetes (T1D) in a population based sample. Methods: Data was taken from the ung@hordaland study, a large population based study in Hordaland County, Norway, conducted in 2012,. In all, 9883 adolescents aged 17-19 years (53% girls) provided self-reported data on both diabetes and a range of instruments used to asess mental health symptoms, including depression, anxiety obsessive-compulsive behaviours, hyperactivity, impulsivity, inattention, perfectionism, resilience, sleep problems and eating behavior.

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

Results: 40 adolescents were classified as having T1D, (prevalence 0.4%). We found adolescents with T1D did not differ from healthy peers on any of the mental health measures. Conclusion: This is one of the first population-based studies to examine mental health of adolescents with T1D. There was no evidence of increased psychopathology across a wide range of mental health measurements. These findings contradict previous studies, and suggest that T1D is not associated with an increased risk of psychosocial problems. CORRESPONDING AUTHOR: Prof. B. Sivertsen, Norwegian Institute of Public Health, Norway, [email protected] O422 DIABETES-RELATED ILLNESS BELIEFS IN BRITISH SOUTH ASIANS: A MIXED METHODS STUDY N.R. Patel1, Bundy Christine1, C. Blickem1, A.P. Kennedy2, C.A. ChewGraham3, D. Reeves1 1 University of Manchester, MANCHESTER, United Kingdom 2 University of Southampton, SOUTHAMPTON, United Kingdom 3 Keele University, KEELE, United Kingdom Introduction: Diabetes is disproportionally high among British South Asians (BSAs) compared to the general UK population. Good selfmanagement is important to manage diabetes. The Common Sense Model (CSM) indicates beliefs about diabetes play an important role in successful Diabetes Management (DM). Little is known about the sociocultural context in shaping beliefs about diabetes in BSAs within the CSM. We aimed to explore the influence of the sociocultural context on illness beliefs in BSAs with diabetes. Methods: A mixed methods, sequentially embedded design was used. 67 participants recruited using random and purposive sampling in North West England, completed a questionnaire measuring illness beliefs, fatalism, and health outcomes; 37 participants completed a social network survey interview and semi-structured interviews to explore the sociocultural context. Findings were analysed using SPSS and thematic analysis. Results: Analysis of the quantitative data suggests that emotional and practical support for DM from social networks was related to concern beliefs, emotional distress and health outcomes (p<0.05). After multivariate analysis, emotional support was a significant predictor of concern beliefs and emotional distress related to diabetes (p<0.05). Analysis of the qualitative data suggests that a fatalistic attitude influences DM and beliefs, alternative food ‘therapies’ are sought, often sanctioned by the family. Conclusion: Diabetes-related illness beliefs and management appear to be shaped by the sociocultural context. The CSM should move beyond the notion of ‘self’ to capture the wider determinants of behaviour which could facilitate the development of culturally sensitive interventions to modify or adapt beliefs to support DM in BSAs. CORRESPONDING AUTHOR: Dr N.R. Patel, University of Manchester, United Kingdom, [email protected] O423 ROLE OF DOPAMINE IN HYPNOTIC ANALGESIA UNDER VISCERAL STIMULATION S. Fukudo1, J. Morrishita2, S. Watanabe3, K. Kawahata2, N. Ishizu2, Y. Tanaka2, S. Watanuki2, K. Hiraoka2, M. Kano2, M. Kanazawa2, K. Yanai1, M. Tashiro2 1 Tohoku University, Graduate School of Medicine, Miyagi, Japan 2 Tohoku University, Sendai, Japan 3 Tohoko University, Sendai, Japan Background & Aims: Hypnotic suggestion changes pain via modifying activity of the regional brain. However, there is no study of the effect of hypnotic suggestion on the neurotransmitter release in the brain under visceral pain. We hypothesized that hypnotic suggestion would increase dopamine release in the basal ganglia in normal healthy men.

S127

Methods: Subjects were 15 healthy males. Dopamine D2 receptor neuroimaging was performed with 11C-raclopride positron emission tomography (PET) twice in all subjects. Colorectum was distended with the barostat with or without hypnotic analgesic suggestion in 10 subjects. Distention effect on D2 neuroimaging was clarified with comparison between distention and no distention in 5 subjects. 11C-raclopride was injected at the end of a set of bag inflation and the visceral stimulation was repeated after the ligand injection. Changes in D2 binding in the caudate nucleus and putamen were analyzed using Statistical Parametric Mapping 5. Results: Binding potential of left caudate nucleus was significantly less at analgesic suggestion with colorectal distention than at no suggestion with colorectal distention (p < 0.01). Binding potential of left putamen was significantly less at analgesic suggestion with colorectal distention than at no suggestion with colorectal distention (p < 0.05). D2 binding potential in the caudate or putamen showed no significant difference between at distention alone and at no distention. Conclusion: These data suggest that analgesic suggestion with hypnosis under visceral sensory inputs induces dopamine release in the caudate nucleus and putamen and that endogenously released dopamine inhibits binding exogenously administered 11C-raclopride to D2 receptors. CORRESPONDING AUTHOR: S. Fukudo, Tohoku University, Graduate School of Medicine, Japan, [email protected] O424 PSYCHOPATHOLOGICAL PERSPECTIVE OF IRRITABLE BOWEL SYNDROME D.L. Dumitrascu 2nd Medical Dept. Iuliu Hatieganu University of Medicine and Pharmacy, Cluj-Napoca, Romania Introduction The irritable bowel syndrome (IBS) is a perfect application for the biopsychosocial (BPS) model (Tanaka et al 2011). Given the implication of personality and mood factors in the expression of IBS symptoms, there is a need to evaluate the role of psychosocial factors and the therapy addressed to them. The aim of this presentation is to update our previous review on this topic (WJG 2012), to undertake a systematic review of therapies addressed to psychological factors in IBS and to show the efficiency of training MDs to approach IBS throught he paradigm of BPS medicine. Methods 1.We reviewed all papers on psychosocial factors on IBS and selected only original papers on Medline, Embase and Thomson Reuters. 2. A systematic review of the publications on these factors has been carried out. 3. We compared a group of IBS pts who were managed by doctors trained according to the usual curricula and compared with a matched group of patients managed by doctors trained in BPS medicine. Results. 1. There is an increasing number of psychosocial factors involved in IBS, more than 30 have been approached by different studies. 2. The systematic review shows the superiority of BCT, hypnotherapy and other interventions in IBS. 3. Pts with IBS treated by doctors trained in BPS medicine have better symptom scores and are better satisfied than the control group. Conclusions. The psychosocial factors cannot be ignored in the approach of IBS. CORRESPONDING AUTHOR: D.L. Dumitrascu, 2nd Medical Dept. Iuliu Hatieganu University of Medicine and Pharmacy, Romania, [email protected] O425 INSULAR ACTIVITY DURING RECTAL DISTENSION ASSOCIATED WITH INCREASED COLONIC RESPONSE TO CRH IN IBS SUBJECTS M. Kano1, S. Fukudo2, T. Muratsubaki1, M. Yagihashi1, J. Morishita1, Y. Tanaka1, H.G. Ly3, P. Dupont3, L. Van Oudenhove3, M. Kanazawa2 1 Tohoku University, Miyagi, Japan 2 Tohoku University, Graduate School of Medicine, Miyagi, Japan 3 University of Leuven, Leuven, Belgium

S128

Introduction: This study examined the hypothesis that intravenous injection of Corticotropin releasing hormone (CRH) induces increased rectal motor responses in irritable bowel syndrome (IBS) compared to healthy controls, and to relate these motor responses to brain responses during rectal distension. Methods: Nineteen subjects with IBS and 20 healthy controls participated in the study. In the first experiment, CRH 2μg/Kg was administrated intravenously and plasma cortisol and phasic volume events (PVEs), as index of fine colonic contractions, measured. In a second experiment, the brain response to mechanical rectal distension was measured among the same subjects by functional magnetic resonance imaging (fMRI). Results: There was no significant difference between IBS and controls in cortisol levels after CRH injection. IBS subjects demonstrated a significant increase in number of PVEs after CRH injection compared to before injection (p=0.002), whereas controls did not. In the fMRI session, rectal balloon distension induced increased brain activity in the visceral pain matrix in both IBS subjects and controls. IBS subject increased activity in the right rolandic opeculum/insula significantly more than controls. Activity of the insula was correlated with the number of the PVEs after CRH injection (over both groups) in the first experiment (all brain data was significant at p<0.001, uncorrected). Conclusion: These results suggest that CRH injection induces exaggerated fine contractions of the colon in IBS subjects independent of cortisol secretion. Increased insular response to rectal distension may be related to the increased colonic motor response to CRH. CORRESPONDING AUTHOR: M. Kano, Tohoku University, Japan O426 PRECEDING STIMULATION AFFECTS BRAIN ACTIVATION IN RESPONSE TO VISCERAL STIMULATION T. Hamaguchi1, S. Fukudo2, M. Kano3, M. Kanazawa2, M. Itoh2, K. Yanai2 1 Saitama Prefectural University, School of Health and Social Services, Saitama, Japan 2 Tohoku University, Graduate School of Medicine, Miyagi, Japan 3 Tohoku University, Miyagi, Japan Background: Repetitive colonic stimuli may induce visceral hyperalgesia in patients with irritable bowel syndrome. The order of distension (moderate preceded by mild and vice versa) has been suggested to influence brain activation and the experience of hypersensitivity, offset analgesia, and anticipation. We hypothesized that brain activation arising from colonic distention with an intense initial stimulation would differ from that arising from intense stimulation preceded by mild stimulation. Methods: In 45 nonclinical healthy subjects, brain processing of visceral sensation induced by colonic distension was examined using H215O positron emission tomography. Subjective feelings regarding the stimuli were also measured. The descending colon was stimulated using six patterns of three bag pressures (0, 20, and 40 mm Hg). To evaluate the neural sensitization to visceral stimulation arising from the precedence effect, the effects of a 20- or 40-mm Hg distention after a sham or 20- or 40-mm Hg distension were analyzed using statistical parametric mapping. Results: The midbrain, insula, and cerebellum, were more strongly activated by a 20-mm Hg distension with a preceding 40-mm Hg distention than by a 20-mm Hg distention without a preceding stimulation (p < .0001). Conversely, a sham stimulation after the experience of an intense stimulation activated the midcingulate cortex, compared with a sham stimulation without the experience of actual visceral stimulation (p < .0001). Conclusions: By directly comparing different patterns of visceral stimuli, preceding visceral stimuli may affect neural sensitization and desensitization in healthy subjects, including elevated midbrain, insula, and midcingulate cortex. CORRESPONDING AUTHOR: Prof. T. Hamaguchi, Saitama Prefectural University, School of Health and Social Services, Japan, [email protected]

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

O427 MINDFULNESS REDUCES CUE-INDUCED FOOD CRAVING: AN FMRI STUDY T. Murastubaki 1 , S. Fukudo 1 , M. Kano 2 , A. Sekiguchi 3 , Y. Ishigaki4, S. Sawada1, K. Kondo1, Y. Kotozaki2, A. Sasaki1, J. Morishita2, M. Kanazawa1, H. Katagiri1, R. Kawashima2 1 Tohoku University, Graduate School of Medicine, Miyagi, Japan 2 Tohoku University, Miyagi, Japan 3 Tohoku University, Medical Megabank Organization, Miyagi, Japan 4 Iwate Medical University School of Medicine, Miyagi, Japan Background & Aims: One way that mindfulness may facilitate control of eating is through the reduction of craving to food cues. There are two candidate neural pathways that link this strategy to reduced craving. One of the pathways may recruit regulatory brain regions in a top-down manner, like cognitive reappraisal. Another pathway may be associated with reduction of craving by directly decreasing craving-related neural reactivity. The present work exams whether mindfulness can reduce selfreported and neural markers of cue-induced craving. Methods: Nineteen healthy subjects viewed and made ratings of food and neutral images while undergoing fMRI. Subjects were trained and instructed to view these images passively, with cognitive reappraisal or mindfulness. Results: Mindfulness reduced self-reported craving to food images. This strategy reduced neural activity in a craving-related region of left insula but did not increase neural activity in an inhibition-related region. Cognitive reappraisal similarly reduced self-reported craving to food images. It increased neural activity in an inhibition-related region of left dorsolateral prefrontal cortex and left ventrolateral prefrontal cortex, but did not reduce neural activity in a craving-related region of insula. Conclusion: These results provide an initial indication that mindfulness may induce modification without relying on top-down pathways to one’s experience in ways that can help reduce subjective and neural reactivity to food cues. CORRESPONDING AUTHOR: T. Murastubaki, Tohoku University, Graduate School of Medicine, Japan O428 NEUROTICISM AND IRRITABLE BOWEL SYNDROME J. Tayama1, S. Fukudo2, N. Nakaya3, T. Hamaguchi4, T. Tomiie5, M. Shinozaki2, T. Saigo1, S. Shirabe1 1 Center for Health and Community Medicine, Nagasaki University, NAGASAKI, Japan 2 Tohoku University, Graduate School of Medicine, Miyagi, Japan 3 Tohoku Medical Megabank Organization, Tohoku University, Miyagi, Japan 4 Saitama Prefectural University, School of Health and Social Services, Saitama, Japan 5 Health Sciences University of Hokkaido, SAPPORO CITY, Japan Background & Aims: Previous studies have reported that patients with irritable bowel syndrome (IBS) show high neuroticism. However, the precise association between the IBS subtypes and the degree of neuroticism in younger populations is largely unknown. We tested our hypothesis that subjects with diarrhea-predominant IBS may have a higher degree of neuroticism than subjects without IBS or those with other subtypes of IBS. We also verified the additional hypothesis that the severity of neuroticism might be correlated with the severity of IBS in younger populations. Methods: We conducted a cross-sectional survey of 557 university students, ranging in age from 18 to 21 years. Presence/ absence of IBS and determination of the IBS subtype was by the Rome II Modular Questionnaire, while the severity of IBS was determined by the IBS severity index (IBS-SI). The degree of neuroticism was evaluated using the Maudsely Personality Inventory (MPI). The presence/absence of psychological distress was measured with the K6 scale.

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

Results: Neuroticism scores in the subjects with diarrhea-predominant IBS were significantly higher than those in the non-IBS subjects or subjects with constipation-predominant IBS. The neuroticism scores were significantly correlated with the IBS-SI scores in all subjects with IBS. Conclusion: These results suggest that neuroticism is involved in the pathophysiology of IBS in young subjects, especially in that of the diarrhea-predominant subtype. CORRESPONDING AUTHOR: J. Tayama, Center for Health and Community Medicine, Nagasaki University, Japan O429 RISK OF BIAS IN HIV-TREATMENT ADHERENCE INTERVENTION TRIALS: A SYSTEMATIC REVIEW AND META-ANALYSIS OF RCTS M. de Bruin University of Aberdeen, ABERDEEN, United Kingdom Background: Randomized controlled trials (RCTs) can offer strong causal evidence on intervention efficacy, but this can be threatened by multiple sources of bias. More needs to be understood regarding the extent to which these sources of bias exist and how they may impact on intervention effects in health behavior intervention reviews. This paper examines risk of bias in HIV-treatment adherence intervention trials.Methods: A systematic review of HIV medication adherence intervention trials. Automated and manual searches utilizing multiple electronic bibliographic databases were conducted for RCTs published between January, 1996 and May, 2012. Data was abstracted for study characteristics, methodological study quality characteristics to assess the risk of bias (using Prevention Research Synthesis (PRS) efficacy review criteria and Cochrane Review guidelines), and intervention effect estimates for adherence and viral load. A meta-analysis, including meta-regression techniques, was conducted to assess the relationship between risk of bias and study characteristics and intervention effects.Results: 80 RCTs were obtained and coded. There was substantial risk of bias due to inadequacies in (the reporting of) random sequence generation, allocation concealment, and the intent-totreat analysis, and the use of subjective behavioral measures. Risk of bias did not impact intervention effect estimates for HIV viral load, but did influence effect sizes for adherence behaviors.Conclusions: Improving on trial design and the comprehensiveness of trial design reports are needed to improve the quality of the evidence on HIV-treatment adherence trials. CORRESPONDING AUTHOR: Prof M. de Bruin, University of Aberdeen, United Kingdom, [email protected] O430 PSYCHOLOGICAL IMPACT, ADHERENCE AND HEALTHRELATED QUALITY OF LIFE IN ADVANCED CHRONIC KIDNEY DISEASE H. Garcia-Llana1, eduardo Remor2, Rafael Selgas1 1 Hospital Universitario La Paz, Madrid, Spain 2 Universidad Autonoma de Madrid, Madrid, Spain INTRODUCTION: Health related quality of life (HRQOL) has been widely studied in the field of dialysis patients. However, there are few studies that include relationships of psychosocial variables and adherence to treatment with HRQOL METHOD: This presentation describes the outcomes of a PhD dissertation including 3 consecutive studies. I. A systematic literature review to synthesize available information on the role that psychological variables have on HRQOL of Advanced Chronic Kidney Disease (ACKD); II. A cross-sectional study to observe the relationship between depression, anxiety, perceived stress, adherence and HRQOL on 61 dialysis patients (home-based dialysis and hemodialysis); III. An intervention study to test a motivational interviewing-based program in improving psychological variables, adherence, HRQOL and biological markers of renal failure

S129

progression on a sample of 42 pre-dialysis patients.RESULTS: A small number of studies regarding psychological variables, adherence and HRQOL were available on literature, the majority of them were focused in the role of depression. The predictors of HRQOL in 61 dialysis patients were: work status, gender and depression. Motivational interviewingbased program in pre-dialysis patients improved levels of adherence and HRQOL, decreased depression and anxiety, and biological markers were better controlled. CONCLUSION: Screening for early indicators of anxiety/stress, depression and difficulties to complying with treatment in dialysis population is mandatory. A program based on a motivational interviewing approach is effective for promoting adherence and wellbeing in ACKD patients on pre-dialysis stage. CORRESPONDING AUTHOR: DR. H. GARCIA-LLANA, HOSPITAL UNIVERSITARIO LA PAZ, Spain, [email protected] O431 PERCEPTUAL AND PRACTICAL DETERMINANTS OF AED ADHERENCE, PERCEPTIONS OF MEDICATION INFORMATION AND DECISION-MAKING IN EPILEPSY S.C.E. Chapman1, S.M. Sisodiya2, R. Eade3, R. prof Horne1 1 UCL School of Pharmacy, LONDON, United Kingdom 2 UCL Institute of Neurology, LONDON, United Kingdom 3 Epilepsy Society, LONDON, United Kingdom Introduction: Adherence to antiepileptic medicines (AEDs) ensures that patients receive the full benefit from their medication. However, approximately 20-40% of patients do not adhere to their AEDs, increasing risk of seizures, hospitalization, mortality and treatment costs. The current study aimed to assess adherence and key modifiable predictors of AEDs nonadherence in epilepsy . Methods: In this cross-sectional study, 1010 epilepsy patients currently taking AEDs completed an anonymous, voluntary online survey hosted by the UK charity Epilepsy Society. They answered a questionnaire about their adherence to AED, their perceptions of AED, practical barriers to adherence to AEDs, their perceptions of shared decision-making about AEDs and their views on sources of information about AEDs. Results: More than half of respondents were ambivalent about their medications, being convinced of their need for AEDs while having high concerns about the potential adverse effects of AEDs. Both these perceptual factors, and practical barriers to medication-taking, predicted adherence to AEDs (all ps < 0.001), perceptions of information from healthcare providers (p < 0.01), and empowerment in decision-making about treatment (p = 0.03). Conclusions: Potentially modifiable perceptual and practical factors are associated with key areas of unmet needs for epilepsy patients including perception of information, involvement in treatment decisions and adherence to AEDs. Interventions tailored to address patients’ concerns about potential adverse effects of AEDs, doubts about need for AEDs and practical barriers may improve outcomes for epilepsy patients. CORRESPONDING AUTHOR: Dr S.C.E. Chapman, UCL School of Pharmacy, United Kingdom, [email protected] O432 DOES A CARDIOVASCULAR EVENT CHANGE ADHERENCE TO STATIN TREATMENT IN PATIENTS WITH TYPE 2 DIABETES? F.M. de Vries1, P. Denig2, S. Vegter1, H.J. Bos1, M.J. Postma1, E. Hak1 1 University Groningen, Department of Pharmacy, GRONINGEN, The Netherlands 2 University of Groningen, University Medical Center Groningen, GRONINGEN, The Netherlands Introduction: Statins are associated with a reduced risk of cardiovascular events. To be effective, adherence is essential. We assessed the effect of a cardiovascular event on statin adherence in type 2 diabetes patients.

S130

Method: A matched cohort study was conducted among diabetes patients initiating statins for primary prevention in the Groningen University IADB.nl pharmacy database. Patients experiencing a cardiovascular event (index date) were matched to a reference patient without such an event on gender, age, initiation date, follow-up period and adherence level before the event. Adherence rates were measured as percentages of days covered (PDC), and shifts in adherence levels (non-adherent/partial-adherent/full-adherent) and rates around the event were evaluated. Results: We could match 375 of the 855 eligible index patients with a reference. Index patients had on average a PDC of 81% after the index date; reference patients had a PDC of 71% (P<0.001) while both had a PDC of 79% before the index date. Index patients were 4.5 times more likely than reference patients to shift from non-adherent to full-adherent [95%CI 1.1-18.8] and 1.8 times more likely to shift from partial-adherent to full-adherent [95%CI 1.2-2.6]. In the index group, 26% of patients became more adherent after the event. In contrast, 20% of patients became less adherent. Conclusion: The occurrence of a cardiovascular event appeared to avert the declining statin adherence rate observed in diabetes patients without such an event. However, one in five patients became less adherent after the event, indicating that there are still important health benefits to achieve. CORRESPONDING AUTHOR: F.M. de Vries, University Groningen, Department of Pharmacy, The Netherlands, [email protected] O433 MEDICATION TAKING BEHAVIOUR AND ADVERSE HEALTH OUTCOMES IN COMMUNITY DWELLING OLDER PATIENTS C.C. Cahir1, T. Fahey2, C. Teljeur3, K. Bennett1 1 Trinity College Dublin, DUBLIN 8, Ireland 2 Royal College of Surgeons in Ireland, DUBLIN, Ireland 3 Health Information and Quality Authority, DUBLIN, Ireland Introduction: To determine the association between medication taking behaviour and adverse drug events (ADEs), health related quality of life (HRQOL) and hospitalisation in older community dwelling patients. Methods: Retrospective cohort study of 855 patients aged ≥70 years from 15 general practices. Medication non-adherence was measured by: (i) Medication Possession Ratio (MPR) using national pharmacy claims data; and (ii) self-report using the Morisky Medication Adherence Scale. ADEs and hospitalisation were measured by patient medical record and self-report for the previous 6 months. HRQOL was measured using EQ5D. Multilevel Poisson and linear regression were used to examine how the number of ADEs, utility and hospitalisation varied by non-adherence after adjusting for covariates; socioeconomic status, deprivation, comorbidity, number of drugs, functional disabilities, social support and health insurance. Results: 263 (31%) patients were non-adherent based on pharmacy claims data (MPR<80%) and 302 (35%) self-reported non-adherence to their medication. Non-adherence (MPR<80%) was not significantly associated with any ADEs but self-reported non-adherent patients had an increased risk of any ADEs (IRR 1.18; 95% CI 1.05, 1.33, p<0.01). Nonadherent patients had a significantly lower mean HRQOL utility (MPR coefficient, -0.11, SE 0.03, p<0.001; self-report coefficient, -0.06, SE 0.01, p<0.001) and an almost two-fold increased risk in the expected rate of any hospitalisation (MPR IRR 1.75; 95% CI 1.42, 2.15, p<0.001; self-report IRR, 1.53; 95% CI 1.16, 2.01, p<0.01) compared to adherent patients. Conclusion: Non-adherence was significantly associated with adverse health outcomes. Developing methods to assist older adults in managing their medications may increase their quality of life. CORRESPONDING AUTHOR: Dr C.C. Cahir, Trinity College Dublin, Ireland, [email protected]

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

O434 SLEEPCARE: TOWARDS IMPROVING ADHERENCE IN AUTOMATED PERSONALIZED SLEEP COACHING F. Griffioen-Both1, Corine Horsch2, Siska Fitrianie2, Sandor Spruit1, Jaap Lancee3, Rogier M. van Eijk1, Willem-Pa Brinkman2, Robbert J. Beun1 1 University Utrecht, UTRECHT, Nederland 2 Technical University Delft, DELFT, Nederland 3 University of Amsterdam, AMSTERDAM, Nederland Introduction. Improving assignment adherence is a major challenge in all types of cognitive behavior therapies (CBT), in particular in self-help therapies from books and internet sites. In the Sleepcare-project, methods from persuasive technology are investigated to improve adherence of selfhelp CBT for insomnia (CBT-I). The project aims at a generic coaching system that integrates persuasive strategies with protocols for behavior change in CBT-I, and the underlying supporting technology. Method. A fully automated mobile coaching application is being developed that offers a therapy for insomnia (e.g. sleep restriction, relaxation exercises). The first phase of the project consists of the translation of existing CBT-I protocols and persuasive strategies (e.g. commitment, notification) into a formal model and the implementation of the model. Results. The translation process resulted in an ontology containing classes and properties of both the CBT-I therapy and persuasive techniques. In addition, a rule-based system and specialized dialogue techniques were developed that control the interaction between coach and coachee. All techniques were implemented in a fully automated smartphone application and integrated in a research infrastructure for empirical testing. Conclusion. Existing CBT-I protocols were successfully implemented into a fully functioning automated coaching system. The use of an explicit ontology and a rule-set enables the separation of domain specific knowledge of insomnia and the applied persuasive strategies. In the second phase, the system will be extended with additional strategies (e.g. reciprocity, motivational support). The resulting coaching system will be empirically tested in an RCT. CORRESPONDING AUTHOR: dr. F. Griffioen-Both, University Utrecht, Nederland, [email protected] Interventions addressing the psychological concerns and needs of women in the breast cancer context S435 INTERVENTIONS ADDRESSING THE PSYCHOLOGICAL CONCERNS AND NEEDS OF WOMEN IN THE BREAST CANCER CONTEXT K. Sherman1, Phyllis Butow2, Eveline Bleiker3, Linda Cameron4 1 Centre for Emotional Health, SYDNEY, Australia 2 Psycho-Oncology Co-operative Research Group (POCOG), University of Sydney, SYDNEY, Australia 3 Netherlands Cancer Institute, AMSTERDAM, Nederland 4 University of California, MERCED, United States of America Breast cancer presents individuals with challenges on many levels. Women recently diagnosed with breast cancer face difficult decisions about treatment, particularly which type of surgery to choose. Following the active treatment phase, many women experience constant doubts about remaining cancer free. Another aspect that presents a number of psychological and emotional challenges to women in the breast cancer context is the process of undergoing genetic risk assessment for breast cancer susceptibility. The aim of this symposium is to present results of randomised controlled trials of three psychosocial interventions designed to enhance the psychological wellbeing of women undergoing genetic testing or faced with a diagnosis of breast cancer. The first intervention, BRECONDA, a web-based interactive decision aid is designed to facilitate decision making in women considering breast reconstruction following mastectomy. Results indicate high levels of user acceptability and

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

patient benefits regarding the quality of the overall decisional process. The second 5-session intervention focuses on helping cancer survivors conquer fears of cancer recurrence. Findings from the trial indicate a high level of acceptability to both patients and therapists alike and potential benefits of decreased psychological morbidity. The final intervention addresses psychosocial issues that arise in women who undergo genetic risk assessment through the use of a Psychosocial Aspects of Hereditary Cancer questionnaire in the genetic counselling sessions. Findings indicate that the questionnaire led to increased discussion of problems in the counselling session, increased wellbeing and decreased general distress of women undergoing genetic risk assessment. Implications of these intervention approaches will be discussed. CORRESPONDING AUTHOR: Associate Prof. K. Sherman, Centre for Emotional Health, Australia, [email protected] S436 THE BRECONDA TRIAL: FACILITATING SURGICAL DECISION MAKING OF WOMEN WITH BREAST CANCER K. Sherman1, Diana Harcourt2, Laura-Kat Shaw1, Caleb Winch3, Paul Brown4, Linda Cameron4, Thomas Lam3, John Boyages3 1 Centre for Emotional Health, SYDNEY, Australia 2 Centre for Appearance Research, University of the West of England, BRISTOL, United Kingdom 3 Macquarie University Cancer Institute, SYDNEY, Australia 4 University of California, MERCED, United States of America Introduction. Women requiring mastectomy for breast cancer are faced with the difficult preference-based decision regarding whether, and how, to restore breast shape after surgery. We have developed a web-based decision aid, BRECONDA, to assist with this decision-making. In a randomised controlled trial we compared the efficacy of BRECONDA with a general educational pamphlet. Methods. Women diagnosed with breast cancer recruited into the multicentre web-based trial completed baseline questionnaires and were then randomly assigned into either: 1) Intervention (INT) - received unlimited access to BRECONDA and an electronic information booklet about breast reconstruction; or, 2) Control (CONT) - received the information booklet alone. Participants then completed 1- and 6-month follow-up questionnaires assessing satisfaction with information, decisional conflict and knowledge. INT participants completed ratings of perceived usefulness of the intervention and appropriateness of the content. Results Mixed method analyses (random intercept, maximum likelihood) demonstrated that INT participants reported lower decisional conflict at 1-(INT=27.3; CONT=35.5) and 6-(INT=23.7; CONT=30.9, p=.014) month follow-up, and greater satisfaction with information (1-month: INT=3.7, CONT=3.4; 6-months: INT=3.9, CONT=3.6, p=.005). INT participants indicated high to very high overall satisfaction with the intervention (78%), reporting that it was easy to use (75%), presented a balanced view (74%) and that the amount of information was “just right”(77%). Conclusions. These findings indicate that intervention participants benefitted regarding their overall decisional process quality with greater satisfaction with information, lower decisional conflict and high user acceptability of the intervention, supporting the feasibility and acceptability of implementing BRECONDA into oncological surgery practice. CORRESPONDING AUTHOR: Associate Prof. K. Sherman, Centre for Emotional, Health, Australia [email protected] S437 CONQUER FEAR - HELPING CANCER SURVIVORS BETTER MANAGE FEAR OF CANCER RECURRENCE P. Butow1, B. Thewes1, A. Smith1, J. Fardell1, J. Gilchrist2, J. Turner3, L. Sharpe1, A. Girgis4

S131 1

University of Sydney, SYDNEY, Australia Macquare University Hospital, Clinical Psychology, SYDNEY, Australia 3 School of Medicine, University of Queensland, BRISBANE, Australia 4 6. South Western Sydney Clinical School, University of NSW, SYDNEY, Australia 2

Introduction Fear of cancer recurrence (FCR) is the most prevalence unmet need reported by cancer survivors, and is associated with increased psychological morbidity and excessive or insufficient medical surveillance. Our group is evaluating in a RCT, a novel 5-session intervention for FCR based on the Common Sense Model of Illness, Self-Regulatory Executive Function Model and Relational Frame Theory. Methods Adult early-stage breast or colorectal cancer survivors (n=160), seen at participating clinics, reporting an FCR score in the clinical range on the FCRI, are randomized to the intervention or attention control (relaxation therapy) arms post consent. Both intervention and control treatments are delivered by 23 psychologists/psychiatrists working in hospital oncology clinics across Australia. Participants complete measures immediately prior to and after the intervention, and 3 months and 6 months later. Results. In a preliminary pilot completed by 4 trained clinicians and 8 patients, all patients completed the intervention, and on average reported a decrease in FCR of 8/36 points on the FCR severity subscale (p=0.002, effect-size 1.9); an increase in quality of life of 13/100 points on the FACT-G (p=0.1, effect-size 0.7) and a decrease of 18/75 points on the impact of events scale (p=0.03, effect-size 1.2). Patients rated therapy helpfulness on average as 8/10. Conclusions: The intervention appears acceptable to both patients and therapists, feasible and likely effective; RCT results will be discussed. CORRESPONDING AUTHOR: Prof P. Butow, University of Sydney, Australia, [email protected] S438 SCREENING FOR PSYCHOSOCIAL PROBLEMS IN GENETIC COUNSELING FOR BREAST CANCER E.M.A. Bleiker1, W. Eijzenga1, D.E.E. Hahn1, M.G.E.M. Ausems2, L. vd Kolk1, N.K. Aaronson1 1 The Netherlands Cancer Institute, AMSTERDAM, Nederland 2 University Medical Center Utrecht, UTRECHT, Nederland Introduction Individuals who request genetic counseling and testing for (breast) cancer may experience a variety of problems. To facilitate the detection of these problems the Psychosocial Aspects of Hereditary Cancer (PAHC) questionnaire was developed. In this study we investigated the frequency of specific problems, and evaluated the efficacy of the PAHC questionnaire in facilitating communication and decreasing distress. Methods In a randomized controlled trial, high-risk individuals were randomized to either: 1) an intervention group, in which the counselors received feedback on the experienced psychosocial problems as indicated on the PAHC-questionnaire, or 2) a control group, where the counselors did not receive this information. The Cancer Worry Scale and Hospital Anxiety and Depression Scale were completed before randomization and four weeks after the genetic counseling. Results Data from 246 counselees, including 150 with a personal and/or family history of breast cancer, showed that 80% experienced psychosocial problems related to ‘living with cancer’, 37% related to ‘genetics’, 34% to ‘children’, 24% to ‘family and social issues’, 20% to ‘general emotions’, and 13% to ‘practical problems’. During the counseling session, significantly more problems were discussed in the intervention group. At follow-up, cancer worries and general distress levels were significantly lower in the intervention group than the control group.

S132

Conclusions One-fifth of the counselees reported emotional problems. Far more counselees experienced problems related to ‘living with cancer’ and ‘genetics’. The PAHC questionnaire is effective in facilitating communication about the specific psychosocial problems, and in reducing distress and cancer worries on the short term. CORRESPONDING AUTHOR: dr E.M.A. Bleiker, The Netherlands Cancer Institute, Nederland, [email protected] Diabetes disparities among U.S. Latinos/hispanics from diverse backgrounds S439 SOCIOCULTURAL AND PSYCHOSOCIAL ASPECTS OF D I A B E T E S P R E VA L E N C E , C O M P L I C AT I O N S A N D MANAGEMENTS IN ADULTS N. Schneiderman1, A.J.M. Boulton2, A.E. Moncrieft1, L. Vileikyte2 1 University of Miami, CORAL GABLES, United States of America 2 University of Manchester, MANCHESTER, United Kingdom Diabetes has become a growing global epidemic. Psychosocial and lifestyle behavioral factors contribute substantially to increasing prevalence of diabetes and can play a major role in decreasing this prevalence, understanding and dealing with the complications of diabetes and participating in its management. In this symposium Dr. Boulton, the moderator, will introduce the symposium from the perspective of the European Association for the Study of Diabetes by describing diabetes as a growing global epidemic. Dr. Schneiderman will focus on diabetes prevalence results from the NIH multi-center Hispanic Community Health Study/ Study of Latinos (HCHS/SOL). The HCHS/SOL found a high prevalence of diabetes among Hispanics/Latinos but considerable diversity as a function of Hispanic background. The low rates of diabetes awareness, diabetes control and health insurance in conjunction with negative association between diabetes prevalence and both household income and education, raise important health policy issues. Dr. Moncrieft will focus on the results of the CALMD randomized controlled trial that aimed to improve physical activity, nutrition and stress management in disadvantaged adults with type 2 diabetes. Beneficial effects on weight, glycemic control, renal function and depressive symptoms were observed. Dr. Vileikyte examined the role of psychological stress in diabetic foot ulcer healing. Her findings indicate that chronic inflammation related to psychological stress appears to delay healing of foot ulcers in diabetics, but does not influence adherence to offloading. Dr. Boulton will discuss the importance of current findings in understanding and dealing with diabetes as a growing global epidemic. CORRESPONDING AUTHOR: Professor N. Schneiderman, University of Miami, United States of America, [email protected] S440 DIABETES DISPARITIES AMONG U.S. LATINOS/HISPANICS FROM DIVERSE BACKGROUNDS N. Schneiderman1, M. Llabre1, C. Cowie2, J. Barnhart3, M. Carnethon4, L. Gallo5, A. Giachello4, G. Heiss6, R. Kaplan3, L. LaVange7, Y. Teng6, L. Villa-Caballero5, L Aviles-Santa8 1 University of Miami, CORAL GABLES, United States of America 2 National Institute of Diabetes, BETHESDA, United States of America 3 Albert Einstein College of Medicine, BRONX, United States of America 4 Northwestern, CHICAGO, United States of America 5 San Diego State University, SAN DIEGO, United States of America 6 University of North Carolina, Chapel Hill, CHAPEL HILL, United States of America 7 Department of Health and Human Services/FDA, SILVER SPRING, United States of America 8 Natioanl Heart, Lung, and Blood Institute, National Institutes of Health, BETHESDA, United States of America

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

Introduction: This ongoing longitudinal, multi-center study sponsored by the National Institutes of Health (NIH) examined differences in prevalence of diabetes and rates of awareness and control among adults from diverse Hispanic/Latino backgrounds in the Hispanic Community Health Study/Study of Latinos (HCHS/SOL). Methods: The HCHS/SOL, a prospective, multi-center, population based study, enrolled, from within the US during 2008 to 2011, 16,415 18-74 year old people of Hispanic/ Latino descent. Diabetes was defined by either fasting plasma glucose (FPG), impaired glucose tolerance (IGT) 2-h after a glucose load, hemoglobin A1c (A1c) or documented use of hypoglycemic agents (scanned medications). Results: Diabetes prevalence varied from 10.2% in South Americans and 13.4% in Cubans to 17.7% in Central Americans, 18.0% in Dominicans and Puerto Ricans and 18.3% in Mexicans (p<0.0001). Prevalence related positively to age (p < 0.0001), body mass index (p < 0.0001) and living in the US longer than 10 years including those born in the US, (p = 0.0232), but was negatively related to education (p=0.0005) and household income (p=0.0043). Rate of diabetes awareness was 58.7%, adequate glycemic control (A1c<7%, 53 mmol/mol) was 48.0%, and having health insurance among those with diabetes was 52.4%. Conclusions: Present findings indicate a high prevalence of diabetes but considerable diversity as a function of Hispanic background. The low rates of diabetes awareness, diabetes control and health insurance in conjunction with the negative associations between diabetes prevalence and both household income and education among Hispanics/Latinos in the US have important implications for public health policies. CORRESPONDING AUTHOR: Professor N. Schneiderman, University of Miami, United States of America, [email protected] S441 BEHAVIORAL LIFESTYLE INTERVENTION DECREASES RISK IN TYPE 2 DIABETES A.E. Moncrieft1, M Llabre1, JR McCalla1, M Gutt1, MD Gellman1, R Goldberg2, N Schneiderman1 1 University of Miami, CORAL GABLES, United States of America 2 University of Miami, Miller School of Medicine, MIAMI, United States of America Introduction: The purpose of this study was to determine the effect of a multi-component behavioral intervention on weight loss, glycemic control, depressive symptoms and renal function in a community-based sample of mostly minority, low-income adults with type 2 diabetes and marked depressive symptoms. Methods: A sample of 111 adults with type 2 diabetes (mean age = 54.81 years, 71% female, 85% Hispanic, mean household income $14,382) participated in a randomized controlled trial titled Community Approach to Lifestyle Modification in Diabetes. Eligible participants were overweight or obese, reported pronounced symptoms of depression, but had no evidence of cardiovascular or renal disease. Participants were assigned to a 1 year, 17-session intervention (n = 54) aimed at improving diet, physical activity, and stress management, or to usual care (n = 57). Outcomes included weight, glycosylated hemoglobin (HbA1c), Beck Depression Inventory-II (BDI-II) score, and estimated glomerular filtration rate (eGFR) assessed at baseline, 6months, and 12-months post randomization. Latent growth modeling was used to examine intervention effects on each outcome. Results: Using ‘intent-to-treat’ analyses, the intervention resulted in decreased weight (B = -.322, SE = .124, p = .01), HbA1c (B = -.066, SE = .028, p = .017), and depressed affect (B = - 1.009, SE = .226, p < .001), and improved eGFR (B = .742, SE = .318, p = .020) relative to usual care. Conclusions: Multicomponent behavioral interventions targeting weight loss and depressive symptoms as well as diet and physical activity are efficacious in the management of type 2 diabetes. CORRESPONDING AUTHOR: A.E. Moncrieft, University of Miami, United States of America, [email protected]

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

S442 PSYCHOLOGICAL STRESS AND DIABETIC FOOT ULCER HEALING L. Vileikyte University of Manchester, MANCHESTER, United Kingdom Introduction: Research indicates that psychological stress (PS) delays acute wound repair. Here we explored the potential role and the mechanisms by which PS impedes chronic, diabetic foot ulcer (DFU) healing. Methods: PS was conceptualized and measured as generalized (Perceived Stress, Hospital Anxiety/Depression and STAXI-Anger) and DFUspecific distress (NeuroQoL-Interpersonal Burden, Patient Interpretation of Neuropathy-Amputation Worry and Anger at Docs). Explored were two potential mechanisms by which PS impacts DFU healing: PSinduced immunomodulation and non-adherence to DFU offloading. DFU-specific biomarkers were determined via quantification of immunohistochemical tissue localization and/or normalized biopsy gene expression. Systemic biomarkers were measured from patient serum via ELISA. Adherence to Offloading (AO) was assessed using a validated, computerized dual activity monitor method. One hundred ten type 2 diabetes patients (84% male; mean age 57 yrs) with plantar neuropathic DFU (Texas Classification: 69% grade 1A; 11% 1B; 16% 2A; and 4% 2B) completed at baseline the self-report PS measures. Results: DFU healing at 6 weeks was less likely in patients reporting more severe DFUspecific distress and depression, linked to increased systemic IL6. In contrast, AO associated with faster DFU healing was not predicted by PS measures. However, more severe DFU and DFU-related restrictions in activities of daily living significantly predicted better AO, while greater neuropathic unsteadiness significantly predicted worse AO. Conclusions: These findings indicate that PS-related chronic inflammation may contribute to DFU non-healing but do not support the role of PS as a determinant of adherence to offloading. CORRESPONDING AUTHOR: L. Vileikyte, University of Miami, United States of America, [email protected] The nocebo effect: promising avenues of new research S443 THE NOCEBO EFFECT: PROMISING AVENUES OF NEW RESEARCH K.J. Petrie University of Auckland, AUCKLAND, New Zealand Chair: Keith J Petrie, University of Auckland Discussant: Winfried Rief, University of Marburg Expecting to experience side effects from a medical treatment is consistently associated with the reporting of nocebo effects. The nocebo effect has important clinical implications, as it is associated with adverse drug reactions, non-adherence and poor persistence with medical treatment. This symposium brings together researchers from Hamburg, Marburg, London and Auckland to present their recent studies on the nocebo effect. The first paper presents data from a representative general population sample and shows perceived sensitivity to medicines is relatively common in the general population and is associated with symptom reporting and GP visits. The second paper looks at side effects reported by women receiving endocrine therapy for breast cancer and shows expectations of side effects prior to treatment are a powerful predictor of adverse reactions to endocrine therapy. The third study looks at the nocebo effects of branding medicine. The study reports that generic labelled tablets have a larger associated nocebo effect than brand name tablets. This has implications for the increasing use of generic medicine worldwide. The discussant for the symposium is Professor Winfried Rief from Marburg University who is a world authority on nocebo and placebo effects, symptom reporting and cognitive behavioural therapy.

S133

CORRESPONDING AUTHOR: Prof K.J. Petrie, University of Auckland, New Zealand, [email protected] S444 PERCEIVED SENSITIVITY TO MEDICINE AND ITS ASSOCIATION WITH SYMPTOM REPORTING AND MEDICAL UTILIZATION K.J. Petrie1, A. Grey1, R. Horne2, K. Faasse1 1 University of Auckland, AUCKLAND, New Zealand 2 School of Pharmacy, University College London, LONDON, United Kingdom Introduction: Perceived sensitivity to medicines is the belief that one is especially sensitive to the actions and side effects of medicines. Perceived sensitivity to medicine has been associated with poor adherence, decreased persistence with treatment and increased side effects. These beliefs are common among clinical populations, but its prevalence in the general population is unknown. In this study we investigated the prevalence of perceived sensitivity to medicine and its association with symptom reporting and the use of medical care. Method: A nationally representative sample of 1000 participants was recruited using random digit dialling sampling. Participants completed the Perceived Sensitivity to Medicines scale, and the General Assessment of Side Effects Scale which assesses the frequency of 46 physical symptoms in the past 7 days. We also gathered demographic information, medical visits and medication use over the previous 12 months. Results: Rate of perceived sensitivity were common in the general population with 20% of the population indicating that their body was very sensitive to medicines. Higher perceived sensitivity to medicines was associated with increased rates of symptom reporting (p < .001) as well as higher number of GP visits during the previous year. Female gender, unemployment, and older age were also associated with heightened perceptions of sensitivity to medicines. Conclusion: Perceived sensitivity to medicines is relatively common in the general population and is associated with symptom reporting and GP visits. In clinical settings perceived sensitivity to medicines is likely to have significant effects on non-adherence and increased side effect reporting. CORRESPONDING AUTHOR: Prof K.J. Petrie, University of Auckland, New Zealand, [email protected] S445 NOCEBO EFFECTS IN BREAST CANCER TREATMENT: EXPECTATIONS PREDICT SIDE EFFECTS OF ENDOCRINE THERAPY Y.N. Nestoriuc1, P. Von Blanckenburg2, F. Schuricht2, U. Albert3, W. Rief2 1 University of Hamburg, HAMBURG, Germany 2 Philipps-University of Marburg, MARBURG, Germany 3 University Medical Center, MARBURG, Germany Introduction Adverse side effects from endocrine therapy result in decreased quality of life and non-adherence in over 50 percent of breast cancer patients. The majority of side effects occur treatment unspecific, suggesting a role of psychological factors. Methods This was a longitudinal study investigating the nature and incidence of adverse side effects and their association to treatment related expectations. Postoperative patients with hormone receptor-positive breast cancer were assessed before the start of adjuvant treatment. All patients received standardized information about endocrine therapy in addition to hospital routine care. Primary outcome was side effects at three month follow-up. Expectations about side effects were measured after standardized information provision. Secondary outcomes included quality of life (EORTC) and adherence. Results Of 331 patients screened, 138 were included at baseline and 107 provided follow-up data. After three month of endocrine therapy, patients reported high rates of adverse effects counting non-specific side effects.

S134

Hierarchical regression analysis showed that pre-treatment expectations predicted side effects after controlling for relevant medical parameters (i.e., cancer staging, type of endocrine therapy, age, menopausal status, baseline symptom-rates, depression, anxiety). The full model explained 31 percent of variance in side effect reporting (F(8,98)=5.528, p<.01), with a significant increment of 6 percent accounted for by pre-treatment expectations (ß=.26, p<.01). Conclusion Expectations are a genuine predictor of side effects in breast cancer patients. Optimizing individual expectations before the start of pharmacotherapy might be a promising strategy to prevent nocebo side effects and improve patients´ quality of life during longer term drug intake. CORRESPONDING AUTHOR: Prof. Dr. Y.N. Nestoriuc, University of Hamburg, Germany, [email protected] S446 NOCEBO RESPONDING FOLLOWING TREATMENT WITH BRAND NAME OR GENERIC TABLETS K. Faasse, KJ Petrie The University of Auckland, AUCKLAND, New Zealand Introduction: Negative perceptions of generic medicines are prevalent, and have the potential to increase treatment side effects through the influence of expectations on nocebo responding. The purpose of the study was to investigate the reported efficacy and side effects following treatment of headaches with both active ibuprofen and placebo treatments which were labelled as either brand name or generic drugs. Method: Undergraduate students experiencing regular headaches (n = 48) were recruited to participate. Using a within-subjects design, each participant received each treatment (branded active, branded placebo, generic active, generic placebo) in random order. Degree of pain before and after treatment, as well as the presence and intensity of treatment side effects, was assessed. Results: Participants reported significantly higher side effect scores after taking the generic placebo than after taking the branded placebo after controlling for baseline symptom scores and negative affect, p < .05. A similar pattern was seen with the active tablets. Side effects most commonly reported following placebo administration were drowsiness, dizziness, dry mouth, stuffy or runny nose, and nausea. Conclusion: Generic labelled tablets have a larger associated nocebo effect than brand name tablets. This has implications for the use of generic drugs worldwide, and indicates the need for interventions to improve perceptions of generic medicines. CORRESPONDING AUTHOR: Dr K. Faasse, The University of Auckland, New Zealand, [email protected] O447 ANXIETY IS ASSOCIATED WITH INCREASED MUSCULOSKELETAL PAIN SENSITIVITY DURING EXPERIMENTALLYINDUCED SYSTEMIC INFLAMMATION IN HUMANS S. Benson, A. Wegner, L. Rebernik, T. Roderigo, M. Schedlowski, S. Elsenbruch University Hospital Essen, ESSEN, Germany Introduction: Immune mechanisms may contribute to the pathophysiology of musculoskeletal pain. One experimental approach to study the effects of a systemic immune activation on pain is the administration of low-dose bacterial endotoxin (i.e., lipopolysaccharides). The aim of this study was to analyze the association between psychological state and trait variables and hyperalgesia during endotoxin-induced systemic inflammation in healthy volunteers. Methods: N=81 men and women were included in this re-analysis of three randomized, double-blind studies. All subjects received an iv injection of low-dose endotoxin. Pressure pain thresholds for the low back (erector

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

spinae muscle), calf (gastrognemicus muscle), and shoulder region (deltoid muscle) were assessed at baseline and 2h post injection using a handheld algometer, along with state anxiety (State-Trait-Anxiety-Inventory, STAI) and plasma cytokines (TNF-alpha, Interleukin (IL)-6, -8, -10). Trait anxiety and depression were measured with the Hospital Anxiety and Depression Scale (HADS). Results: Endotoxin application led to significant increases in plasma cytokines and state anxiety, and to significantly decreased pain thresholds (i.e., increased musculoskeletal pain sensitivity) (all p<0.001 vs. baseline). Whereas state anxiety and HADS-depression scores did not correlate with endotoxin-induced hyperalgesia, we observed significant associations between HADS-anxiety and increased pain sensitivity of the low back (r=.29, p<0.01) and shoulder region (r=.30, p<0.01). Conclusions: Based on an experimental model of acute systemic inflammation, our data provide further evidence for a role of immune processes in the pathophysiology of somatic pain syndromes. Higher anxiety may contribute to increased pain sensitivity during immune activation, and thus constitute a risk factor for pain chronification. CORRESPONDING AUTHOR: PD Dr. S. Benson, University Hospital Essen, Germany, [email protected] O448 ACUTE STRESS INFLUENCES MOMENTARY PAIN IN WOMEN SUFFERING FROM FIBROMYALGIA J.M. Doerr University of Marburg, MARBURG, Germany Objectives: Stress may affect experiences of pain. At a micro-longitudinal level, this association has not been studied sufficiently. In our study, we examined how acute stress may impact momentary pain in fibromyalgia, also considering neuroendocrinological changes and physical activity as possible mediators. Methods: A sample of sixteen female fibromyalgia patients (52.8±7.8yrs) reported subjective stress and pain levels six times a day for 14 consecutive days. Salivary cortisol and alpha-amylase were analyzed from samples obtained at the same time points. Patients wore wrist actigraphs for the assessment of physical activity. Results: Hierarchical multilevel models showed that momentary pain levels were best predicted by stress at the same time point (p<.001), with an additional effect of stress at the previous time point (p=.043). Both high and low mean physical activity one hour prior predicted lower momentary pain levels (p=.021), an effect which was suppressed by previous stress levels. Higher momentary cortisol (p=.049), but not alpha-amylase, levels were associated with higher momentary pain levels. Biological variables at the previous time point did not predict pain levels. A flattened alphaamylase slope (p=.003) and heightened overall cortisol output (p=.047) predicted heightened daily pain levels. Neither cortisol nor alpha-amylase were mediators in the relationship between stress and pain. Discussion: Perceived acute stress was shown to be an amplifier of momentary pain levels in patients with fibromyalgia. There is also evidence for an effect of neuroendocrine mechanisms on pain. Further, patients experienced momentary pain relief with either high or low physical activity. CORRESPONDING AUTHOR: J.M. Doerr, University of Marburg, Germany, [email protected] O449 COGNITIVE MECHANISMS OF CHANGE IN MULTIDISCIPLINARY TREATMENT OF PATIENTS WITH CHRONIC WIDE SPREAD PAIN A. De Rooij1, J. Dekker2, M. De Boer2, M. Van der Leeden1, L.D. Roorda1, M. Steultjens3 1 Reade, AMSTERDAM, Nederland 2 VU University Medical Center, AMSTERDAM, Nederland 3 Glasgow Caledonian University, GLASGOW, United Kingdom

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

Objective: To evaluate the contribution of improvement in negative emotional cognitions, active cognitive coping, and control and chronicity beliefs to the outcome of the multidisciplinary treatment in patients with chronic widespread pain (CWP). Design: Prospective cohort study. Patients: One hundred twenty diagnosed with CWP, who completed a multidisciplinary pain program. Methods: Data were used from baseline, 6 months and 18 months followup measurements. Longitudinal relationships were analyzed between changes in cognitions and outcome, using generalized estimated equations. Outcome domains included: pain, interference of pain in daily life, depression, and global perceived effect. Cognitive domains included: negative emotional cognitions, active cognitive coping and control and chronicity beliefs. Results: Improvements in negative emotional cognitions were associated with improvements in all outcome domains, in particular with improvement in interference of pain with daily life and depression (between baseline and 6 months and 6 and 18 months). Improvements in active cognitive coping were associated with improvements in interference of pain in daily life (between baseline and 6 months). Finally, improvements in control and chronicity beliefs were associated with improvements in pain and depression (between 6 and 18 months). Conclusions: Improvement in negative emotional cognitions seems to be a key mechanism of change of multidisciplinary treatment in CWP. Improvement of active cognitive coping and improvement of control and chronic timeline beliefs may constitute mechanisms of change as well, although the evidence is less strong. CORRESPONDING AUTHOR: A. De Rooij, Reade, Nederland O450 A RANDOMIZED-CONTROLLED TRIAL OF AN INTERNETBASED COGNITIVE-BEHAVIORAL INTERVENTION FOR NONSPECIFIC CHRONIC PAIN M.J. De Boer, G.J. Versteegen, K.M. Vermeulen, R. Sanderman, M.M.R.F. Struys University Medical Center Groningen, GRONINGEN, Nederland Background: Cognitive-behavioral treatment can nowadays be delivered through the internet. This form of treatment can have various advantages with regard to availability and accessibility. Previous studies showed that internet-based treatment for chronic pain is effective compared to waiting-list control groups. Methods: We conducted a randomized controlled trial comparing an internet-based cognitive-behavioral intervention with e-mail therapist contact to a face-to-face cognitive-behavioral group intervention. Of the 72 participants who were randomly assigned to an internet or a group course, 50 participants completed the intervention. Participant were assessed at base-line (T0), immediately after the 7-week course (T1) and at the booster session two months later (T2). Pain-related catastrophizing was the primary outcome measure. Pain intensity, fatigue, pain-related interference, locus of control, pain coping, global health related quality of life and medical expenses were secondary outcome variables. Results: Significant improvement was found on catastrophizing, pain coping, locus of control and aspects of global health-related quality of life in both the internet and the group course directly after the course and at the booster session. Pain intensity was improved in both courses at the booster session. At T2, improvement in catastrophizing, pain intensity, pain coping and some quality of life dimensions was significantly greater in completers of the internet course than in the group course. Furthermore, the internet course was cost-effective compared to the group course. Conclusions: We conclude that the internet-based cognitive-behavioral intervention was at least as effective as the face-to-face group intervention, and on some outcome measures appeared to be even more effective.

S135

CORRESPONDING AUTHOR: Dr. M.J. De Boer, University Medical Center Groningen, Nederland, [email protected] O451 THE EFFECTIVENESS OF THE MITCHELL METHOD RELAXATION TECHNIQUE FOR THE TREATMENT OF FIBROMYALGIA: A RCT A. Amirova1, Mark Cropley2, Alice Theadom3 1 National Medical University, ALMATY, Kazakhstan 2 School of Psychology, University of Surrey, GUILDFORD, United Kingdom 3 Institute for Applied Neuroscience Auckland University of Technology, AUCKLAND, New Zealand Introduction: The aim of the study was to evaluate the effectiveness of the Mitchell Method Relaxation Technique (MMRT), a mind-body therapy, for reducing symptoms associated with fibromyalgia. Method: A threearm randomized controlled trial was used to compare effectiveness of a self-applied MMRT (n= 67) against an unspecified attention (n = 66) and a usual care (n = 56) group. Primary outcomes were self-reported fatigue, pain and sleep, secondary outcomes included daily functioning and quality of life, depression (dysphoria) and coping, and process outcomes were anxiety and perceived stress. Outcome measures were assessed at baseline, following the intervention (four weeks) and one-month post intervention (eight weeks). Results: There was combined postintervention improvement on outcomes in the MMRT group (p<.005) with small effects for sleep problems (d=0.29, p<.05), sleep inadequacy (d=0.20, p<.05), and medium effects on fatigue (d=0.47, p<.05). The onemonth follow-up fatigue score was not significantly different to the postintervention score (p=.25) indicating short-term sustainability of the treatment effect. The effects on sleep problems and sleep inadequacy was not sustained. Pre-post change in pain levels did not significantly vary across three groups (p=.59) yet medium effects of the MMRT was present (d=.54). There was a dose effect, as pain significantly decreased when the MMRT was practiced three times a week (p<.005). Conclusion: The MMRT is an effective management for pain, sleep problems, and fatigue. High relative risk reduction rates for fatigue (37%) and pain (42.8%) support clinical significance and recommendation of the MMRT as an adjuvant treatment for people with fibromyalgia. CORRESPONDING AUTHOR: A. Amirova, National Medical University, Kazakhstan, [email protected] O452 USING ILLNESS PERCEPTIONS TO CLUSTER CHRONIC PAIN PATIENTS PARTICIPATING IN A CHRONIC PAIN SELFMANAGEMENT PROGRAM L. Frostholm, M.M. mimi Mehlsen Aarhus University Hospital, AARHUS N, Denmark Introduction: This study examines whether Leventhal’s common-sense model may usefully help classify chronic pain patients based on their illness representations, since these have been shown to influence outcomes in chronic pain conditions. Methods: 424 chronic pain patients enrolled in a RCT on the effect of the chronic pain self-management program (CPSMP) completed a revised version of the Illness Perception Questionnaire (IPQ-R). Cluster analyses using established guidelines were carried out and associations to selfreported pain, physical symptoms (SCL-12), illness worry (Whiteley-7), and anxiety (SCL-ANX) and depression (SCL-DEP) six months later were examined. Results: The cluster analysis resulted in 3 meaningful clusters based on patients’ perceptions: 1) patients feeling very certain about the nature of their chronic pain condition, 2) patients reporting very high levels of uncertainty and moderate levels of perceived consequences and emotional representations, (3) and patients reporting very high levels of consequences and emotional representations as well as a cyclical timeline

S136

perspective. The three groups were comparable with respect to long timeline perspective and perceived control. The groups differed in their levels of self-reported physical symptoms (F=6.13, p=0.003), anxiety (F=18.36, p<0.0001), depression (F=31,79, p<0.0001), and illness worry(F=10.35, p<0.0001) but not in self-reported pain on a VAS scale (F=0.43, p=0.65). Conclusions: The preliminary results from this study support the potential of using illness perceptions to cluster chronic pain patients. Further analysis will show whether these groups may help clarify outcome of participation in the SPSMP. In the longer perspective, this study may help customize psychological treatments for chronic pain. CORRESPONDING AUTHOR: phd L. Frostholm, Aarhus University Hospital, Denmark, [email protected] O453 THE ASSOCIATIONS BETWEEN SUPPORTIVE LEADERSHIP BEHAVIOR AND THE COSTS OF ABSENTEEISM AND PRESENTEEISM J.S. Schmid1, M.N. Jarczok1, D. Sonntag1, J.E. Fischer1, R.M. Herr2, B. Schmidt1 1 University of Heidelberg, MANNHEIM, Germany 2 University of Amsterdam, AMSTERDAM, Nederland Introduction: Recent research has identified supportive leadership behavior (SLB), which denotes supportive involvement and fairness of a supervisor, as a determinant of self-rated health. Thus, better SLB should be associated with lower presenteeism (attending work while being sick) and absenteeism (being absent from scheduled work due to sickness) and, in the end, reduce cost-of-illness. However evidence on these association as well as estimates for such indirect costs are missing. This study investigates associations between SLB, presenteeism and absenteeism and quantifies related indirect costs. Methods: Cross-sectional data from a German industrial sample (n=4395 response rate 50%) were used. SLB was measured with a validated scale consisting of 4 items. The dependent variables, self-reported presenteeism and absenteeism, were dichotomized (>5 days). Logistic regressions were performed and adjusted for various confounders. SLB was categorized into tertiles and costs for presenteeism and absenteeism were calculated by estimating lower and upper boundaries for each tertile. Results: The lowest SLB tertile, compared to the highest, was associated with higher presenteeism (OR 2.33, CI [1.96;2.78]) and absenteeism (OR 1.84, CI [1.57;2.17]). Compared to high SLB, the costs of low SLB for absenteeism are between 801.90 € and 2,756.82 € higher per person and year. For presenteeism this difference ranges between 91.21 € and 734.37 € per person and year. Conclusions: SLB might be a potential leverage to reduce absenteeism and presenteeism and the associated indirect costs. Consequently, supervisors have the possibility to contribute substantially to the health and productivity of employees.

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

occupational setting itself can significantly contribute to this development. Recent research has identified supportive leadership behavior (SLB), which denotes supportive involvement and fairness of a supervisor, as a determinant of self-rated health, but evidence linking SLB to depression is missing. This study aimed to identify associations between SLB and depression, and further aimed to determine if SLB predicts depression independent of other established work stress factors (i.e., effort-reward imbalance, ERI). Methods: Cross-sectional data from a German industrial sample (n=4395 response rate 50%) were used. SLB was measured with a validated scale consisting of 4 items (Alpha=.92). Depression was measured with the mental health inventory 5 (MHI-5). Linear and logistic regression models were calculated with potential confounders (sex, age, job position, smoking, drinking, physical activity, and ERI). Results: Better SLB was positively associated with lower depression ratings (ß=.29; p=0.00, age and sex adjusted), albeit attenuated when adding work stress and additional confounders (SES, ERI, health behavior) to the model (ß=0.14;p=0.00). The same pattern emerged when depression was dichotomized on the basis of a clinical cutoff (OR=0.74; 95% CI 0.69-0.80, in the fully adjusted model). Conclusion: Poor SLS might be a potential risk factor for depression in working adults. These findings suggest that leadership behavior may become a relevant target for future interventions studies aiming to improve health in the work force. CORRESPONDING AUTHOR: Dr. B.S. Schmidt, Heidelberg University, Germany, [email protected] O455 ORGANIZATIONAL JUSTICE IS ASSOCIATED WITH TINNITUS FINDINGS FROM A CROSS-SECTIONAL STUDY R.M. Herr1, A. Loerbroks2, J.A. Bosch3, M. Seegel1, M. Schneider1, B. Schmidt1 1 Heidelberg University, MANNHEIM, Germany 2 University of Düsseldorf, DÜSSELDORF, Germany 3 University of Amsterdam, AMSTERDAM, The Netherlands

O454 SUPPORTIVE LEADERSHIP AND DEPRESSION IN APPARENTLY HEALTHY WORKING ADULTS B.S. Schmidt1, M. Seegel1, M.N. Jarczok1, R.M. Herr1, A. Loerbroks2, J.A. Bosch3, J.E. Fischer1 1 Heidelberg University, MANNHEIM, Germany 2 University of Düsseldorf, Düsseldorf, Germany 3 University of Amsterdam, Amsterdam, Nederland

Introduction: Tinnitus is the perception of a noise in the human ear without an external source. Research has identified work-related stress, and its potential consequences such as depression and burnout, as a risk factor. Perceived fairness at work (organizational justice) is a significant source of occupational distress and predictive of depression and burnout. Therefore, we aimed to determine the association of organizational justice with tinnitus symptoms, and additionally explored a mediating role of depression and burnout. Methods: Cross-sectional data from a sample of 1,632 employees (response rate 40.8%; 51.6% male; mean age = 41.3 years with SD = 9.4) were used. Tinnitus was assed by self-report (self- or physiciandiagnosed; n = 207; 13,9%). Organizational justice and its subcomponents (interactional and procedural justice), burnout, and depression were measured by validated questionnaires. For analyses, scales were centered by z-transformation. Results: Perceived organizational justice was inversely related to tinnitus. This association was independent of demographics, job characteristics (including potential noise exposure), lifestyle, and body mass index (OR = 0.754; 95% C.I. = 0.649 to 0.876). Adjusted mediation analyses revealed a potential mediation by burnout (95% C.I. indirect effect: -0.188 to -0.066), and depression (95% C.I. indirect effect: -0.160 to -0.043). Conclusion: Organizational justice appeared to be strongly related to tinnitus. As depression and burnout seemed to mediate this association, these findings suggest several potential targets for intervention studies, starting with organizational justice.

Introduction: The prevalence of depression is rising in Western countries. This development impacts the work force and, in fact, it is the

CORRESPONDING AUTHOR: R.M. Herr, Heidelberg University, Germany, [email protected]

CORRESPONDING AUTHOR: J.S. Schmid, University of Heidelberg, Germany, [email protected]

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

O456 JOB STRAIN, EFFORT-REWARD IMBALANCE AND DEPRESSIVE SYMPTOMS: A THREE-WAVE LONGITUDINAL STUDY IN GERMAN PHYSICIANS J.L. Li1, M. Weigl2, A. Loerbroks1, C. Degen1, A. Müller1, J. Glaser3, A. Angerer1 1 University of Düsseldorf, DÜSSELDORF, Germany 2 Ludwig-Maximilians-University, MUNICH, Germany 3 University of Innsbruck, INNSBRUCK, Austria Introduction: Cumulative evidence on work stress and depression has been available, but very few studies were able to draw on data from multiple repeated measurements and made full use of this data structure. We aim to examine the impact of work stress on depressive symptoms in a sample of German physicians, a high risk group for mental disorders, using multilevel techniques. Method: This is a three-wave longitudinal study in 507 German physicians. The measure of work stress was based on the job strain model and the effort-reward imbalance model at all three waves, while the depressive symptoms were assessed with the State-Trait Depression Scales contemporaneously. Multilevel linear regression was employed to quantify longitudinal associations between job strain, effort-reward imbalance and depressive symptoms over three waves. Results: When using all three-wave full data, it was found that both job strain and effort-reward imbalance were significantly associated with depressive symptoms in the mutually adjusted statistical model. However, effort-reward imbalance (coefficient = 2.01, P < 0.0001) appeared to have more explanatory power as a model of work stress than job strain (coefficient = 0.71, P < 0.0048). Conclusions: In future, multilevel modeling, which provides an appropriate approach for longitudinal study with repeated measurements, deserves more attention in occupational epidemiological research. According to our findings, it is suggested that interventions of work stress reduction, especially improving the reciprocity between effort and reward at work, may be effective in preventing depression in physicians. CORRESPONDING AUTHOR: J.L. Li, University of Düsseldorf, Germany, [email protected] O457 THE ASSOCIATION OF EFFORT-REWARD-IMBALANCE AND ASTHMA: FINDINGS FROM TWO CROSS-SECTIONAL STUDIES A. Loerbroks1, R.M. Herr2, J. Li1, J.A. Bosch3, M. Seegel2, M. Schneider2, P. Angerer1, B. Schmidt2 1 University of Düsseldorf, DÜSSELDORF, Germany 2 Heidelberg University, MANNHEIM, Germany 3 University of Amsterdam, AMSTERDAM, Nederland Introduction: While work stress has been found to predict a broad range of health outcomes, evidence for an association with asthma is sparse and inconclusive. Moreover, this limited body of work is characterized by the use of non-validated work stress measures. It was the aim of this study to examine the association of asthma with work stress based on the wellestablished effort-reward imbalance (ERI) model. Methods: We drew on data from two cross-sectional studies. Study 1 was conducted in 2013 in a large pharmaceutical company (n=1,632). Study 2 was based on the 2011 wave of the population-based German SocioEconomic Panel (n=8,388). ERI was assessed by using validated scales. Asthma was determined by self-reported physician diagnoses. Associations between ERI components (z-scores) and asthma were estimated by logistic regression-based multivariate odds ratios (ORs) with 95% confidence intervals (CIs). Results: In both studies, a one standard deviation increase of the ERI score (reflecting higher work stress) was associated with elevated odds of asthma (Study 1: OR=1.48, 95%CI=1.13-1.95, and Study 2: OR=1.22, 95%CI=1.08-1.38). Analyses of the separate ERI components suggested

S137

modest associations of effort and reward with asthma in both studies (Study 1: OR effort =1.25, 95%CI=0.95-1.64, OR reward =0.67, 95%CI=0.51-0.87, and Study 2: OR effort =1.15, 95%CI=1.01-1.31, OR reward =0.87, 95%CI=0.79-0.97). Positive associations between overcomittment and asthma were only found in Study 1 (OR=1.32, 95%CI=1.01-1.72), but not in Study 2 (OR=0.99, 95%CI=0.87-1.12). Conclusions: This is the first study to show that work stress as measured by an established operationalization is associated with asthma. CORRESPONDING AUTHOR: Dr. A.L. Loerbroks, University of Düsseldorf, Germany, [email protected] O458 IMPACT OF ORGANIZATIONAL POLICIES AND PRACTICES ON WORKPLACE INJURIES IN A HOSPITAL SETTING T.H. Tveito1, G. Sembajwe2, L.I. Boden3, J.T. Dennerlein4, G.R. Wagner5, C. Kenwood6, A.M. Stoddard6, S.E. Reme1, K. Hopcia7, D. Hashimoto8, W.S. Shaw9, G. Sorensen5 1 Uni Research/Uni Health, BERGEN, Norway 2 The City University of New York, NEW YORK CITY, United States of America 3 Boston University School of Public Health, BOSTON, United States of America 4 Northeastern University, BOSTON, United States of America 5 Harvard School of Public Health, BOSTON, United States of America 6 New England Research Institutes, WATERTOWN, United States of America 7 University of Illinois at Chicago, CHICAGO, United States of America 8 Partners Health Care, BOSTON, United States of America 9 Liberty Mutual Research Institute for Safety, HOPKINTON, United States of America Introduction This study aimed to assess the relationships between perceptions of organizational practices and policies, social support, and injury rates among workers in hospital units. Methods A sample of 1230 hospital workers provided survey data on organizational practices and policies (OPP), job flexibility, and social support. Demographic data and unit injury rates were collected from the hospitals’ administrative databases. Results Injury rates were lower in units where workers reported higher OPP scores and high social support. These relationships were mainly observed among the registered nurses (RNs). Perceptions of job flexibility, social support, and OPPs were not correlated between RNs and personal care assistants (PCAs). The RNs perceived co-worker support and OPPs as less satisfactory than the PCAs. Conclusions Employers aiming to reduce injuries in hospitals could focus on good OPPs and a supportive work environment. CORRESPONDING AUTHOR: H. Tveito, Uni Research/Uni Health, Norway, [email protected] O459 ACTIVITY LEVELS OF INDIVIDUALS WITH ARTHRITIS ARE CHARACTERIZED BY DIFFERING PAIN-RELATED COGNITIONS AND OUTCOME EXPECTATIONS N.C. Gyurcsik, M.A. Cary, J.D. Sessford, L.R. Brawley, K.S. Spink University of Saskatchewan, SASKATOON, Canada INTRODUCTION. Physical activity (PA) is an effective arthritis selfmanagement strategy when performed at sufficient levels (recommended=150+ mins/week). Sufficiently active (S-PA) people seem undaunted by arthritis pain. However, is this the case for the insufficiently active (InS-PA) and completely inactive? The individual difference factors of arthritis pain acceptance and pain anxiety have been studied in S-PA people, as have negative disease outcome expectations (OEs: e.g., postPA joint stiffness), but are unexamined in other adult PA groups. This

S138

study examined if arthritis-related pain intensity, pain acceptance, pain anxiety, and negative OEs (OEs-distress; OEs-likelihood) differed among S-PA, InS-PA, and inactives. METHODS. North American participants completed online measures of two-week PA, pain-related social cognitions, and negative OEs. S-PA (n=83), InS-PA (n=89), and inactive (n=103) participants were identified. A significant MANOVA of group differences for assessed variables was detected (p<.001). Follow-up analyses revealed that the S-PA group reported significantly greater pain acceptance, less pain anxiety, and less distressing negative OEs compared to the InS-PA and inactives (p‘s<.01). S-PA and InS-PA did not differ in pain intensity and OEs-likelihood (p‘s>07), whereas S-PA reported less pain and higher OEs-likelihood than inactives (p<.01). Inactives reported more pain intensity and anxiety, less pain acceptance, greater negative OEs-likelihood and OEs-distress than the InS-PA (p‘s<.01). CONCLUSIONS. These results capture the first comparison of all three PA groups relative to arthritis pain-related cognitions and OEs and suggest individual differences. Identifying differences in psychological characteristics may aid understanding of their influence on differential adherence between groups and offer insight about tailoring arthritis-PA interventions. CORRESPONDING AUTHOR: Dr. C. Gyurcsik, University of Saskatchewan, Canada, [email protected] O460 VOCATIONAL SCHOOL TEACHERS' CURRENT PRACTICES AND ACCEPTABILITY OF INTERVENTION STRATEGIES TO REDUCE SITTING IN CLASSROOM H.P. Laine1, S.H. Hynynen2, V.A-S. Araujo-Soares3, P.J. Jallinoja4, A.H. Haukkala2, T.L Lintunen5, N.H. Hankonen2 1 University of Helsinki, HELSINKI, Finland 2 Department of Social Research, University of Helsinki, HELSINKI, Finland 3 Institute of Health and Society, Newcastle University, NEWCASTLE, United Kingdom 4 National Consumer Research Centre, HELSINKI, Finland 5 Department of Sport Sciences, University of Jyväskylä, JYVÄSKYLÄ, Finland Introduction: As school days include long periods of uninterrupted sitting, interventions aimed to reduce sitting among youth could be directed to teachers controlling the classroom behavior. The aim of the study was to examine 1) what strategies are used in vocational schools to reduce sitting and 2) what kind of attitudes teachers have toward potential intervention strategies to reduce sitting 3) whether teacher experience and gender are associated with these attitudes. Methods: Altogether 196 teachers from 23 Finnish vocational schools replied to an electronic survey, including questions about current strategies used to interrupt sitting, equipment available, and attitudes to intervention strategies. Findings: 81% of the core subject teachers and 93% of the vocational teachers found it important that their school aims at decreasing uninterrupted sitting. However, 22% of core subject and 44% of vocational teachers included activities to reduce sitting in their classes. Of all teachers, 89% consider using at least one of the six candidate intervention strategies. Environmental changes such as replacing chairs were less acceptable. Experience (years) as a teacher was not associated with current strategy use or attitudes. Compared to men, women had more positive attitudes regarding potential intervention strategies (p =.009) among vocational teachers. Conclutions: A limited number of Finnish vocational school teachers are effectively using strategies to decrease uninterrupted sitting. However, teachers find the proposed intervention techniques acceptable and feasible to be implemented in their schools. CORRESPONDING AUTHOR: M.Soc.Sci H.P. Laine, University of Helsinki, Finland, [email protected]

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

O461 FEASIBILITY AND ACCEPTABILITY OF AN mHEALTH INTERVENTION TO PROMOTE PHYSICAL ACTIVITY: THE PROPELS STUDY K.L. Morton1, H. Eborall2, S. Sutton1, W. Hardeman1 1 University of Cambridge, CAMBRIDGE, United Kingdom 2 University of Leicester, LEICESTER, United Kingdom Introduction: Mobile technologies for health (mHealth) represent a promising strategy for reducing type 2 diabetes (T2D) risk. The PROPELS trial investigates whether group-based education supplemented by a textmessaging intervention is effective in promoting physical activity (PA) among people at high risk of T2D. This study aimed to assess the feasibility and acceptability of the text messaging intervention prior to trial evaluation. Method: We conducted three formative focus groups (phase 1; N=15) to determine perceptions about the utility of mHealth for increasing PA to inform intervention content. We conducted four pre-testing focus groups (phase 2; N=20) to refine intervention content, and a pilot study (phase 3; N=13) to assess feasibility and acceptability. Participants, aged 39-78 years, had recently participated in group-based education for people at T2D risk. Results: Phase 1: Emergent themes included: benefits of mHealth (e.g., providing support and immediate feedback capabilities) and factors inhibiting effectiveness (e.g., unrealistic expectations about mHealth and impersonal messages). Phase 2 highlighted the need for tailored SMS content and two-way SMS interaction, and informed language and content of messages. Phase 3: Participants identified no particular difficulties in using text-messaging regardless of SMS experience and it appeared to fit well with everyday routines. Conclusions: The iterative mHealth development and piloting process was invaluable for informing text-messaging content, frequency and language. Findings also indicated that text-messaging is feasible and acceptable to older adults at high T2D risk as a means to provide ongoing support for PA. CORRESPONDING AUTHOR: Dr L. Morton, University of Cambridge, United Kingdom, [email protected] O462 YOUNG ADULT'S PREFERRED THEORY-BASED BEHAVIOURAL AND TECHNOLOGICAL FEATURES IN A MOBILE PHYSICAL ACTIVITY APPLICATION S.J. te Velde, A. Middelweerd, S. Khodabaks, L. Belmont, J. Brug VU university medical center, EMGO Institute for Health and Care Research, AMSTERDAM, Nederland Introduction: Social influences and self-efficacy are important determinants of physical activity (PA) that should be targeted in mobile PA interventions. Mobile applications (apps) are promising, but recent reviews show that PA apps hardly include theory-based features. Therefore, this study will identify behaviour change techniques (BCTs) and technological features that target social determinants and/or self-efficacy and assess preferences of its use in PA apps in young adults. Methods: Young adults (n~100) will complete on-line questions assessing PA, social influences and self-efficacy using validated questionnaires. Furthermore, preferences for PA promoting BCTs (e.g. monitoring, goal setting) and performance expectancy, effort expectancy, social influence, facilitating conditions and perceived playfulness, based on the Unified Theory of Acceptance and Use of Technology. Descriptive statistics will describe which behavioural and technological features young adults perceive as important, interesting and useful for PA promotion. Associations of PA, social determinants and self-efficacy with the preferences will be estimated by regression analyses.

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

Results: Results (available in May/June 2014) will describe the behavioural and technological preferences of young adults, e.g. proportions of participants that prefer self-efficacy promoting features (e.g. goal setting, receiving feedback, persuasion) features in an app. Regression coefficients will describe the strength of PA, social determinants and self-efficacy with the specific preferences. Conclusion: Results from the survey will directly be used in the development of a new theory-based mobile intervention called active2gether, aiming to empower young adults to be more physically active. CORRESPONDING AUTHOR: dr. S.J. te Velde, VU university medical center, EMGO Institute for Health and Care Research, Nederland, [email protected] O463 DOES CARDIAC REHABILITATION LEAD TO CHANGES IN PHYSICAL ACTIVITY HABITS? A SYSTEMATIC REVIEW N. ter Hoeve1, H.J.G drs Van den Berg-Emons2, R.T. drs. Van Domburg2, H.J. prof. Stam2, M. m.d. Sunamura3, B.M.A. drs. Huisstede2 1 Capri Cardiac Rehabilitation\Erasmus MC, ROTTERDAM, Nederland 2 Erasmus MC, ROTTERDAM, Nederland 3 Capri Cardiac Rehabilitation, ROTTERDAM, Nederland Aims: While improvements in physical activity levels are known to have several health benefits for patients with acute coronary syndrome (ACS) and are therefore animportant goal of cardiac rehabilitation (CR), it is not clear whether current CR programmes provide sufficient long-lasting changes in physical activity habits. Therefore, the aim of this study was to systematically review the scientific literature regarding the short-term and long-term effects of CR on physical activity levels in patients with ACS. Methods: PubMed, Embase, Cinahl and Pedro were systematically searched for relevant randomized controlled trials (RCTs). Two reviewers independently selected articles, extracted data on outcome measures for physical activity levels and assessed the methodological quality of RCTs. A best-evidence synthesis was used to summarize the results. Results: A total of 28 RCTs were included. In the short term (0-6 months after completion of CR) we found strong evidence for higher physical activity levels in CR programme participants than in non-participants. Evidence for long-term maintenance (≥6 months after completion of CR) of these effects was conflicting. At most, limited evidence was found in favour of one specific type of CR. Conclusion: Current CR programmes for patients with ACS are effective in improving short-term physical activity habits. Despite the well-known benefits of being active, it would appear that current standard CR programmes are not producing the desired long-term effects. Since study conclusions are based mainly on self-report measures, which are known to have restricted validity, future research should use objective instruments with a high validity. CORRESPONDING AUTHOR: N. ter Hoeve, Capri Cardiac Rehabilitation\Erasmus MC, Nederland, [email protected] O464 EFFECTS OF A 12-WEEK ENDURANCE TRAINING PROGRAM ON THE SUBJECTIVE AND THE PHYSIOLOGICAL STRESS RESPONSE S.K. Klaperski, Reinhard Fuchs University Freiburg, FREIBURG, Germany Introduction: Stress is a major threat for health. Previous research suggested that physical exercise might reduce this threat by leading to biological adaptations which contribute to a reduced stress response

S139

(known as Cross-Stressor Adaptation hypothesis; Sothmann, 2006). The present study is the first to experimentally examine the effects of an endurance exercise program on the cardiovascular and endocrine stress response to a standardized psychosocial stressor. Methods: We randomly assigned 149 healthy, inactive male employees to a 12-week exercise training, relaxation training, or a waitlist control group; the final sample consisted of 96 subjects. Before and after intervention we assessed the groups’ physical fitness (lactate testing) and subjective stress responses, and we compared their physiological stress responses to the Trier Social Stress Test for Groups in terms of salivary free cortisol, heart rate (HR) and heart rate variability (HRV). Results: Pre-post intervention comparisons showed that the physiological stress reactivity changed differently between the groups (p < .05). Only the exercise group showed a reduced cortisol (p < .01, d = .64) and HR response (p < .001, d = .77), and an increase in HRV (p < .001, d = .54) post intervention. Furthermore, only endurance training reduced subjective stress reactivity (p < .01, d = .34) and increased fitness (p = .02, d = .31). Conclusion: A 12-week exercise program led to a more beneficial stress response than no training or relaxation training. Thus, our findings suggest that the Cross-Stressor Adaptation hypothesis is valid for cardiovascular and endocrine stress reactivity. CORRESPONDING AUTHOR: Dr. S.K. Klaperski, University Freiburg, Germany, [email protected] O465 A LADDER LOGIC MODEL TO INFORM THE DESIGN OF INCENTIVE TRIALS FOR WOMEN AROUND CHILDBIRTH B. Hoddinott1, H. Morgan2, G. Thomson3, N. Crossland3, S.U. Dombrowski1 1 University of Stirling, STIRLING, United Kingdom 2 University of Aberdeen, ABERDEEN, United Kingdom 3 University of Central Lancashire, PRESTON, United Kingdom Introduction. Smoking in pregnancy and not breastfeeding are socially patterned behaviours. Two mother-and-baby groups in disadvantaged UK areas were co-applicants in a mixed methods study to inform incentive trial design, as co-designing interventions with service users is recommended. Methods. Three systematic reviews. 74 qualitative interviews and 16 focus groups with 88 pregnant women/recent mothers/partners; 53 service providers; 24 experts/decision makers. 3 conference interactive discussions (63 attendees). Surveys of incentive acceptability (1144 general public; 497 health professionals) and a discrete choice experiment (320 women with a smoking history) were undertaken. Iterative data collection and analysis resulted in a logic model which was adapted to engage service user perspectives. Results. The logic model applies the metaphor of a ladder to understand how key components of incentive programmes (rungs) are perceived in relation to life course and context facilitators (rungs) and barriers (missing or damaged rungs) and behaviour change theory. Tailored programmes delivered by specialist teams to enable women to bolster their individual capabilities and family/social network support were important. Monitoring to set short term goals, visual outcome verification and multiple community locations were valued. Rigid, prescriptive interventions which place the onus on the woman to behave ‘correctly’ may risk women feeling judged and pressurised. To avoid losing face, women may disengage with services. Conclusion. An incentive ladder logic model has face validity with service users for the design of trial ‘rungs’ that fit with everyday life ‘rungs’ as incentives alone were considered unlikely to succeed in either reach or effectiveness. CORRESPONDING AUTHOR: Prof. M. Hoddinott, University of Stirling, United Kingdom, [email protected]

S140

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

O466 RISKS VERSUS BENEFITS OF MEDICATION USE DURING PREGNANCY: WHAT DO WOMEN PERCEIVE? B. Mulder1, C.C.M. Schuiling-Veninga1, L.P. Morsink2, M.J. Bijlsma1, E. Van Puijenbroek3, J.G. Aarnoudse4, E. Hak1, T.W. De Vries2 1 Universitiy of Groningen, GRONINGEN, Nederland 2 Medical Center Leeuwarden, LEEUWARDEN, The Netherlands 3 Netherlands Pharmacovigilance centre Lareb, 'S-HERTOGENBOSCH, Nederland 4 University Medical Center Groningen, GRONINGEN, Nederland

Conclusion: Consistent with health behaviour change theories, beliefs/ expectations and knowledge regarding GWG are likely to influence weight gain over the course of pregnancy. Women tend to expect they will gain more weight than they desire, and it is possible these selfdefeating expectations are more closely linked to future weight gain during pregnancy. These findings have implications for interventions designed to prevent GWG.

Introduction: Understanding risk perceptions is essential in designing good risk communication strategies. The aim of this study was to evaluate the perception of risks, benefits and major concerns regarding medication use during pregnancy. Methods: Questionnaires were handed out to all pregnant women who attended the hospital for first and second line care facilities. Patients were asked to score the risks (Risk-score(RS)) and benefits (Benefit-score(BS)) of several drug groups on a scale from 1-7. In addition questions were asked about the level of concern of different unfortunate events resulting from medication use. Results: (Results are preliminary) A total of 118 eligible women completed the questionnaire (response rate 77%). More than 80% of the women used medication during pregnancy. Pregnant women were most concerned about having a child with a congenital birth defect (38.1%), a miscarriage (36.0%) or having a child with an allergic disease (25.7%), respectively as a result of drug use. Paracetamol(RS=2.0,BS=4.7), antacids(RS=2.8,BS=4.2) and antibiotics(RS3.3,BS=4.6) were perceived relatively low in risk and high in benefit. Sedatives(RS=5.2,BS=3.9) and NSAIDS(RS=4.7,BS=3.2) were perceived relatively high in risk and low in benefit. Conclusion: This is the first study that shows the different concerns that pregnant women have regarding medication use during pregnancy. Though it has been reported that women overestimate teratogenic risk of medication use, this study showed that most of the drugs were perceived relatively low in risk and high in benefit. Health care providers in first and second line care obstetric facilities can take this into account when counselling pregnant women.

O468 HEALTHY EATING AND LIFESTYLE IN PREGNANCY: THE HELP TRIAL MAIN RESULTS A. Simpson1, Elinor John1, Dunla Cassidy1, Rebecca Playle1, Karen Jewell1, Helen Stanton1, David Cohen2, The HELP study team1 1 Cardiff University, CARDIFF, United Kingdom 2 University of South Wales, WALES, United Kingdom

CORRESPONDING AUTHOR: BSc B. Mulder, Universitiy of Groningen, Nederland, [email protected] O467 EXPECTATIONS, DESIRES AND KNOWLEDGE OF GESTATIONAL WEIGHT GAIN S. McPhie, H. Skouteris, B. Hill Deakin University, MELBOURNE, Australia Aim: To explore the relationship between expectations, desires and knowledge regarding gestational weight gain (GWG) on GWG status. Method: 166 pregnant women were tracked during pregnancy. At 16-18 weeks gestation participants provided information on how much weight they expected, and would like (desired) to gain over the course of their pregnancy. For knowledge of GWG, women were asked about the minimum and maximum amount of weight they thought they should gain during pregnancy for their baby’s health. This was then compared to the Institute of Medicine’s (IOM) 2009 GWG guidelines for each prepregnancy BMI category. Findings: On average women expected they would gain 3.55kg (SD = 2.66) more than they desired. The findings of hierarchical regressions revealed that expectations about how much weight women would gain during pregnancy (as reported early on in pregnancy) explained a greater proportion of the variance of GWG (in late pregnancy, ie., at 36 weeks gestation) than desires about GWG. The majority of women with excessive GWG overestimated both the maximum and minimum amount of weight they needed to gain during pregnancy to have a healthy baby.

CORRESPONDING AUTHOR: Dr E. McPhie, Deakin University, Australia, [email protected]

Introduction: Maternal obesity is associated with pregnancy and birth complications. Excess weight gain during pregnancy can lead to longterm maternal obesity and is linked with child obesity. This study aims to test a theory based weight management intervention for obese pregnant women, which targets physical activity and healthy eating. Methods: This cluster RCT was designed following a successful pilot. 598 pregnant women were recruited across 20 maternity units. Women in the intervention group were invited to attend a weekly weight management group. Control participants received usual care. Women were followed up at 36 weeks gestation and 6 weeks, 6 months and 1 year postpartum. The primary outcome is BMI at 1 year postpartum. Secondary outcomes include pregnancy weight gain, quality of life, mental health, waist-hip ratio, child weight, diet, physical activity, pregnancy and birth complications, breast feeding, social support, self- regulation and self-efficacy. A health economic and process evaluation were also conducted. Results: Baseline characteristics were similar between groups. Most women thought that controlling their weight was important and around half were confident that they could control their weight during the pregnancy. Over a third of the women weighed themselves daily or weekly. Follow-up data are currently being analysed and results for all outcomes will be complete by March 2014. Conclusions: This large trial is in an under-researched area and is highly topical. If successful, this intervention could reduce health risks as well as impacting the wider family and lead to reduced healthcare costs. CORRESPONDING AUTHOR: Dr A. Simpson, Cardiff University, United Kingdom, [email protected] O469 WEIGHT MANAGEMENT IN PREGNANCY: EXPERIENCES OF ‘HEALTHY EATING AND LIFESTYLE IN PREGNANCY (HELP)' TRIAL D.M. Cassidy1, L. Copeland1, E. John1, Y. Moriarty1, H. Stanton1, K. Jewell2, S.A. Simpson1 1 Cardiff University, CARDIFF, United Kingdom 2 Cardiff & Vale University Health Board, NHS Wales, CARDIFF, United Kingdom Introduction: Around 1 in 5 pregnant women in the UK are obese. Obesity is linked to poor health and pregnancy complications. We developed and tested in a cluster RCT a theory-driven group-based intervention for obese pregnant women, facilitated by midwives and slimming world consultants. It was designed to support women during pregnancy to moderate weight gain and reduce BMI one year after birth, in addition to targeting secondary health and wellbeing outcomes. This paper aims to explore participants’ and facilitators’ experiences of engaging in the ‘HELP’ trial and provide insight into study processes.

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

Methods: 62 participants took part in semi-structured telephone interviews. Participants were sampled across 20 centres; by treatment, weight change and group attendance; with data collected at 6 months and 1 year post-birth. 3 focus groups were conducted with intervention facilitators across 10 centres, post-intervention delivery. Interview and focus group data were analysed using thematic analysis. Results: Accounts offered insights into the group environment, the importance of social support and the value of facilitators’ involvement. Reasons for compliance were highlighted, with discussion on aspects that encouraged women to change diet and activity behaviours, barriers to engaging with the intervention, and researcher influence in control sites. Participants and facilitators emphasised the benefits of weight management group support in pregnancy. Conclusions: Engaging in the intervention can help women to adopt healthier habits leading to avoidance of excess weight gain during pregnancy and better outcomes for mother, baby and the wider family. Accounts will facilitate interpretation of the main study findings. CORRESPONDING AUTHOR: D.M Cassidy, Cardiff University, United Kingdom, [email protected] O470 RISK FACTORS FOR THREATENED MISCARRIAGE: A NESTED CASE-CONTROL STUDY C. Zhu1, H. Yang2, Q. Geng3, L. Chen3 1 Guangzhou Medical University, GUANGDONG, China 2 Guangzhou Women and Children’s Medical Centre, Guangzhou Medical University, GUANGZHOU, China 3 Bureau of Health Care, Health Department of Guangdong Province, GUANGZHOU, China Objective: To examine the association between behavioural and lifestyle risk factors, and risk of threatened miscarriage. Methods: A hospital-based nested case-control study was undertaken on 611 women presenting with threatened miscarriage and 1222 asymptomatic age-matched controls, between July 2011 and December 2012 in Guangzhou. Pregnant Women were questioned about sociodemographic, behavioural, lifestyle and other factors in their most recent pregnancy. Main outcome measured was threatened miscarriage. Results: After adjustment for confounding, the following were independently associated with increased risk: exposure to computer radiation, eating spicy food prior to conception, exposure to hormone drugs during pregnancy, eating fried food during pregnancy, infectious disease during pregnancy and eating dairy products during pregnancy. Eating high-fat food during pregnancy was associated with reduced risk. We did not confirm an association with caffeine consumption, smoking, alcohol consumption, or stress. Conclusions: The results confirm that advice to encourage a healthy diet and behavioral habits might help women in early pregnancy (or planning a pregnancy) reduce their risk of threatened miscarriage. Findings regarding exposed to hormone drugs during pregnancy, eating spicy food prior to conception, exposed to computer radiation, eating fried food during pregnancy, infectious disease during pregnancy and eating dairy products during pregnancy, and the reduced risk associated with eating high-fat food during pregnancy, are noteworthy. CORRESPONDING AUTHOR: C. Zhu, Guangzhou Medical University, China, [email protected] O471 AFFORDABILITY OF ALCOHOL AND EDUCATIONAL DISPARITIES IN ALCOHOL-RELATED MORTALITY: A TIME-SERIES ANALYSIS K. Herttua1, P. Mäkelä2, P. Martikainen1 1 University of Helsinki, HELSINKI, Finland 2 National Institute for Health and Welfare, HELSINKI, Finland

S141

Introduction. Prices of alcohol and income tend to influence how much people buy and consume alcohol. They may be combined into one measure, affordability of alcohol. Research on the association between affordability of alcohol and alcohol-related harm, and how this association varies across sub-populations is scarce or non-existent We estimated effects of affordability of alcohol on alcohol-related mortality according to gender and education. Methods. Time series analysis was applied to the quarter-annually aggregations of deaths and affordability of alcohol in Finland for the period 1988-2007 to assess the association between affordability of alcohol and alcohol-related mortality. Mortality was defined using information on both underlying and contributory causes of death. To measure affordability of alcohol, we calculated affordability index using information on personal income and alcohol prices. Analyses were carried out for men and women aged 30-79 years in four educational categories. Results. During the 20-year follow-up about 62,500 persons died from alcohol-related causes. Alcohol-related mortality rate was more than 3fold among those with only a basic education as compared with those with tertiary education. Among men and women with only a basic education, an increase (decrease) of 1% in the affordability of alcohol was associated with increases (decreases) of 1.2% (95% CI 0.7 to 1.7) and 0.8 (95% CI 0.3 to 1.3), respectively, in alcohol-related deaths per 100,000 person-years. Among higher educated, this association was weak or non-existent. Conclusion. These data suggest that better affordability of alcoholic is associated with higher alcohol-related mortality among low-educated men and women. CORRESPONDING AUTHOR: Dr K. Herttua, University of Helsinki, Finland, [email protected] O472 SMOKING PREFERENCES AMONG YOUNG ADULTS: RESULTS FROM A DISCRETE CHOICE EXPERIMENT P. Brown Univ California Merced, MERCED, United States of America Background: Policy makers in the US currently discount nearly 50% the benefits of smoking cessation efforts due to the lost “consumer surplus”from current smoking. Yet many of the assumptions underlying rational theories of smoking behavior are inconsistent with complex multidimensional decisions associated with smoking, and preference for smoking are unknown. The purpose of this study is to examine the importance of smoking to smokers relative to non-smokers using a discrete choice experiment (DCE). Methods: A DCE was administered to 123 smokers and non-smokers between the ages of 18 and 25, with the participants asked to choose between smoking options (cigarettes, hookah, e-cigs, or cigars) on the basis of cost , place to smoke, level of addictiveness, smell of smoke, and long term health impacts. Results: The results suggest that long term health risks were the most significant influence on smoking decisions (54%), followed by the type of product (11%), the place where they smoked (10%), and the kick or satisfaction from smoking (10%). As expected, there are significant differences between heavy, causal, or light smokers in the importance of these factors (as measured by willingness to pay). However, the differences were not significant enough to explain the high level of discounting of the future benefits associated with stop smoking campaigns. Conclusion: The results have implications for our understanding of the impact of addiction on consumer choice and for policy relating to the benefits associated with stop smoking campaigns. CORRESPONDING AUTHOR: Professor P. Brown, Univ California Merced, United States of America, [email protected]

S142

O473 COOPERATION BETWEEN GENERAL PRACTITIONERS AND PULMONOLOGISTS IN AN ASTHMA/COPD SERVICE: BASELINE DESCRIPTION AND LONGITUDINAL OUTCOMES E.I. Metting, R.A. Riemersma, R. Sanderman, J.W.H. Kocks, T. Van der Molen University of Groningen, University Medical Center Groningen, Groningen, The Netherlands Introduction: Diagnosing asthma and COPD can be difficult which can lead to underdiagnosis and subsequently undertreatment. In 2007, the Asthma/COPD(AC)-service was implemented in the North of the Netherlands to advice General Practitioner(GP)s in diagnosing, treating and managing their asthma and COPD patients by involving local pulmonologists. Methods: GPs may refer respiratory patients for single/yearly follow-up assessments by the AC-service. Spirometry, medical history, health status (Clinical COPD Questionnaire (CCQ)) and asthma control (Asthma Control Questionnaire (ACQ)) are assessed. The pulmonologist inspects the data online, without seeing the patient, and sends the GP the results along with a diagnosis and treatment advice. Digital data from all assessments are available for research. If a change in medication was advised by the pulmonologist, patients were scheduled for an additional 3 months follow-up assessment. Results: Until now, the service has included ~12.000 patients (mean age = 54±19 years, 44% male) from 359 GPs and ~2000 new patients are included yearly. In 78% of the assessments, the PM was able to diagnose patients based on online information (45% asthma, 17% COPD, 7% asthma/COPD overlap). Follow up results after 12 months show that the proportion of patients with ≥1 exacerbations/year decreased from 37% at baseline to 26% after 12 months (n=1062, p<0.00). Conclusion: The AC-service has proven to be a feasible collaboration system between GPs and pulmonologists and the large database provides unique possibilities for further research. The principles of this service might also be applicable in other areas or other (chronic) diseases. CORRESPONDING AUTHOR: E.I. Metting, University of Groningen, University Medical Center Groningen, The Netherlands, [email protected] O474 INSTITUTIONALIZING A COLLABORATIVE MODEL OF CARE FOR DEPRESSION AND DIABETES IN A LARGE HEALTHCARE ORGANIZATION K.J. Coleman, Mark Dr. Dreskin, Gabrielle Dr. Beaubrun, Phillip Dr. Tuso, Jialuo Liu, Magdalena Pomichowski, Alisa Aunskul Kaiser Permanente Southern California, PASADENA, United States of America Introduction: Collaborative care for the treatment of mental and physical health conditions within primary care settings has been shown to be effective in controlled trials. We present preliminary outcomes from efforts to institutionalize this model in a large integrated healthcare system. Methods: Patients were enrolled in the Care for Mental, Physical and Substance Use Syndromes (COMPASS) program as part of a large U.S. federally-funded project to implement a proven chronic care model for depressed patients with cardiovascular disease. COMPASS is six months, involves monthly contact from nurse care managers, and weekly case review by a team of consulting physicians and psychiatrists. Eligible adult patients were those who met the following criteria: Patient Health Questionnaire (PHQ9) scores 10 or higher and either cholesterol (LDL-C) values > 100 mg/dL (heart disease) or hemoglobin A1c (HbA1c) > 8.0% (diabetes). Data were collected from electronic medical records and analyzed using dependent t-tests for those patients having had at least three months of treatment and follow-up measures (n = 91).

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

Results: Patients had significant improvements in PHQ9 scores (-7.09 ± 5.85; p < .001) and HbA1c (-1.01 ± 1.47; p < .001) and trends towards improvement in LDL-C (-7.24 ± 30.79). There were challenges in implementation including changing scope and practice of physicians, psychiatrists, and existing depression care managers and the implementation of intensive care management in the environment of high volume, low cost healthcare. Conclusions: Preliminary results from COMPASS provide support for the institutionalizing proven collaborative care models in high volume primary care settings. CORRESPONDING AUTHOR: Dr. K.J. Coleman, Kaiser Permanente Southern California, United States of America, [email protected] O475 EFFORTS TO SECURE UNIVERSAL ACCESS TO HIV/AIDS TREATMENT: A COMPARISON OF BRICS COUNTRIES J. Sun1, A.D. a Bertoldi2, A. Boing3, M.P.T. Silveira4, S. McGee5, L.E. Ziganshina6, V.N. Khaziakhmetova6, R.M. Khamidulin6, M.R. Chokshi7, F. Suleman8 1 National Institute of Hospital Administration, PR, China 2 Departamento de Medicina, Social da Faculdade de Medicina da Universidade Federa, PELOTAS, Brazil 3 Department of Public Health, Federal University of Santa Catarina, SANTA CATARINA, Brazil 4 Instituto de Biologia, Departamento de Fisiologia e Farmacologia, Universidade F, PELOTAS, Brazil 5 Healthcare Policy Analyst, Market Access, SANOFI, BRICS Medicines Alliance Theme, PRETORIA, South Africa 6 Department of Basic and Clinical Pharmacology, Kazan Federal University, KAZAN, Russian Federation 7 Public Health Foundation, BRICS Medicines Alliance Country Leader, NEW DELHI, India 8 Discipline of Pharmaceutical Sciences, University of KwaZulu-Natal, KWAZULU, South Africa The BRICS countries represent 43% of the global population. Their political and economic leadership in achieving universal access to ARVs is critical for the developing world to achieve the above target. Their positive experiences demonstrate that it is feasible to extend access to ART to the people in need even in a resource-poor setting. Their experiences in securing universal access to ARVs could serve as an engine of innovative health solutions for the other developing countries and contribute in the global effort toward strengthening health systems. This article illustrates how the BRICS countries have been building their focused leadership, making important high level commitments and national policy changes, and improving their health systems, in addressing the HIV/AIDS epidemics in their respective settings. The most important contributors to the success of responses to HIV/AIDS include: creating a legal basis for healthcare as a fundamental human right; political commitment to necessary funding for universal access and concrete actions to secure equal quality care; comprehensive system to secure demands that all people in need are capable of accessing prevention, treatment and care; active community involvement; decentralization of the management system considering the local settings; integration of treatment and prevention; taking a horizontal approach to strengthen health systems; full use of the TRIPS flexibility; and regular monitoring and evaluation to serve evidence based decision making. This study touches cross-cutting research topics of governance, accountability and participation; health equity and rights; and complexity science and people-centered systems. CORRESPONDING AUTHOR: J. Sun, National Institute of Hospital Administration, China, [email protected]

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

O476 PRACTICE-BASED CARE MANAGEMENT OUTPERFORMS HEALTH PLAN DISEASE MANAGEMENT IN TARGETING AND ENGAGEMENT J. Holtrop1, LA Green2, A.A. Emeott3, Z. Luo4, Q. Chen4, M. Tao5, H. Chang5 1 University of Colorado Denver, AURORA, United States of America 2 University of Alberta, EDMONTON, Canada 3 University of Michigan, ANN ARBOR, United States of America 4 Michigan State University, EAST LANSING, United States of America 5 Blue Cross Blue Shield of Michigan, DETROIT, United States of America Introduction: Many US health insurance plans identify patients with chronic disease (e.g., diabetes) via claims data, contact them independently of their physicians, and offer disease management services. An alternative strategy is for primary care practices to identify the care needs of their patients and assume the chronic disease management role. This study compared the effectiveness of practice-delivered (PDCM) versus health plan-delivered care management (HPDCM). Methods: This was a mixed methods prospective cohort comparative effectiveness study. Quantitative data included target and engagement rates drawn from monthly data reports, clinical data from EMRs and registries, and utilization data from claims records. Qualitative data included interviews and observations of practices and staff members. Results: PDCM targeted a higher percentage of patients for care management than HPCDM (36% versus 16%). Of all potential patients with chronic disease, the overlap of the targeting efforts (PDCM actual targeted overlapping with HPDCM hypothetically targeted) was only 7%. HPDCM targeted higher risk patients than PDCM (average 4.3 versus 2.9 mean calculated risk scores). PDCM members were engaged at a higher rate (51%) than HPDCM (13% in similar populations). Clinical and utilization outcomes were not significantly different across the two models. Qualitatively, practices with well-trained care managers and part of the practice team were more satisfied with care management and utilized it routinely, as compared to practices with centralized care managers. Conclusion: Integrated PDCM with well-trained care managers was advantageous in this study. Improving targeting of high risk and high cost patients may maximize clinical and costs benefits. CORRESPONDING AUTHOR: Associate Prof PhD Holtrop, University of Colorado Denver, United States of America, [email protected] Patients' perceptions of illness and treatment: targets for interventions to support medication adherence S477 PATIENTS' PERCEPTIONS OF ILLNESS AND TREATMENT: TARGETS FOR INTERVENTIONS TO SUPPORT MEDICATION ADHERENCE R. Horne1, K.J. Petrie2 1 UCL School of Pharmacy, LONDON, United Kingdom 2 University of Auckland, AUCKLAND, New Zealand This symposium will present novel research illustrating how an understanding of patients’ common-sense model of illness and treatment can be incorporated into the design of interventions to optimise adherence to medication for long term conditions. We present three studies in three long-term conditions where regular use of maintenance medication is an essential component of self-management: osteoporosis, inflammatory bowel disease (IBD) and HIV-infection. We begin with a prospective longitudinal follow-up study of medication beliefs and adherence in osteoporosis. Taking a theory-based approach we show that baseline assessments of specific and general medication beliefs, somatosensory

S143

amplification and mental quality of life, are predictive of adherence 12 months later, identifying potentially modifiable targets for interventions to improve adherence. In the second paper we present an RCT of an intervention which attempts to address these adherence-related medication beliefs (e.g. doubts about necessity and concerns about potential harm) using a tailored intervention delivered through a web-based platform for patients with IBD. We report the feasibility and acceptability of the web-based delivery vehicle and the effects of the intervention of beliefs and reported adherence. The third presentation reports the effectiveness of a smartphone app linking personal adherence to real-time simulated visual displays of viral activity on adherence to antiretroviral therapy, illness perceptions, and medication beliefs. Use of the app led to significantly higher adherence and more positive perceptions of HIV and medication. Our Discussant will outline the implications for theory, research and practice CORRESPONDING AUTHOR: Professor R Horne, UCL School of Pharmacy, United Kingdom, [email protected] S478 POTENTIALLY MODIFIABLE DETERMINANTS OF INTENTIONAL NON-ADHERENCE TO ORAL BISPHOSPHONATES IN WOMEN WITH POSTMENOPAUSAL OSTEOPOROSIS M. Shedden-Mora1, Y. Nestoriuc1, S.R. Heisig1, U.S. Albert2, A. Kurth3, W. Rief4, P. Hadji2 1 University of Hamburg, HAMBURG, Germany 2 Clinic for Obstetrics and Gynaecology, Philipps University Marburg, MARBURG, Germany 3 Clinic for Orthopaedics, Ratingen Hospital, RATINGEN, Germany 4 Department of Clinical Psychology and Psychotherapy, Philipps University Marburg, MARBURG, Germany Introduction. Adherence to oral bisphosphonates in women with postmenopausal osteoporosis (PO) is crucial to prevent fractures and sustain mobility. However, poor adherence to oral compared to intravenous bisphosphonates has been reported in the real-world BonViva Intravenous Versus Alendronate (VIVA) study. The aim of this prospective embedded study was to explore the role of medication beliefs as potentially modifiable determinants of intentional non-adherence to oral bisphosphonates. Method. Of n=1491 women with PO receiving oral Alendronate in the VIVA study, n=439 patients (29.4%; age 71 years; PO duration 3.2 years) were included in a prospective analysis measuring baseline beliefs about medicines (BMQ), somatosensory amplification (SSAS), and quality of life (SF-12). Intentional non-adherence was defined as discontinuation due to lack of compliance or patient’s request, in contrast to unintentional causes such as serious adverse events. Results. At 12 months, 55 patients (12.5%) showed intentional nonadherence, 51 patients (11.6%) showed unintentional non-adherence and 333 patients (75.9%) continued treatment. Using MANOVA, the groups did not differ with respect to age, comorbidities, risk factors or previous treatments. At baseline, intentional non-adherers reported more negative beliefs about harmfulness (p<.001) and overuse (p<.01) of medicines in general, more concerns about their medication (p<.05), higher levels of somatosensory amplification (p<.05) and lower mental quality of life (p<.05), compared to adherers or unintentional nonadherers. Conclusions. Negative medication beliefs are potentially modifiable predictors of intentional non-adherence to oral bisphosphonates. The results serve as a basis to develop early stage interventions targeting medication beliefs in order to improve adherence in PO. CORRESPONDING AUTHOR: Dr. C Shedden-Mora, University of Hamburg, Germany, [email protected]

S144

S479 MODIFYING ADHERENCE-RELATED BELIEFS IN INFLAMMATORY BOWEL DISORDER: RANDOMIZED CONTROLLED TRIAL OF AN ONLINE INTERVENTION S. Chapman1, A. Sibelli2, P. Bondarek1, R. Horne1, A. Forbes3, R. Driscoll3, A. Chater3 1 UCL School of Pharmacy, LONDON, United Kingdom 2 Kings College London, LONDON, United Kingdom, Nederland Introduction: Nonadherence to maintenance medication is common in inflammatory bowel disease (IBD) and increases flare-ups and healthcare costs. Patients who doubt their need for maintenance medication for IBD or are concerned about its potential adverse effects are less likely to adhere. Online intervention provides a potential cost-effective vehicle for the delivery of tailored interactive support to change these adherence-related beliefs. Method: Participants with IBD taking azathioprine or mesalazine were recruited through a hospital clinic and patient organisation website. Patients’ barriers were ‘profiled’ using an automated online questionnaire and those who reported doubts about need or concerns about adverse effects of medication were randomized to control (n=140) and intervention groups (n=129). Intervention participants received components comprising interactive exercises and persuasive messages tailored to their profiled concerns and doubts. The impact of the intervention on beliefs, reported practical barriers and adherence was assessed at 1 and 3 months post-intervention. Results: Intervention participants had lower concerns about medication than controls at 1 month follow-up, Intervention m=2.51, Control m = 2.87, t(134)=2.61, p = 0.01, but this was not maintained at 3 months. There were no significant differences between groups' necessity beliefs or reported adherence at either follow-up point. Conclusions: These results demonstrate that online interventions can be a useful tool for providing tailored support to patients with IBD at risk of nonadherence, and in particular concerns about medication. Future work will include interviews to assess the acceptability of the intervention, to inform development of online support services for IBD patients. CORRESPONDING AUTHOR: Dr CE Chapman UCL School of Pharmacy, United Kingdom, [email protected] S480 A SMARTPHONE APP USING PERSONALIZED HEALTHRELATED VISUAL IMAGERY IMPROVES ADHERENCE TO ANTIRETROVIRAL THERAPY K.J. Petrie1, A. Perera1, Mark Thomas1, J. Moore2 1 University of Auckland, AUCKLAND, New Zealand 2 Media Lab, Massachusetts Institute of Technology, BOSTON, United States of America Background: Mobile technologies have potential to deliver adherence promoting interventions which are cost-effective and scalable. This study aimed to investigate the effectiveness of a smartphone app that links personal adherence to real-time simulated visual displays of viral activity on adherence to antiretroviral therapy (ART), illness perceptions, and medication beliefs. These features were intended to modify participants’ conceptualisations of their HIV infection and ART regimen to improve adherence. Methods: 28 people on ART were randomised to use either a standard or augmented version of the smartphone app. The augmented version contained additional components which illustrated to the participant, in real-time, the estimated concentrations of antiretroviral agents in blood plasma as well as personalized immune activity. Adherence to ART was assessed at baseline and three month follow-up using HIV viral load, pharmacy dispensing and self-report. Results: Participants who received the augmented application displayed a significantly higher level of self-reported adherence to ART at follow-up,

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

compared to individuals who received the standard version (p = .02). Further, there was a significant decrease from baseline to three month follow-up in the proportion of participants classified as non-adherent in the augmented group p = .03). Greater usage of the app was associated with improvements understanding of HIV infection and perceived necessity for ART. Conclusion: This research suggests that a smartphone app, using personalised health-related imagery and real-time feedback, is effective in facilitating adherence to ART. Study findings indicate that such an intervention may have considerable utility in the clinical care of people with HIV infection. CORRESPONDING AUTHOR: Prof KJ Petrie, University of Auckland, New Zealand, [email protected] Lifestyle related chronic diseases in developing countries: addressing this urgent health priority using innovative approaches S481 LIFESTYLE-RELATED CHRONIC DISEASES IN DEVELOPING COUNTRIES: ADDRESSING THIS URGENT HEALTH PRIORITY USING INNOVATIVE APPROACHES B. Oldenburg1, X. Zhong2, S. Van de Vijver3, B. Potemans4 1 University of Melbourne, MELBOURNE, Australia 2 Anhui Centre for Disease Control, HEFEI, China 3 African Population Health Research Centre, NAIROBI, Kenya 4 General Practice Centre, THE HAGUE, The Netherlands Like most developed countries, lifestyle-related chronic conditions (NCDs) are now also the major cause of disease burden in almost all of these countries as well. Yet, most of what we know about how to prevent and control chronic NCDs in developing countries is still derived from research conducted in high income countries. Therefore, it is very important to carefully adapt interventions before they are implemented in developing countries. This process of adapting interventions from high income to developing countries will be demonstrated by describing three chronic disease prevention projects being undertaken in developing countries. In China, Dutch primary care experts in collaboration with expertise from the Anhui Provincial Center for Disease Control have adapted strategies previously evaluated in developed countries for the prevention of cardiovascular disease. Program implementation was then formally evaluated in 7 other communities in Anhui Province. In the Indian project, the Australian-Indian research team undertook a comprehensive needs assessment in Kerala, India to understand more about how local culture, health behaviours and living circumstances influence diabetes prevention. These findings have informed the design and evaluation of a peer support intervention program to prevent diabetes in India. Finally, an African research institute in Kenya is collaborating with a Dutch team to develop an innovative model for prevention and treatment of hypertension in Nairobi slums. The intervention program has been based on a needs assessment incorporating extensive consultations with local stakeholders and the community. The discussant will then consider the learnings from these projects and the implications. CORRESPONDING AUTHOR: Professor F Oldenburg, University of Melbourne, Australia, [email protected] S482 A COMPREHENSIVE APPROACH FOR PREVENTING CARDIOVASCULAR DISEASES AND DIABETES IN THE CHINESE PRIMARY CARE SETTING X. Zhong Anhui Center for Disease Control, HEFEI, China Introduction: Although cardiovascular diseases (CVD) and diabetes are major contributors to China’s NCD crisis, there is little evidence from

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

China how to deal with this situation. We describe the adaptation and implementation evaluation of a Dutch approach to prevent CVD in Anhui Province in China over the past 5 years. Methods: After piloting in one community, the following steps were implemented in a further 7 communities by primary care professionals from local community health centers: 1. A self-test screening questionnaire (SAQ), based on the Dutch concept of Prevention-Consultation, was administered to adults >35 years (DETECTION). 2. Participants with a high risk score were then invited to undergo additional clinical and laboratory tests to confirm their level of risk (CONFIRMATION of RISK). 3. High risk participants were then encouraged and supported to change their lifestyle (LIFESTYLE EDUCATION and COUNSELING). Results: The screening questionnaire was completed by 9067 participants in seven communities with 30% identified as “high-risk”. 34% had further clinical assessment and almost half of these individuals participated in lifestyle education. The SAQ proved to be a reliable and easy tool to use. Conclusion: Although further improvements to the program have been recommended, the evaluation of the implementation of this three-step screening and disease program in urban communities demonstrates its potential feasibility, acceptability and future scalability in the Chinese context. CORRESPONDING AUTHOR: Dr X Zhong, Anhui Center for Disease Control, China, [email protected] S483 LIFESTYLE CHANGE IN INDIA: HOW DO LOCAL CULTURE, HEALTH BEHAVIORS AND LIVING CIRCUMSTANCES INFLUENCE DIABETES? B. Oldenburg University of Melbourne, MELBOURNE, Australia Introduction: Diabetes has become a major public health challenge in India. Factors relevant to implementation of diabetes prevention programmes in resource-constrained countries have been understudied. We describe the findings from research aimed at informing the development, implementation and evaluation of a scalable, community-based diabetes prevention programme in Kerala, India (K-DPP). Methods: Data were collected from three main sources: (1) a systematic review of key research literature; (2) a review of relevant policy documents; and (3) focus groups conducted among individuals with prediabetes in Kerala, India. These findings were then used to identify how local culture, health behaviours and living circumstances influence diabetes prevention. Results: Prevalence of key lifestyle-related risk factors for diabetes particularly related to poor nutrition and sedentariness - are increasing rapidly in Kerala. We have identified key environmental and personal determinants of these rapid lifestyle changes, including: poor accessibility to and availability of health services; traditional cultural values and norms; optimistic bias and other misconceptions related to disease risk; and low expectations regarding one’s ability to influence health and disease outcomes. Conclusions: We have adapted the findings to develop a groupbased peer support programme that will address cultural and family determinants of lifestyle risks, including family decisionmaking regarding adoption of healthy nutrition and increased levels of physical activity. This intervention is being currently evaluated in a cluster randomised community-based trial of diabetes prevention in rural areas of Kerala. CORRESPONDING AUTHOR: Professor F Oldenburg, University of Melbourne, Australia, [email protected]

S145

S484 DEVELOPMENT AND IMPLEMENTATION OF CARDIOVASCULAR PREVENTION PROGRAM IN THE SLUMS OF NAIROBI, KENYA S.J.M. van de Vijver African Population and Health Research Center, NAIROBI, Kenya Introduction: Cardiovascular diseases (CVD) are one of the main causes of morbidity and mortality in sub-Saharan Africa, with hypertension being the main risk factor. National health systems are weak and cannot cope with this epidemic, specifically in the neglected urban areas. The purpose of this study is to develop and implement an innovative and feasible model for prevention and treatment of hypertension in slums. Methods: Through a mixed method of analyzing results from prevalence and intervention studies on CVD and focus group discussion with key informants in the slums of Nairobi an intervention model was developed based on the estimated costs and health benefits. The final model is expected to be highly cost-effective and costs around 1USD/per person in the community. The intervention contains community awareness, home-based screening, patient and provider incentives to seek and deliver treatment specifically for hypertension, and adherence support. Results: The intervention is currently implemented in Korogocho, where more than 5000 inhabitants have been screened and referred to the local clinic for treatment and lifestyle. The screening is highly appreciated by the community with low rates of refusal but it is challenging to convince high risk patients to come to clinic and be compliant with treatment, as it has to compete with other priorities due to insecurity, mobility and comorbidity. Conclusion: This CVD prevention program in the slums has the potential to be highly cost-effective and affordable for governments and NGO’s in the area, although practical challenges during the implementation should be carefully addressed. CORRESPONDING AUTHOR: S.J.M. van de Vijver, African Population and Health Research Center, Kenya, [email protected] Promoting physical activity accross the perinatal period: what strategies are most likely to be effective? S485 PROMOTING PHYSICAL ACTIVITY ACROSS THE PERINATAL PERIOD: WHAT STRATEGIES ARE MOST LIKELY TO BE EFFECTIVE? P.F. Van der Pligt1, M. van Poppel2, E. Olander3, J. Jelsma2 1 Deakin Univeristy, MELBOURNE, Australia 2 VU University Medical Center, AMSTERDAM, The Netherlands 3 City University, LONDON, United Kingdom Chair: Ms Paige van der Pligt, Deakin University, Australia. Objectives:1) Presentation of research investigating patterns of physical activity (PA) and sedentary behaviours across the perinatal period and associations with maternal physical and mental health. 2) A review of behavioural interventions to promote PA during pregnancy and results of recent studies that have assessed strategies to promote PA and reduce sedentary behaviour during pregnancy and following childbirth. 1. Mireille van Poppel will present data from the Vitamin D And Lifestyle Intervention (DALI) study. Physical activity patterns and sedentary behaviours of obese, pregnant women will be reported. The association between mental health status and PA levels in women during pregnancy will be examined. 2. Ellinor Olander will present findings from a review of interventions targeting PA during pregnancy. Results of a meta-analysis and moderatoranalysis assessing the association across studies, between presence or absence of specific behaviour change techniques within interventions and change in PA behaviour will be reported. 3. Paige van der Pligt will present data from the Melbourne Infant Feeding Activity and Nutrition Trial (INFANT). First time mother’s PA

S146

and sedentary behaviours at 3 months postpartum will be reported. Associations of PA levels and sitting time with advice provided by clinicians during the postpartum period will be assessed. 4. Judith Jelsma will present results from the HeLP-GDM study, a postpartum behavioural intervention targeting overweight women with a history of GDM. Effectiveness of the intervention in reducing barriers to PAwill be reported along with psychosocial determinants including social support and self-efficacy. CORRESPONDING AUTHOR: Ms PF Van der Pligt, Deakin Univeristy, Australia, [email protected] S486 PHYSICAL ACTIVITY AND MENTAL HEALTH IN OBESE PREGNANT WOMEN: RESULTS FROM DALI STUDY M.M. van Poppel1, L. De Wit1, J. Jelsma1, A. Bogaerts2, F.J. Snoek1 1 VU University Medical Center, AMSTERDAM, Nederland 2 Limburg Catholic University College, HASSELT, Belgium Purpose. To examine the association between mental health and objectively measured physical activity in obese pregnant women in Europe. Methods. Data gathered before 20 weeks of gestation in the vitamin D and lifestyle intervention for the prevention of gestational diabetes mellitus (DALI) study were used. Time spent in moderate-to-vigorous physical activity (MVPA) and sedentary behaviour were objectively measured with accelerometers. Emotional well-being was measured with the WHO well-being index (WHO-5), and the Cambridge Worry Scale was used to assess pregnancy-related worries. In addition, socio-demographic characteristics, lifestyle factors, and perceptions and attitude regarding weight management and physical activity were measured. Results. A total of 98 obese pregnant women from Austria, Belgium, Ireland, Italy, the Netherlands, Poland and Spain were included. Women had a mean age of 31.6 (SD 5.8) years, a pre-pregnancy BMI of 34.1 kg/ m2 (4.3), and were on average 15.4 (SD 2.8) weeks pregnant. 27.1% of the women had WHO-5 scores indicative of depressed mood (<50) and most frequently endorsed pregnancy-related worries pertained to own and the baby’s health. Women with a depressed mood spent 1.59 times less time in MVPA compared to women reporting good well-being. No differences in MVPA levels were found for women with no, some or a lot of pregnancy worries. Sedentary behaviour was not associated with either depressed mood or worrying. Conclusions. Lower physical activity is associated with depressed mood but not with pregnancy-related worries in obese pregnant women. Improving mood in these women at high risk for gestational diabetes warrants special attention. CORRESPONDING AUTHOR: Dr. MM van Poppel, VU University Medical Center, Nederland, [email protected] S487 HOW DO WE HELP WOMEN KEEP PHYSICALLY ACTIVE IN PREGNANCY? A SYSTEMATIC REVIEW AND META-ANALYSIS E. Olander1, L. Atkinson2, D.P. French3 1 City University London, LONDON, United Kingdom 2 Coventry University, COVENTRY, United Kingdom 3 University of Manchester, MANCHESTER, United Kingdom Introduction: Despite numerous benefits associated with physical activity (PA) during pregnancy, most pregnant women decrease their PA behaviour. Previous reviews have identified what behaviour change techniques (BCTs) are associated with PA, but have excluded pregnant women. The aim of this review was to identify which BCTs are associated with PA behaviour during pregnancy. Methods: A search of six databases identified 24 controlled trials reporting PA behaviour. Included interventions had to measure PA at

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

least twice in pregnancy. Studies without a control group were excluded. All intervention descriptions were coded with the 40 item CALO-RE taxonomy of BCTs. Meta-analysis and moderator analyses were conducted to examine the association across studies between the presence or absence of specific BCTs within interventions and changes in PA behaviour. Results: The overall effect size for intervention effects was d = 0.41. Sixteen BCTs were identified in the included interventions, of which ten were associated with a decreased level of PA. Two BCTs (‘prompt self-monitoring of behavioural outcome’ and ‘provide instruction on how to perform the behaviour’) were associated with a higher level of PA compared to when these BCTs were not included in the intervention. Conclusion: These findings suggest that interventions should include practical elements such as instructing women how to be active and encourage self-monitoring of the outcomes of their PA (for example weight). These results also suggest that caution may be needed in applying BCTs that are associated with PA behaviour in non-pregnant adults as they may be less suitable for pregnant women. CORRESPONDING AUTHOR: Dr K Olander, City University London, United Kingdom, [email protected] S488 HOW DOES CLINICIAN ADVICE INFLUENCE PHYSICAL ACTIVITY PATTERNS AND SEDENTARY BEHAVIOURS DURING THE POSTPARTUM PERIOD? P.F. Van der Pligt, K. Campbell, M. Teychenne, K. Ball, K. Hesketh, D. Crawford Deakin Univeristy, MELBOURNE, Australia Introduction: Physical activity (PA) during the postpartum period is associated with numerous physical and mental health benefits and increases the chance of women returning to pre-pregnancy weight. Yet, little is known about how best to promote PA and reduce sedentary behaviours following childbirth. The aim of this study was to investigate PA and sedentary behaviours and assess associations with provision of clinician advice in first-time mothers. Method: Women (n=448) enrolled in the INFANT Extend study completed a cross-sectional questionnaire at ~3 months postpartum. The validated Active Australia Survey assessed PA time for the previous week. Women self-reported time spent sitting and PA advice received by a healthcare practitioner. Independent t-tests assessed associations between advice and time spent walking, in moderate and vigorous activity and sitting. Logistic regression assessed associations between advice and meeting Australian PA recommendations (≥150 minutes/week). Results: Mean time spent walking (251.97±196.78 minutes/ week) was greater than moderate (36.68±88.58 minutes/week) or vigorous activity time (61.74±109.96 minutes/week). Overall, 76.1% of women met recommendations for PA and 51.9% received PA advice. Mean sitting time was significantly greater on weekdays (242.09±230.40 minutes/week) than on weekend days (175.60±118.70 minutes/week)(p<0.05). There was no significant association between PA advice and time spent sitting, walking in moderate or vigorous activity or with meeting PA recommendations. Conclusions: Walking was the most prevalent PA. Despite many women meeting PA recommendations, advice did not influence PA patterns or sedentary behaviour. Identification of strategies to promote postpartum PA is needed. Further examination regarding content of postpartum clinician advice is warranted. CORRESPONDING AUTHOR: Ms PF Van der Pligt, Deakin Univeristy, Australia, [email protected]

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

S489 HELP-GDM LIFESTYLE INTERVENTION AFFECTS PSYCHOSOCIAL DETERMINANTS IN OVERWEIGHT POSTPARTUM WOMEN WITH RECENT GESTATIONAL DIABETES MELLITUS J.G.M. Jelsma EMGO+ Instute for Health and Care research, AMSTERDAM, Nederland Introduction: Overweight women with a history of gestational diabetes mellitus (hGDM) have an elevated risk to develop type 2 diabetes. This study investigated how a postpartum behavioral lifestyle intervention could affect certain psychosocial determinants and barriers in this population. Methods: 59 women with a body mass index of 25 or higher and a hGDM, participated in a randomised controlled trial on average 26 months postpartum. The intervention (n=29) consisted of two face to face and five telephone lifestyle coaching sessions, in which women received self-help material and a pedometer to help in making lifestyle change. Throughout the intervention the participants were encouraged by healthy messages on postcards and phone. The control group (n=30) received no intervention. At baseline and six months self-efficacy, social support and barriers related to physical activity and nutrition were measured with a questionnaire especially developed for women with a hGDM. Linear regression analysis were applied (p<0.05) to test for intervention effects. Results: The intervention was effective in improving social support (β 3.4, p=0.00; β 2.1, p=0.02), increasing self-efficacy (β -2.1, p=0.03; β 4.3, p=0.00) and reducing barriers (β -3.0, p=0.02; β -3.8, p=0.01) for respectively physical activity and nutrition. Negative beta’s indicate a positive intervention effect for barriers and self-efficacy, for social support positive beta’s indicate a positive intervention effect. Conclusion: These intervention effects on behavioral change determinants are promising, although obviously, actually changing physical activity and nutrition behavior is essential for achieving the associated health benefits. CORRESPONDING AUTHOR: J.G.M. Jelsma, EMGO+ Instute for Health and Care research, Nederland, [email protected] O490 PERCEIVED DISCRIMINATION INDUCES PSYCHOLOGICAL AND PHYSIOLOGICAL STRESS IN IMMIGRANTS S. Fischer, R. Mewes, J. Strahler, U.M. Nater University of Marburg, MARBURG, Germany Introduction. Both everyday experiences of discrimination, as well as mental and physical ill health are frequently reported in Turkish immigrants living in Germany. The question arises whether and how perceived discrimination translates into negative health consequences. One possible mediator is the stress that is exerted by perceived discrimination. It was the main objective of the current study to investigate these mechanisms in Turkish immigrants living in Germany. Methods. Thirty healthy, male Turkish immigrants underwent both a control and a discrimination condition. They repeatedly rated acute perceived discrimination and stress on a visual analogue scale. Heart rate, skin conductance levels, and salivary alpha-amylase as indicators of the autonomic nervous system, and salivary cortisol as an indicator of the hypothalamus-pituitary-adrenal axis were assessed at multiple time points during both conditions. Perceived chronic discrimination and chronic stress were measured via questionnaires. Results. There was a significant time by condition effect in the subjects’ ratings of discrimination (p < .001) and stress (p = .001). Similarly, the subjects’ heart rate (p < .001) and alpha-amylase levels (p = .005) increased over time during the discrimination but not during the control condition. No differences were found for skin conductance levels. Four subjects were cortisol responders (2.5 nmol/l cut-off), and these subjects

S147

were characterized by higher levels of perceived chronic discrimination and chronic stress. Conclusions. Perceived discrimination induces psychological and physiological stress in immigrants. In the long term, repeated experiences of discrimination might render immigrants vulnerable to the development of mental and physical ill health. CORRESPONDING AUTHOR: S. Fischer, University of Marburg, Germany, [email protected] O491 INJUSTICE AT WORK IS ASSOCIATED WITH REDUCED GLUCOCORTICOID SENSITIVITY: FINDINGS FROM A CROSS-SECTIONAL STUDY R.M. Herr1, A. Loerbroks2, J.E. Fischer1, J.A. Bosch3 1 Heidelberg University, MANNHEIM, Germany 2 University of Düsseldorf, DÜSSELDORF, Germany 3 University of Amsterdam, AMSTERDAM, The Netherlands Objective: Organizational justice refers to perceived fairness at the workplace. Low levels of organizational justice have been identified as a major source of distress and predict poor health. Dysregulation of immunological and inflammatory pathways may, in part, underlie these associations. The present study investigated the association of organizational justice with in vivo leukocyte glucocorticoid sensitivity, or glucocorticoid receptor resistance (GCR). GCR reflects impaired feedback control of leukocyte function via hypothalamus-pituitary-adrenal (HPA) activity. Methods: Organizational justice was assessed among 547 male factory workers (mean age = 46 years, SD = 9) by validated questionnaire. The cortisol awakening response (CAR) was measured and fasting blood was collected to assess leukocyte subset distribution (neutrophils, lymphocytes, and the neutrophil/lymphocyte ratio). Associations were assessed by linear regression models, adjusted for a range of demographic and lifestyle variables. Results: In individuals reporting high justice at work (based on median split), cortisol and hematologic parameters showed the expected significant associations (all betas ≥ .23; all p-values ≤ .001). In contrast, among those experiencing low justice at work cortisol was unrelated to hematologic parameters, indicative of impaired glucocorticoid regulation. These regression coefficients were significantly different between groups (all pvalues ≤ .03), and appeared robust against adjustments for a range of potentially confounding variables. Conclusion: Low organisational justice at work was related to an impaired ability of endogenously released cortisol to regulate leukocyte distribution. These findings identify a novel pathway by which organizational justice may affect health. CORRESPONDING AUTHOR: RM Herr, Heidelberg University, Germany, [email protected] O492 JOB STRAIN AND BIOMARKERS OF STRESS IN FEMALE SHIFT WORKERS IN LABORATORY AND FIELD CONTEXT K.K. Karhula1, M. Härmä1, M. Sallinen1, H. Lindholm1, A. Hirvonen1, M. Elovainio2, M. Kivimäki1, J. Vahtera1, S. Puttonen1 1 Finnish Institute of Occupational Health, HELSINKI, Finland 2 National Institute of Health and Welfare, HELSINKI, Finland Introduction. Long-term stress is associated with changes in biomarkers of stress: lowered salivary cortisol peak in the morning and elevated evening level, and with flatter diurnal slopes and decreased daily production of salivary alpha-amylase. Shift work and night work, in particular, may be associated with elevated cortisol secretion. We studied the simultaneous associations of high job strain (combination of low control and high demands) and shift work to biomarkers of stress.

S148

Methods. Female health care professionals (nurses/nursing assistants, n=95) were recruited from wards that belonged to the top (high strain group, HJS, n=42) or bottom quartiles on job strain (low strain group, LJS, n=53) based on a survey in the Finnish Public Sector Study. A laboratory study included Trier Social Stress Test (TSST) and a threeweek field study included sleep diary, actigraphy and three daily saliva samples during pre-selected, circadian rhythm and recovery controlled morning and night shift and day off. Results. The TSST induced a sharp rise in cortisol secretion in the HJS group compared to the LJS group but the difference was not statistically significant (p=0.088). The HJS group had on average higher alphaamylase levels 30 minutes after awakening in the morning shift (p=0.016). Cortisol awakening response was higher in the LJS group in the day-off compared to the HJS group (p=0.048). Conclusion. Within a sample of female shift workers, exposure to stressful work environment was associated with minor changes in stress biomarkers. CORRESPONDING AUTHOR: MSc KK Karhula, Finnish Institute of Occupational Health, Finland, [email protected] O493 PILOT STUDY ON CONDITIONING CORTISOL AND ITS PSYCHOPHYSIOLOGICAL EFFECTS J. Tekampe Leiden University/Radboudumc, LEIDEN, Nederland Conditioned physiological effects of pharmacological agents may offer relevant clinical opportunities to improve treatment for patients with various conditions. While conditioned pharmacological immune effects are shown, less is known about endocrine effects, e.g. on the key stressregulatory parameter cortisol. This study aimed to test a paradigm to condition cortisol levels and assess its psychophysiological effects under non-stress and stress conditions. A randomized placebo-controlled conditioning paradigm was applied in 10 healthy, premenopausal women. In three sessions, an association was established between an unconditioned stimulus (100 mg hydrocortisone or placebo) and a novel-tasting beverage (conditioned stimulus). During three additional sessions, all participants were re-exposed to the conditioned stimulus paired with a placebo pill under non-stress and stress (Trier Social Stress Test) conditions. Feasibility and safety of the design were assessed and effect sizes of cortisol, alpha-amylase, and negative affect were investigated. The design was found to be feasible and safe. Under non-stress conditions, cortisol and alpha-amylase were lower in the experimental than in the control group, whereas negative affect was similar in both groups. During stress exposure, cortisol, alpha-amylase, and negative affect were lower in the experimental than in the control group. These preliminary findings suggest that it might be possible to condition endogenous cortisol which may affect the psychophysiological stress response. The same design is currently used to verify these findings in a larger sample.

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

from diurnal HPA-axis activity has been inconclusive, whereas the predictive value of stress-induced HPA-axis functioning has not been studied before. The aim of this study was to predict mental disorders from morning and stress-induced HPA-axis functioning during a three-year follow-up from awakening and stress-induced HPA-axis functioning. Methods Data were used from 561 TRAILS participants, a prospective cohort study of Dutch adolescents. Saliva samples were collected at age 16 at awakening, 30 min after awakening and before, during and after a social stress test. Mental disorders were assessed three years later with the Composite International Diagnostic Interview (CIDI). Results A flat recovery slope predicted disorders with a first onset after the experimental session (OR = 1.27, p = .04). The cortisol awakening response marginally significantly predicted new disorders (OR = 0.77, p = .06). Recovery revealed smaller, non-significant odds ratios when predicting new onset affective or anxiety disorders, major depressive disorder, or dependence disorders in three separate models, corrected for all other new onsets. Conclusions Our results suggest that delayed recovery of the HPA-axis and maybe reduced CAR are indicators of a more general risk status and may be part of a common pathway to psychopathology. Delayed recovery suggests that individuals at risk for mental disorders perceived the social stress test as less controllable and less predictable. This is a novel finding that has implications for (clinical) practice. CORRESPONDING AUTHOR: dr. E Nederhof, University of Groningen, Nederland, [email protected] O495 THE ROLE OF SAMPLE TIMING IN SALIVARY ALPHAAMYLASE RESPONSES TO ACUTE STRESS T.N. Nagy1, R. Van Lien2, G. Willemsen2, G. Proctor3, M. Efting4, M. Fülöp1, G. Bárdos1, E.C.I. Veerman5, J.A. Bosch4 1 Eotvos Lorand University, BUDAPEST, Hungary 2 VU University, AMSTERDAM, Nederland 3 King’s College London, LONDON, United Kingdom 4 University of Amsterdam, AMSTERDAM, Nederland 5 Academic Centre of Dentistry (ACTA), AMSTERDAM, Nederland

O494 PREDICTING MENTAL DISORDERS FROM HPA-AXIS FUNCTIONING: A THREE-YEAR FOLLOW-UP IN THE TRAILS STUDY E. Nederhof1, F.V.A. Van Oort2, E.M.C. Bouma1, O.M. Laceulle1, A.J. Oldehinkel1, J. Ormel1 1 University of Groningen, GRONINGEN, Nederland 2 Erasmus Medical Center, ROTTERDAM, Nederland

Background: The status of salivary alpha-amylase (sAA) as a marker of sympathetic nervous system (SNS) activity has been challenged, partly because the parasympathetic nervous system (PNS) co-regulates sAA release. Since the PNS and SNS show asynchronous changes over the course of an acute stressor, it is conceivable that the effects of stress on sAA partly depend on sample timing. This idea was tested in the present study, using two laboratory stressors know to elicit distinct patterns of autonomic activation. Method: 34 healthy participants underwent an 8.5-minute cognitive stressor (memory-search task; MT) and an 8.5-minute cold pressor task (CPT). Timed unstimulated saliva samples were collected at six timepoints, while blood pressure and cardiac autonomic activity was assessed. Heart rate variability (RMSSD) was used as a proxy for vagal tone and the pre-ejection period (PEP) as a proxy for sympathetic drive. Results: Both stressors increased anxiety. sAA was unaffected during the MT while sympathetic activity and parasympathetic withdrawal was observed. In contrast, at stressor offset sAA increased rapidly, when RMSSD had returned to baseline values while sympathetic activation was still elevated. The CPT caused the expected increase in blood pressure, but did not significantly affect the salivary parameters at any point. Conclusion: thatsAA changes did not reliably follow changes in psychological stress or other indices of SNS activity and the commonly observed sAA increase at stress-offset may not be representative of sAA levels during stress. Adjustment for salivary flow rate appeared to significantly alter the magnitude and direction of sAA responses.

Background HPA-axis functioning has been presumed to play a key role in the development of psychopathology. Predicting affective disorders

CORRESPONDING AUTHOR: TN Nagy, Eotvos Lorand University, Hungary, [email protected]

CORRESPONDING AUTHOR: Drs. J Tekampe, Leiden University/ Radboudumc, Nederland, [email protected]

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

O496 TREATMENT PATHWAYS, EXPERIENCES AND EXPECTATIONS OF WOMEN WITH GYNAECOLOGICAL CANCER IN ODISHA: A QUALITATIVE INQUIRY S. Pati1, A.S. Chauhan2, S. Nayak1, M.A. Hussain3, S. Nayak4 1 Public Health Foundation of India, BHUBANESWAR, India 2 INCLEN, NEW DELHI, India 3 University of Queensland, QUEENSLAND, Australia 4 All Indian Institute of Medical Sciences, BHUBANESWAR, India Introduction: Cancer is the leading cause of death in developing countries. The burden of cancer is increasing in economically developing countries due to population aging and growth. Burden becomes more severe when associated to the population having low literacy rate and economically weak. It is evident from the earlier studies that treatment seeking pathways and care experience has a significant role in deciding time intervals to diagnosis and treatment. Methods: To understand these practices among the women with gynecological cancer in Odisha, we conducted a hospital based cross-sectional inquiry to explore barriers and enablers in treatment seeking pathways, care giving experiences and patient’s expectations of women aged 18+ in tertiary care hospital. To meet the objectives, In-depth Interviews (IDIs) were conducted using structured interview guides with pre-determined themes. Data was collected from seventeen women who had suffered or were suffering from gynecological cancer and received at least one treatment intervention at tertiary level. Results: After financial constraints, low awareness of signs and symptom, no family support, and stigma is the leading cause of treatment delay among the women. Traditional healing practices were often found both as the silo as well as integrated to allopathic treatment pathways. Some respondents also reported enacted and felt stigma to be one of the reason for delayed consultation. Unavailability of diagnostic facilities is quoted by a majority of respondents as the major cause of delayed treatment. Awareness of cancer among the general population and primary-care practitioners, including in alternative systems of medicine, is important. CORRESPONDING AUTHOR: Dr Pati, Public Health Foundation of India, India, [email protected]

O497 DECISION MAKING PROCESSES AND NEEDS OF BREAST CANCER PATIENTS IN AFTERCARE; DEVELOPING A DECISION AID L. A. Klaassen1, C. Hoving1, C. D. Dirksen2, L. J. Boersma2 1 Maastricht University, MAASTRICHT, The Netherlands 2 Maastricht University Medical Centre, MAASTRICHT, The Netherlands INTRODUCTION To develop an efficient decision aid (DA) it is necessary to understand how patients reach their decisions so that the DA can facilitate these processes. Therefore, the aim of the study was to assess which decision processes are used by breast cancer patients when choosing an optimal aftercare trajectory. Additionally, needs and preferences concerning the design of a DA were investigated. METHOD Patients who have finished their treatment for breast cancer less than 1 year ago were asked to participate in one of 4 focus groups consisting of 4 to 6 participants. A semi-structured questionnaire was used to guide the discussion. All focus group data was coded by 2 independent researchers. RESULTS Needs concerning an aftercare DA were highly personal and varied from leaflets to digital aids. Breast cancer patients reported to use both medical information and intuition to make decisions regarding their aftercare. While rational processes were used to consider whether options were relevant, intuitive processes played a major role in forming a preference for one option over the others. All patients indicated they wanted to make these decisions together with their healthcare provider,

S149

but greatly valued autonomy and preferred to make the final decision themselves. CONCLUSIONS Both rational and intuitive processes seem to be important for patients in making a decision about their aftercare and should receive attention in the DA. Autonomy appeared to be important for patients as well. Therefore, a DA should allow patients to make an informed decision while maintaining their autonomy. CORRESPONDING AUTHOR: L. A. Klaassen, Maastricht University, The Netherland, [email protected] O498 ANXIETY AND DEPRESSIVE SYMPTOMS IN COLORECTAL CANCER SURVIVORS: LOW SEVERITY, EXCEPT IN SPECIFIC SUBGROUPS A.M.J. Braamse1, S.T. Van Turenhout2, J.S. Terhaar sive Droste2, G.H. De Groot3, R.W.M. Van der Hulst4, M. Klemt-Kropp5, S.D. Kuiken6, R.J.L.F. Loffeld7, M.T. Uiterwaal8, J.H. Smit1, B.J.W.H. Penninx1, C.J.J. Mulder2, J. Dekker2 1 GGZ inGeest / VU University Medical Center, AMSTERDAM, Nederland 2 VU University Medical Center, AMSTERDAM, Nederland 3 Rode Kruis Hospital, BEVERWIJK, Nederland 4 Kennemer Gasthuis, HAARLEM, Nederland 5 Medical Center Alkmaar, ALKMAAR, Nederland 6 Sint Lucas Andreas Hospital, AMSTERDAM, Nederland 7 Zaans Medical Center, ZAANDAM, Nederland 8 Spaarne Hospital, HOOFDDORP, Nederland Background. Previous studies indicate that colorectal cancer (CRC) survivors suffer from depressive and anxiety symptoms, but results are conflicting. The present study aimed to (1) assess prevalence of anxiety and depressive symptoms in CRC survivors; (2) compare CRC survivors’ anxiety and depressive symptom severity with reference groups; (3) identify risk factors for heightened anxiety or depression symptom severity. Methods. In this cross-sectional study, 91 persons diagnosed with CRC 3.5 to 6 years ago completed the Inventory of Depressive Symptomatology, Beck Anxiety Inventory, Concerns About Recurrence Scale, and EORTC QLQ-C30. Results were compared with data from the Netherlands Study of Depression and Anxiety. Results. The majority of CRC survivors had no (80.6%) or mild (13.6%) anxiety symptoms; 5.8% reported moderate or severe symptoms. The majority had no (59.3%) or mild (33.0%) depressive symptoms; 7.7% reported moderate symptoms. Controlling for sociodemographic variables, anxiety and depressive symptom severity in CRC survivors did not differ significantly from healthy controls, but was lower compared with persons with current or remitted anxiety and/or depressive disorder. Shorter time since diagnosis and higher number of comorbid diseases were risk factors for higher anxiety symptom severity. Having no partner and higher number of comorbid diseases were risk factors for higher depressive symptom severity. Conclusion. The majority of CRC survivors had no or mild depressive and anxiety symptoms. Anxiety and depressive symptom severity did not differ from healthy controls. However, some subgroups were at risk for heightened anxiety and depressive symptoms. CORRESPONDING AUTHOR: A.M.J. Braamse, GGZ inGeest / VU University Medical Center, Nederland, [email protected] O499 PREDICTORS OF PATIENT-REPORTED PHYSICAL OUTCOMES IN COLORECTAL CANCER PATIENTS: A SYSTEMATIC REVIEW A. Rutte, J. Dekker, A.M.J. Braamse, H. Bomhof-Roordink, C.J.J. Mulder, J.S. Terhaar sive Droste, H.M.W. Verheul, R.W.J.G. Ostelo VU University Medical Center, AMSTERDAM, Nederland

S150

Background: Colorectal cancer (CRC) patients may report impaired physical outcomes, including poor physical functioning, poor sexual functioning, and symptoms. We aimed to systematically review predictors for patient-reported physical outcomes in CRC patients. Methods: A systematic, computerised literature search in Medline, EMBASE, PsycINFO, and the Cochrane Library was carried out in December 2012. Studies with ≥150 adult CRC patients were included. We included studies that assessed socio-demographic predictors, diseasespecific predictors and stoma, baseline function and symptoms, psychosocial predictors, and lifestyle predictors. We analysed short-term (0-2 years post-treatment) and long-term (>2 years post-treatment) outcomes separately. Results: Forty-six studies were included, of which 25 were of high methodological quality. For the majority (87%) of the studied associations, weak or inconclusive evidence was provided. Strong to moderate evidence was found for female gender predicting worse short-term physical functioning, worse long-term sexual functioning, and worse longterm urinary incontinence. Being younger at diagnosis predicted worse short-term physical well-being (strong evidence). Being older at surgery and pre-operative urinary incontinence predicted long-term urinary incontinence (moderate evidence). Pre-operative difficulty in bladder emptying predicted long-term difficulty in bladder emptying (moderate evidence). In addition, strong to moderate evidence suggested that stoma did not affect various patient-reported physical outcomes. Conclusion: Gender, age, and pre-operative urinary functioning predict physical functioning, sexual functioning, and symptoms. The majority of the associations, including all psychosocial and lifestyle predictors were only studied once or twice. Further research on these factors is strongly recommended. CORRESPONDING AUTHOR: Prof. Dr. J. Dekker, VU University Medical Center, The Netherlands, [email protected] O500 FATIGUE SCREENING: IDENTIFYING CASES OF CANCERRELATED FATIGUE USING THE SF-36, POMS, AND FATIGUE SYMPTOM INVENTORY M.A Andrykowski1, Paul B. Jacobsen2 1 University of Kentucky, LEXINGTON, United States of America 2 Moffitt Cancer Research Institute, TAMPA, United States of America Purpose: A case definition approach has been developed to identify clinically significant cases of cancer-related fatigue (CRF). This approach employs a clinical interview and specific diagnostic criteria. Identification of cut-off scores on questionnaire measures of fatigue to identify cases of CRF would be useful. Method: Women (n=389) undergoing adjuvant therapy for stage 0-IIIa breast cancer participated in a Fatigue Diagnostic Interview following completion of an initial regimen of chemotherapy (N=204) or radiotherapy (N=185). Participants completed three questionnaire measures commonly used to assess fatigue including the POMS-Vigor subscale (POMS-VIGOR) , SF-36 Vitality subscale (SF-VITALITY), and Fatigue Symptom Inventory Average Fatigue rating (FSI-AVG). Results: 106 women (27%) met clinical criteria for CRF (CRF “cases”). ROC analyses using CRF caseness as the criterion yielded areas-underthe-curve (AUC) as follows: SF-VITALITY (AUC=.810), POMSVIGOR (AUC=.803), and FSI-AVG (AUC=.794). (AUC’S > .75 are generally considered as representing good overall accuracy). Cut-off scores yielding optimal discrimination (i.e., good sensitivity and specificity) for identifying CRF cases were: SF-VITALITY (scores <35), POMS-VIGOR (scores >8), and FSI-AVG (rating > 4). Using these cut-off scores, sensitivity (SE), specificity (SP), positive predictive value (PPV), and negative predictive value (NPV) were: SF-VITALITY (SE=.74, SP=.71, PPV=.49, NPV=.88), POMS-VIGOR (SE=.81, SP=.63, PPV=.45, NPV=.90), and FSI-AVG (SE=.84, SP=.60, PPV=.44, NPV=.91).

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

Conclusions: Each of the three questionnaire measures demonstrated more reasonable accuracy in identifying “cases”of CRF. Use of the cutoff scores identified for each may be an efficient means of screening and identifying individuals for further clinical evaluation to determine clinically significant CRF cases. CORRESPONDING AUTHOR: Professor A Andrykowski, University of Kentucky, United States of America, [email protected] O501 BLOOD AND MARROW STEM CELL TRANSPLANTATION TYPE PREDICTS EMPLOYMENT AT 1 YEAR POST-TRANSPLANT E.M. Morrison, S. Ehlers, C. Bronars, C. Patten, W. Hogan, S. Hashmi, D. Gastineau Mayo Clinic, ROCHESTER, United States of America Introduction: Recipients of blood and marrow stem cell transplantation (SCT) contend with functional and emotional threats to well-being well beyond the peri-transplant period. Autologous transplant is associated with fewer medical complications and better overall recovery than allogeneic transplant (Wong et al., 2010), differentially impacting adjustment. Employment status is an important marker for post-transplant adjustment (Kirchhoff et al., 2010). The present study examined whether transplant type predicted employment status 1 year after SCT. Methods: Participants (N=268) from a larger prospective cohort of 565 patients completed a lifestyle survey 1-year post-SCT. Participants were selected for either fulltime employment or underemployment attributable to health. They were predominately married/partnered (77.6%), Non-Hispanic Caucasian (81.3%), males (63.1%) between the ages of 23-73 (mean: 54 years), who received autologous transplant (78.5%). Employment status was dichotomized to reflect whether it was impacted by health: Full-time (47%) or Underemployed (53%). Results: Transplant type was correlated with age and employment; allogeneic patients were younger (mean: 52 years vs. 56 years autologous) and more likely to be underemployed (73.7% vs. 46.6% autologous). Employment was correlated with gender; females reported more underemployment (63.5% vs. 46.2% of males). Multiple linear regression demonstrated that transplant type significantly predicted employment status, controlling for age and gender (B=.307, p< .01). Autologous patients were more likely to report full-time employment (53.4%) than allogeneic patients (26.3%). Conclusion: These findings have implications for health care economics, personal finances, and long-term adjustment. Patient care that includes managing expectations related to functional ability and returning to work may facilitate posttransplant adjustment. CORRESPONDING AUTHOR: EM Morrison, Mayo Clinic, United States of America, [email protected] O502 IMPROVING THE AVAILABILITY, PROMOTION AND PURCH ASE OF HEALTHY FOODS FROM COMMUNITY SPORTING CLUB CANTEENS: RCT L. Wolfenden1, M. Kingsland1, B. Rowland2, P. Dodds1, K Gillhan3, S.L. Yoong1, M. Sidey4, J. Wiggers1 1 The University of Newcastle, NEWCASTLE, Australia 2 Deakin University, GEELONG, Australia 3 Hunter New England Population Health, NEWCASTLE, Australia 4 Australian Drug Foundation, MELBOURNE, Australia Introduction: Non-elite community sporting clubs represent an attractive setting for health promotion. The aim of this study was to assess the effect of a multi-component intervention on the availability, promotion and purchase of fruit and vegetable and non-sweetened drink products from non-elite community sporting club canteens. Method: The study employed a repeat cross-sectional, parallel group, cluster randomized controlled trial design. Non-elite community football clubs were

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

randomly assigned to an intervention that sought to improve the availability and promotion of fruit and vegetables and non-sweetened drinks in sporting club canteens. The interventions was delivered over 2.5 consecutive winter sporting seasons and provided clubs with resources, training, and support from a dedicated project officer. Telephone surveys of club members were conducted at baseline and 30 months later. Results: Eighty eight sporting clubs and 1143 club members participated in the study. Relative to the control group, at follow-up, clubs allocated to the intervention were significantly more likely to have fruit and vegetable products available at the club canteen (OR=5.13; 95% CI 1.70-15.38), were more likely to promote fruit and vegetable selection using reduced pricing and meal deals (OR=34.48; 95% CI 4.18-250.00) and members of intervention clubs were more likely to report purchase of fruit and vegetable (OR=2.58 95%CI; 1.08-6.18) and non-sweetened drink (OR=1.56; 95%CI 1.09-2.25) products. Conclusion: The findings demonstrate that an implementation intervention within this setting can improve the nutrition environment of sporting clubs and the purchasing of fruit and vegetables and non-sweetened drinks by sporting club members. CORRESPONDING AUTHOR: Dr L Wolfenden, The University of Newcastle, Australia, [email protected] O503 RESULTS OF A RCT: THE EFFECTS OF A VIDEO-DRIVEN AND T E X T- D R I V E N WE B - B A S E D O B E S I T Y P R E V E N T IO N INTERVENTION M.J.L. Walthouwer1, A. Oenema1, L. Lechner2, H. De Vries1 1 Maastricht University, MAASTRICHT, Nederland 2 Open University of the Netherlands, HEERLEN, Nederland Introduction: The purpose of this study was to test the effects of a text and video version of a web-based computer-tailored obesity prevention intervention on BMI, physical activity, and energy intake among Dutch adults. Method: A three group RCT was conducted to compare the effects of the two versions of the intervention with a no intervention control condition. Participants were recruited through companies, occupational health centers, and mass media advertisements. Measurements took place at baseline (N = 2,426) and six months after baseline (N = 1,663). Primary outcome measures were self-reported BMI, physical activity, and energy intake as assessed through online questionnaires. The data were analysed by means of linear regression analyses. Results: Participants’ mean age was 48.3 years and 58.2% was female. Most participants were highly educated (44.9%), while 34.8% had a medium and 20.3% a low level of education. Preliminary analyses showed that participants in the text condition had a lower energy intake (β=-.182, p=0.000) and lower BMI (β=-.045, p=0.009) as compared to the control condition, while participants in the video condition only had a lower energy intake (β=-.190, p=0.000) than those in the control condition. Moderation analyses will be conducted to identify potential differential intervention effects according to level of education. Conclusions: Preliminary results of this study indicate that a textdriven computer-tailored obesity prevention intervention may be more effective than a video-driven computer-tailored intervention, since the text-based version had an effect on both BMI and dietary intake. CORRESPONDING AUTHOR: JL Walthouwer, Maastricht University, Nederland, [email protected] O504 SELF-REPORTED AND PEDOMETER-RECORDED WALKING IN RELATION TO FITNESS AND ADIPOSITY IN A SEDENTARY POPULATION A. Moncrieft, M.M. Llabre, R. Goldberg, M.D. Gellman, J. Pico, M. Gutt, B.E. Hurwitz, N. Schneiderman University of Miami, MIAMI, United States of America

S151

Introduction: Understanding relationships between walking and cardiometabolic risk is particularly relevant in primarily sedentary populations. The purpose of this study was to determine if walking is associated with greater fitness and/or less adiposity in type 2 diabetes (T2DM) and metabolic syndrome (MetS) using both self-reported and pedometer recorded indices of walking. Methods: Participants included 274 individuals (mean age = 53 ± 8.25 years; 61% women) with T2DM (n = 134) or MetS (n = 140). Participants reported time walking in bouts ≥ 10 minutes, and wore pedometers to record daily steps for one week. Multiple group structural equation modeling was used to evaluate relationships between walking and body mass index, waist circumference, and peak VO2. Results: Self-reported walking was related to increased peak VO2, B = 0.188, SE = 0.09, p < .05, and decreased waist circumference, B = -0.590, SE = 0.25, p < .05. Pedometer recorded steps were also related to decreased waist circumference, B = -0.041, SE = 0.02, p = .05, and increased peak VO2, B = 0.017, SE = 0.01, p < .05. Self-report and pedometer measures were not associated, r = .103, SE = .10, p > .05. Conclusions: Increased walking is related to reduced adiposity and improved fitness in T2DM and MetS. Use of both self-report and pedometer recorded methods of walking assessment is warranted as they may assess different constructs and differentially relate to cardiometabolic risk factors. CORRESPONDING AUTHOR: E. Moncrieft, University of Miami, United States of America, [email protected] O505 WHOSE RESPONSIBILITY IS IT? PUBLIC PERSPECTIVES ABOUT OBESITY MANAGEMENT DISCUSSIONS IN CLINICAL CONSULTATIONS J. Hart1, S. Yelland2, A. Mallinson2, Z. Hussain2, S. Peters2 1 Manchester Medical School, MANCHESTER, United Kingdom 2 Manchester Centre for Health Psychology, School of Psychological Sciences, MANCHESTER, United Kingdom Introduction: Tackling obesity is a global priority. Effective approaches include individual and public health interventions, including input from health professionals. It is important, therefore, to understand public views and expectations on the role of health professionals in the management of weight and implementation of behaviour change. Methods: Qualitative, semi-structured interviews were conducted with 31 members of the general public; recruited via weight loss groups and public advertisements. An inductive grounded theory approach was used to analyse the data. Within this, recurring concepts/themes were explored through questioning in future interviews, until thematic saturation was achieved. Results: Three themes summarised views on the role of health professionals in weight management. They were (1) Patient’s Expectations, (2) Control and (3) Responsibility to Initiate Talk. A theoretical model was formulated to demonstrate the association between the themes. Conclusion: Individuals perceive doctors as having a role in their weight management but not always providing the most effective approach. Individuals who perceived their weight management as within their control believed that health professionals held less responsibility to initiate weight management talk. In contrast, those who perceived their weight management as out of their control believed that health professionals held all the responsibility to initiate this talk. Formal training for doctors to enable them to take into account not only their patients’ expectations but also the relationship between control and responsibility to initiate talk may increase the efficacy of their communication, and lead to more frequent/acceptable initiation about weight management and behaviour change. CORRESPONDING AUTHOR: Dr J. Hart, Manchester Medical School, United Kingdom, [email protected]

S152

O506 R I S K C O M M U N I C AT I O N C O N C E R N I N G D I E TA RY SUPPLEMENTS WITH TWO-SIDED HEALTH EFFECTS K.C.J. Curfs1, E. M. Pajor1, A. Oenema1, A. Opperhuizen2, H. de Vries1 1 Maastricht University, MAASTRICHT, Nederland 2 The Netherlands Food and Consumer Product Safety Authority, UTRECHT, Nederland Introduction: Twenty percent of the Dutch population uses multivitamin and mineral supplements. Although dietary supplements can improve health, they might have negative side-effects as well. It is important that these two-sided health effects are communicated to consumers in such a way that they will use this information when making a decision about the use of the product in question. Since research on communication concerning supplements with two-sided effects is scarce, the goal of this study was to investigate how these effects can be communicated effectively in order to improve informed decision making. Method: Experts from a variety of fields were invited to participate in a systematic three-round Delphi study. Their opinion and agreement were assessed on the importance of factors related to communication concerning dietary supplements with two-sided effects. Results: Data collection was completed at the end of June 2014. It was expected that agreement among experts would be reached on the importance of factors such as the lay-out and content of the message, the recipient’s ability and motivation to process the message, risk perception and the degree to which the message was tailored to individual needs. Conclusion: A list of important factors related to communication concerning dietary supplements with two-sided effects was identified. These factors can be used in experimental studies, in which their effects on informed decision making can be tested. CORRESPONDING AUTHOR: K.C.J. Curfs, Maastricht University, Nederland, [email protected] O507 DEVELOPING ATAILORED NUTRITION AND PHYSICAL ACTIVITY INTERVENTION FOR LOW-EDUCATED WOMEN FROM VARIOUS ETHNIC BACKGROUNDS K. Romeike1, A Oenema1, H. De Vries1, L. Lechner2 1 Maastricht University, MAASTRICHT, Nederland 2 Open University of the Netherlands, HEERLEN, Nederland Introduction: Unhealthy nutrition and physical activity (PA) behaviors are prevalent among Dutch, Turkish and Moroccan women in the Netherlands, having a low socioeconomic status (LSES). This project aims to develop two versions of a computer-tailored intervention to improve dietary and PA behavior among Dutch, Turkish and Moroccan LSES women. One version is tailored to cognitive variables, while the other is additionally tailored to ethnic identity (EI), i.e. to which degree someone identifies with an ethnic group. It is hypothesized that additional EItailoring will result in a superior effect and higher acceptance among the target group. Methods: Using the Intervention Mapping protocol, two evidence- and theory-based interventions were developed, based on selfregulation models. Results: Both interventions consisted of a sequence of printed brochures. In intervention 1, messages were tailored on participants’ health behavior, attitude and self-efficacy (SE). Methods used to address these factors were: behavioral feedback to increase awareness of risk behavior, arguments to improve attitude and goal setting to improve SE. In intervention 2, messages and lay-out were additionally tailored on EI. Several EI-categories were distinguished: Turkish/Moroccan traditional, Turkish/Moroccan adjusted to Dutch culture, Turkish/Moroccan in-between (partly traditional, partly adjusted to Dutch culture), Dutch traditional and Dutch modern. Brochures per category differ in lay-out, colors, use of symbols and patterns, role models and advice messages (e.g. referring to food or cultural customs important in the particular EI group). Conclusions: The systematic process of Intervention Mapping

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

resulted in two interventions for the target group. The interventions will be evaluated in an RCT. CORRESPONDING AUTHOR: K. Romeike, Maastricht University, Nederland, [email protected] O508 USING PREVENTION SUBTYPES TO PREDICT FUTURE SMOKING AND ALCOHOL STATUS W. Velicer, S.F. Babbin, L.A. Brick, C.A. Redding, A.L. Paiva University of Rhode Island, KINGSTON, United States of America Introduction. Substance abuse interventions tailored to the individual level have produced effective outcomes for a wide variety of behaviors. One approach employs cluster analysis to identify prevention subtypes. In this study, prevention subgroups were tested for their ability to predict future substance use. Method. Analyses were performed on a baseline sample of 6th graders from 20 Rhode Island middle schools involved in a 36-month intervention. Most adolescents (>95%) reported being in the Acquisition Precontemplation (aPC) stage: not smoking or not drinking and not planning to start in the next six months. For both smoking and alcohol, the baseline aPC sample (N=4158) was randomly split into five subsamples. Cluster analysis was performed within each subsample based on three substance acquisition variables: Pros, Cons, and Situational Temptations. Results. Across all subsamples for both smoking and alcohol, four clusters were identified: (1) Most Protected; (2) Ambivalent; (3) Risk Denial; and (4) High Risk. Baseline cluster membership was used to predict smoking and alcohol status at 12, 24, and 36 months. At all time points, for both smoking and alcohol, the proportion remaining in aPC was consistently the highest for Most Protected, and lowest for High Risk. Conclusions. The ability of the prevention subgroups to predict future smoking and alcohol use provides evidence of the utility of these subtypes for tailoring interventions. CORRESPONDING AUTHOR: Professor F Velicer, University of Rhode Island, United States of America, [email protected] O509 I LIKE IT BUT I DON'T: ATTITUDINAL AMBIVALENCE AND ADDICTIVE BEHAVIOURS D. Lindsay, A. L. Swinbourne James Cook University, TOWNSVILLE, Australia Introduction: Attitudinal ambivalence occurs when individuals endorse both positive and negative attitudes toward the same target object. Ambivalent attitudes are particularly relevant for addictive behaviours, as these behaviours may have both positive and negative evaluations associated with them. For example, drinking alcohol may make someone relaxed but can also produce feelings of nausea. Despite this, the majority of research focused on attitudes toward addictive behaviours assume that these attitudes are either positive or negative, not positive and negative. By assessing ambivalence toward addictive behaviours, a greater understanding of the nature of attitudes underlying these behaviours can be realised. Method: A total of 247 participants (M= 28.76 years) took part in this study. A measure of potential ambivalence, which asks participants to indicate their positive and negative evaluations on split semantic differential scales, was completed for five different health behaviours: drinking on a weekday and a weekend, exercising, increasing fruit and vegetable intake and smoking. Results: Attitudes toward alcohol consumption were found to be the most ambivalent. When examining drinking behaviour, a pattern emerged suggesting that the greater the quantity of alcohol consumed, the more ambivalent participants were toward drinking. Similar patterns were found for smoking. Conclusions: The results suggest that attitudinal ambivalence is an important aspect of addictive behaviours, as participants engaging in greater levels of these behaviours also reported higher levels of ambivalence. This indicates that

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

individuals are aware of the positives and negatives of addictive behaviours but continue to perform them anyway. CORRESPONDING AUTHOR: B Lindsay James Cook University Australia [email protected] O510 EVALUATING SMOKING TREATMENT SERVICES IN THE UK: WHAT FACTORS INFLUENCE QUIT ATTEMPTS? F Dobbie1, R Hiscock2, S Murray1, L Bauld1 1 University of Stirling, STIRLING, United Kingdom 2 University of Bath, BATH, United Kingdom Introduction: A number of countries now offer smoking treatment services. One example is the UK. Here, these services have existed as part of the National Health Service since 1999 and involve a wide range of providers, different types of behavioural support and pharmacotherapy. Services are required to assess the smoking status of their clients at four weeks post quit date and validate this using exhaled CO. Methods: Prospective, longitudinal, cohort study of 3,000 service users in eight parts of England who had made a quit attempt, followed-up at 4 weeks and 1 year. This paper presents short term (4 week) outcomes. Results: The services, at 4 weeks, achieved an overall CO validated quit rate of 44%, rising to 57% when all self-report quitters were included. Interim findings suggest that the most effective support was open groups (which were delivered by specialist stop smoking service advisors) and the least effective support was provided by practice staff in primary care. Age and affluence predicted quitting. Clients who were highly nicotine dependent were less likely to quit but low dependence did not predict quitting. Conclusion: UK treatment services were last evaluated in a prospective study a decade ago (Judge et al, 2005). Our results for short term outcomes are strikingly similar to this earlier study, suggesting that at the national level, these services are still delivering effective treatment for UK smokers. However, there is variation between individual providers and behavioural support types that requires further research. CORRESPONDING AUTHOR: M Dobbie University of Stirling United Kingdom [email protected] O511 EFFECTS OF A REDUCTION-TO-QUIT PROGRAM FOR SMOKING COPD OUTPATIENTS: THE REDUQ STUDY M.E. Pieterse1, P. Hagens1, J. van der Palen2 1 University of Twente, ENSCHEDE, Nederland 2 Medisch Spectrum Twente, ENSCHEDE, Nederland Introduction. Smoking cessation is the most effective means of favourably modifying the progression of COPD. However, due to several failed quit attempts, smoking COPD patients often give up trying. A novel strategy to re-motivate these smokers may be a reduced smoking intervention, enabling them to rebuild self-control and improve their cognitions towards quitting. The REDUQ study evaluates such a reduction-to-quit intervention. Methods In a multi-center RCT 130 COPD outpatients (smoking ≥10 cigarettes per day; with at least two recent failed quit attempts; unwilling to quit but motivated to reduce their smoking) were enrolled. Patients in the REDUQ intervention group received small-group behavioural counselling (8 sessions in 18 months + 4 telephone contacts) delivered by trained pulmonary nurses and free nicotine replacement therapy (NRT). Patients in the self-help control group attended a single information session and received a self-help manual with reduction strategies. As soon as patients expressed readiness to quit, they were referred to the smoking cessation program offered by the outpatient clinic.

S153

Results. No difference in self-reported 7-days point-prevalence abstinence were observed at both 6 months (REDUQ 9.3%; control 7.1%) and 12 months (REDUQ 14.8%; 12.5% Control). An intervention effect was found on the average reduction in cigarettes per day from baseline to 6 months (48.5% vs. 21.6%; F=8.4; N=77; P<.01), but not at 12 months (53.2% vs. 41.3%). Discussion. Although the reduction-to-quit intervention was effective in considerably reducing tobacco exposure at 6 months, this effect was not sustained and did not result in additional abstinence. CORRESPONDING AUTHOR: Dr. M.E. Pieterse University of Twente Nederland [email protected] O512 ADOLESCENT TOBACCO, ALCOHOL AND ILLICIT DRUG USE: DOES RESILIENCE MATTER? RK Hodder1, M Freund2, J Bowman2, L Wolfenden2, K Gillham3, J Dray2, J Wiggers1 1 Hunter New England Population Health/The University of Newcastle, WALLSEND, Australia 2 The University of Newcastle, CALLAGHAN, Australia 3 Hunter New England Population Health, WALLSEND, Australia Introduction: Substance use contributes significantly to the global burden of disease, with initiation primarily occurring during adolescence. Research suggests multiple individual and environmental factors may be protective of adolescent substance use. Such factors have been summarised under the concept of resilience, where individual factors constitute internal resilience (eg. self-esteem) and environmental factors constitute external resilience (eg. family connection). Given the lack of comprehensive peer reviewed evidence regarding the association between resilience and substance use a study was conducted to determine the association between adolescent resilience and substance use. Method: A cross sectional study in 32 Australian secondary schools was conducted. Students in Grade 7 (12-13 years) to Grade 10 (15-16 years) completed an anonymous web-based survey regarding their resilience (overall, internal, external, 14 resilience subscales) and substance use (tobacco, alcohol, marijuana, other illicit substances). Logistic regression analyses were conducted to examine associations between all measures of resilience and substance use. Results: The survey was completed by 9,987 students (participation rate 89.8%). Excluding recent tobacco use, significant associations were found between all but one resilience and substance use relationship tested (101/102). Adolescents with low resilience had significantly greater odds of reporting tobacco [OR range: 1.6 (95%CI:1.3-2.0)-4.6 (95%CI:3.65.8)], alcohol [OR range:1.3 (95%CI:1.1-1.6)-4.4 (95%CI:3.3-6.1)], marijuana (OR range:1.9 (95%CI:1.5-2.5)-6.8 (95%CI:4.9-9.3)] and other illicit drugs use (OR range:2.4 (95%CI:1.8-3.2)-12.3 (95%CI:8.4-18.0)] than those with moderate/high resilience. Conclusions: The pervasiveness of the association found between resilience and substance use suggests enhancing the resilience of adolescents has the potential to reduce adolescent substance use. CORRESPONDING AUTHOR: R.K. Hodder, Hunter New England Population Health/The University of Newcastle, Australia, [email protected] O513 META-ANALYSIS OF THE IMPACT OF DEPRESSION ON SUBSEQUENT SMOKING CESSATION IN CARDIAC PATIENTS F. Doyle1, D. Rohde1, A. Rutkowska1, K. Morgan2, G. Cousins1, H.M. McGee1 1 Royal College of Surgeons in Ireland, DUBLIN, Ireland 2 Perdana University-RCSI, PERDANA, Malaysia Introduction. Smoking cessation is crucial for patients with coronary heart disease (CHD), yet depression may impede cessation success. We

S154

systematically reviewed the prospective association between depression and subsequent smoking cessation in those with CHD in order to quantify this effect. Methods. Electronic databases (PsychInfo, PubMed, CINAHL) were searched (1st January 1990 to 22nd May 2013) for prospective studies of CHD patients which measured depression at baseline (scales, diagnostic interview or antidepressant prescription) and reported smoking continuation/cessation at follow-up. Standardized mean differences (SMD) and associated 95% confidence intervals (CI) were estimated using random effects meta-analysis. Sensitivity analysis explored the impact of limiting meta-analysis to studies using different depression measures, different durations of follow-up, higher quality studies, study design or statistical adjustment of results. Results. From 1451 citations retrieved, 28 relevant articles were identified. Meta-analysis of all available data from 20 unique datasets found that depressed CHD patients were significantly less likely to quit smoking at follow-up (SMD=-.39, 95% CI -.50 to -.29; I2=51.2%, p=.005). Estimates remained largely unchanged for each sensitivity analysis, except for two studies that used antidepressants, which showed a much larger effect (SMD=-.94, -1.38 to -.51; I2=57.7%, p=.124). Conclusions. CHD patients with depressive symptoms are significantly less likely to quit smoking than their non-depressed counterparts. CHD smokers may require aggressive depression treatment to enhance their chances of quitting. CORRESPONDING AUTHOR: Dr F Doyle, Royal College of Surgeons in Ireland, Ireland, [email protected] O514 SCHOOL DROPOUT AND POOR HEALTH: CAUSATION, DIRECT SELECTION, OR INDIRECT SELECTION? H. Bosma, M.J. Theunissen, P. Verdonk, F. Feron Maastricht University, MAASTRICHT, Nederland

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

O515 LOW FAMILY INCOME AND BEHAVIOR PROBLEMS: ARE CHILDREN DIFFERENTIALLY SUSCEPTIBLE? T. Bøe1, Mari Hysing1, Henrik Da Zachrisson2 1 Uni Research, BERGEN, Norway 2 The Norwegian Center for Child Behavioral Development, OSLO, Norway Background. Children who grow up in poor families have been shown to have a higher rate of externalizing problems. Some children may be more sensitive to this influence based on temperamental characteristics. Methods. Linear and semi-log associations between income-to-needs across children’s first three years and internalizing and externalizing problems when they were five were investigated using data from the Norwegian Mother and Child study. Moderation by emotionality when children were one-and-a-half and three years old was assessed. Results. Significant main effects of income-to-needs and emotionality, and a significant interaction effect between income-to-needs and emotionality was found for externalizing problems. No strong associations of income-to-needs with internalizing problems appeared Conclusion. Children in poor families with an emotionally reactive temperament had higher scores on externalizing problems when they were five. CORRESPONDING AUTHOR: T Bøe, Uni Research, Norway, [email protected] O516 ARE SOCIOECONOMIC POSITION, WORK STRESS, AND WORK C O M P L E X I T Y A S S O C I AT E D T O M O B I L I T Y A F T E R RETIREMENT? Pingo Kåreholt, Alexander Darin Mattsson, Charlotta Nilsen Aging Research Center, STOCKHOLM, Sweden

Introduction - School dropout, being related to poor health, is an important public health problem. Finding the early life pathways towards these health inequalities is important for intervention purposes. We set out to examine how processes of causation (prior socioeconomic conditions), direct selection (prior health status) and indirect selection (personality) relate to the link between school dropout and health. Methods - A case-control design, consisting of 330 dropout cases and 330 controls still attending school was used. The respondents, aged between 18 and 23, came from Eindhoven and its environs in The Netherlands. Using logistic regression analyses, the association between dropout and current health was separately controlled for the measures of socioeconomic background, prior health status, and personality. Except for dropout status, all variables were questionnaire-based. Results - Poor current health was linked to school dropout (OR: 4.06; 95% CI 2.07-7.98). Controlling for the socioeconomic background and prior illnesses reduced this odds ratio with only 7 to 8 percent. In addition to the control for socioeconomic background, controlling for the personality measures, low control beliefs in particular, reduced the odds ratio with about 25 percent. Conclusion -As the influence of low control beliefs on the dropout current health association is partly independent of the socioeconomic circumstances in which the respondents grew up (supporting “indirect selection”), personality might be equally fundamental as socioeconomic conditions in generating life-course socioeconomic and health-related pathways for people. Public health professionals might thus consider early interventions in youth with all too fatalistic and powerless mind-sets.

INTRODUCTION: Most people spend a large proportion of their adult life working; hence work conditions may be important for mobility at older ages. Socioeconomic position (SEP) might associate to mobility both from associations to work conditions and from other causes. AIM: To study how work stress, work complexity, and SEP 20+ years earlier associates to mobility after retirement. DATA: Swedish nationally representative samples, from 1968, 1981, and 1991 were re-interviewed 1992, 2002, and 2004, and 2011 (69+) with 2024 years follow-up time (women, n~900; men, n~800). Only persons <65 years and gainfully employed at baseline where included. METHOD: Ordered logistic regressions. VARIABLES: Mobility was measured as number of limitations with walking 100 meters, walking up stairs, rising from a chair, and standing without support. Work conditions was measured as 1) Work stress: Job control, demands, and high strain (low control, high demands). 2) Work complexity with data, people and substantive (general) complexity. Socioeconomic position (SEP) is summarized of social class, education, income, and cash margin. RESULTS: Work stress (high job strain and low control) were significantly associated to worse mobility even after controlling for SEP among women, but not men. Higher substantive complexity and complexity with data had significant associations to better mobility among both sexes, also after controlling for SEP. Work complexity with people was significant among women, but not when controlling for SEP. Higher SEP had significant associations to better mobility after controlling for work complexity and high job strain, but not after controlling for job control.

CORRESPONDING AUTHOR: Dr. H. Bosma, Maastricht University, Nederland, [email protected]

CORRESPONDING AUTHOR: Pingo Kåreholt, Aging Research Center, Sweden, [email protected]

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

O517 THE BURDEN OF BEING POOR AND FEELING STIGMATISED: SOCIOECONOMIC HEALTH INEQUALITIES IN A DUTCH INTERNET-PANEL A.M.W. Simons, D.A.I. Groffen, A. Koster, H. Bosma Maastricht University, MAASTRICHT, The Netherlands Introduction. Because of their meritocratic ideologies and optimistic beliefs in upward social mobility, Western countries create high expectations of being successful. Consequently, individuals are held responsible for their socioeconomic status and might become stigmatised when living at the bottom of the socioeconomic hierarchy. Therefore, we examined whether perceived class-related stigmatisation mediates the association between low socioeconomic status and poor health. Methods. Cross-sectional data on perceived class-related stigmatisation, household income and subjective health were obtained from 1,735 participants (age 16 - 90) of the Dutch Longitudinal Internet Studies for the Social sciences (LISS)-panel. Perceived class-related stigmatisation was measured by six statements regarding perceived stigmatisation because of income, educational level or occupation. Results. Of the participants, 19.3 percent perceived some kind of classrelated stigmatisation. The low income group felt more often stigmatised than the high income group (OR 1.89, CI 95% 1.40-2.57). The association between perceived stigmatisation and health was significant (OR 2.43, CI 95% 1.81-3.76). Low income also was linked to poor health (OR 1.75, CI 95% 1.29-2.38). This association decreased with 17 percent after controlling for perceived stigmatisation (OR 1.62, CI 95% 1.19-2.21). Conclusion. Living at the bottom of the socioeconomic hierarchy was associated with poor health. Part of the association was due to people with low income feeling more often stigmatised. Although this has to be confirmed with longitudinal analyses, the current study suggests that, through class-related stigmatization, societal beliefs in success and own responsibility might harm people and contribute to creating socioeconomic health inequalities. CORRESPONDING AUTHOR: Drs. A.M.W. Simons, Maastricht University, The Netherlands, [email protected] O518 INCOME OR INCOME RANK: UNRAVELLING THE MECHANISM THROUGH WHICH INCOME RELATES TO DEPRESSION H. Osafo-Hounkpatin1, A.M. Wood2, G.D.A. Brown3, G. Dunn1 1 University of Manchester, MANCHESTER, United Kingdom 2 University of Stirling, STIRLING, SCOTLAND, United Kingdom 3 University of Warwick, COVENTRY, United Kingdom Introduction: Social psychology research as well as primate studies indicate that it is an individual’s rank in a social hierarchy, rather than their absolute income that is relevant for depression. Studies that have tested this rank hypothesis (RH) have yielded inconclusive results, partly due to the use of different study populations, statistical methods and choice of reference group. We provide an advanced test of the RH in two separate datasets, allowing direct comparison of results. Methods: Using ordinary least squares regression, we compare the predictions of the RH and the absolute income hypotheses (AIH). We use data from a cohort of 10,317 individuals living in Wisconsin (USA) who completed surveys in 1992 and 2003 and a cohort of 32,900 British individuals from 2002 to 2008. We use various specifications of utility as a function of income so as to account for the non-linear relationship between income and depression (assessed by the Centre of Epidemiologic Studies Depression (CES-D) scale), giving the AIH its best chance. Results: The best fitting model for both current and future risk of depression was one that included, in addition to other demographic variables, income rank within education and gender groups for both populations. Sensitivity analysis using different reference groups produced similar

S155

results. Bootstrapping analysis confirmed that the rank model was a better fit than the alternative model. Conclusion: This study illuminates the role of psychosocial processes linking social rank to depression. The findings may inform strategies to reduce income-related health disparities. CORRESPONDING AUTHOR: H. Osafo-Hounkpatin, University of M a n c h e s t e r, U n i t e d K i n g d o m , h i l d a . o s a f o h o u n k p a t i n @ postgrad.manchester.ac.uk O519 IMPLEMENTING EVIDENCE-BASED CARE IN LOW INCOME COUNTRIES: ACCEPTABILITY AND IMPACT OF A UGANDAN TRAINING COURSE J. Hart1, L. Byrne-Davis1, C. Armitage2, M. Johnston3, H. Slattery4, M. Jackson4, R. McCarthy5, G. Yuill6, G. Byrne7 1 Manchester Medical School, MANCHESTER, United Kingdom 2 Manchester Centre for Health Psychology, School of Psychological Sciences, MANCHESTER, United Kingdom 3 Aberdeen Health Psychology Group, Institute of Applied Health Sciences, ABERDEEN, United Kingdom 4 University Hospital of South Manchester, MANCHESTER, United Kingdom 5 University of Salford, SALFORD, United Kingdom 6 Stepping Hill Hospital, STOCKPORT, United Kingdom 7 Health Education North West, MANCHESTER, United Kingdom Introduction. Understanding the efficacy and efficiency of education programmes is crucial in low and middle income countries where workforce development is key to improving patient health. However, education programmes are typically evaluated in terms of increases in knowledge, skills, satisfaction or patient outcomes meaning there is a lack of evidence about whether, how and in what circumstances, such programmes lead to changes in professional practice. The aim of the present study was to assess behaviour change in a multi-professional group taking part in a oneday course on acute illness management (AIM) in Uganda. Methods. Using questionnaires, we measured knowledge pre and post course; key drivers of behaviour (capability, opportunity and motivation) post-course; and behavior 1 month post-course. Results. 51 health professionals took part in AIM and all took part in the research. There was a significant increase in knowledge pre to post course.There were high levels of self-reported capability and motivation and lower levels of opportunity. Behavioural intentions were significantly higher than self-reported behaviours at 1 month. Knowledge was not predictive of self-reported behaviour 1 month later. Conclusion. It is acceptable to measure behavioural predictors alongside education. We need further work to understand the link between knowledge, predictors and behaviour change; knowledge is necessary but not sufficient to change professional practice. Evaluation of the impact of education by measuring predictors of behaviour allows tailoring of education to increase its efficacy/efficiency. CORRESPONDING AUTHOR: Dr J. Hart, Manchester Medical School, United Kingdom, [email protected] P520 COPING AND DYSFUNCTIONAL BELIEFS ABOUT CAREGIVER BURDEN IN CAREGIVERS OF PATIENTS WITH TERMINAL CANCER I.N.P. Perez Sachez1, M.L.R.G. Maria Luisa Rascon-Gasca2 1 National Autonomous University of Mexico, MEXICO CITY, Mexico 2 National Institute of Psychiatry Ramon de la Fuente Muñiz, MEXICO CITY, Mexico Introduction: Caregiver burden has significant effects on primary caregiver mood and health. The emotional distress of the caregiver increases

S156

in the last stages of illness but there are few studies in caregivers of patients with terminal cancer (CPTC). And also not much is known about cognitive variables in caregiver burden. Aims: (1) Examine the influence of the positive aspects of care, coping and dysfunctional beliefs about caregiver burden, depression and anxiety, (2) Determine if difference exists in the levels of these variables between caregivers of CPTC and caregivers of patients in oncological treatment (CPOT). Methods: The study was cross-sectional and non-experimental, was conducted in the clinic of pain and palliative care of the National Medical Center XXI century in Mexico City, 133 CPTC (n=73) and CPOT (n=60) were interviewed. The average age was 49.05 years (SD=14.01), 42 were male and 91 were women. Results: The CPTC has more caregiver burden (t=3.30, p=.001), anxiety (t=2.13, p=0.35), and depression symptoms (t=2.17, p=0.31) than CPOT; coping, dysfunctional beliefs care and positive aspects of care did not differ (in their score); solely coping subscale showed differences, alternative perception of events (t”“=3.46, p=.001), which was higher among CPOT than CPTC. Regarding the influence of these variables over burden, a Structural Equation Model was performed (x2=569, p=113, df=3), the model showed that dysfunctional beliefs predict coping styles and positive aspects of care and these 3 variables affect the caregiver burden, which in turn predicts anxiety in caregivers. CORRESPONDING AUTHOR: Ms I.N.P.S. Perez Sachez, National Autonomous University of Mexico, Mexico, [email protected] P521 PREDICTING PROSTATE CANCER SCREENING ATTENDANCE AMONG UNDERSERVED AA, WHITE MEN USING THE SOCIALINFLUENCE- EFFICACY MODEL O.G. Dominic1, M.L. Liddell, Morgan2, B.A. Betsy Aumiller3, B.C.K. Brenda C. Kluhsman4, A.S. Angela Spleen3, E.J.L. Eugene J. Lengerich3 1 Highmark, Inc.; Penn State, CAMP HILL, United States of America 2 Psychiatry, Loma Linda University, LOMA LINDA, United States of America 3 The Pennsylvania State University, College of Medicine, Department of PHS, HERSHEY, United States of America 4 Education, Lane Community College at Florence, FLORENCE, United States of America Purpose: This study examined motivating factors (social influences) for prostate cancer(CA) screening with education and with the cognitive element of informed-decision making; the role of social support in prostate cancer screening; and the intention to be screened among medically underserved, low-income African American (AA) men of the recommended screening age of ≥ 50 years and older residing in Pennsylvania. Methods: 151 eligible AA engaged in one of five communitybased prostate screening offered. Pre- and post-test assessing soicodemographic variables (age, sex, race/ethnicity, marital status, education and income levels, poverty level, insurance and employment status) and screening health behaviors (prostate screening status, intention to screen, attitude, belief and knowledge about prostate CA and screening, attendance of a prostate CA education session, role of social influence on their decision-making to get screened, and cues to action to get screened for prostate cancer) were administered prior to screening. Results: We found significant differences (p=0.04) between agreement on the necessity of screening by race/ethnicity. Social construct indicated that on average screening decisions for men in all groups were affected by their social structure. Differences in individual social measures for AA were observed. AA were in agreement with self-efficacy for screening than non-AA men (p=0.03). 42% of the individuals who attended the prostate cancer screening reported being influenced by another individual in their decision to be screened for prostate cancer. Conclusions: A community-based targeted intervention with a church and social support

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

component appears to have potential for improving prostate screening uptake among AA men in Pennsylvania. CORRESPONDING AUTHOR: Conslt;Adj Prof OG Dominic, H ig h m ar k , I nc . ; P en n S t at e , U n it e d S t at e s o f Am er ic a , [email protected] P522 PROSTATE CANCER PATIENTS' EXPERIENCES AT DIAGNOSIS AND PSYCHOLOGICAL WELL-BEING FIVE YEARS LATER: A NATIONWIDE SURVEY U.S. Lehto National Institute for Health and Welfare THL, HELSINKI, Finland Today most prostate cancers are diagnosed in the early phase and men live for many years after the diagnosis. Thus, their well-being and quality of life (QOL) is becoming more important. We investigated patient experiences and psychological well-being of a population-based sample of prostate cancer patients of all ages who received any of the current treatments. In a random national sample (50%) of prostate cancer patients diagnosed in 2004, we collected information on the patients’ experiences at diagnosis and choice of treatment (e.g., how the treatment was selected, patient satisfaction with the care and information they received, psychological reactions). In 2009, we inquiredabout the experiences and measured the patients’ psychological well-being (psychological symptoms, satisfaction with life). 1239 completed questionnaires (73 %) were accepted for the study. Differences between treatments and predictors of psychological wellbeing were investigated using descriptive statistics and regression analysis. Half of the respondents were satisfied with the care and information they received about the cancer and side effects of treatment. Experiences and well-being were most positive after brachytherapy. Patients undergoing prostatectomy or brachytherapy more often participated in selecting their treatment. Negative experiences, such as learning of the diagnosis in an impersonal way and dissatisfaction with the information and care received, predicted poorer well-being. Negative experiences of the care and information received were highly prevalent. Experiences and well-being varied between treatments, usually in favor of brachytherapy. Negative experiences, which improvements in care and communication would likely reduce, seem to have a long-lasting impact on patients’ psychological well-being. CORRESPONDING AUTHOR: Dr. U.S. Lehto, National Institute for Health and Welfare THL, Finland, [email protected] P523 RELATIONSHIPS BETWEEN TAIWAN NURSES' ATTITUDES TOWARD CARING FOR DYING PATIENTS AND END-OF-LIFE CARE BEHAVIORS Y.L. Tsai1, L.Y. Yang2 1 Tri-Service General Hospital, TAIPEI, Taiwan 2 Department of Social Welfare and Social Work, Hsuan Chuang University, TAIPEI, Taiwan Purpose: Nurses have many chances to take care of dying patients and families. The purpose of the study was to explore the relationships between Taiwan Nurses’ attitudes toward caring for dying patients. Methods: The study adopts crossed sectional exploratory design and surveys are used to collect data through using of Chinese-version Frommelt attitudes toward care of the dying (FATCOD) scale, Death Attitude Profile-Revised Scale (DAP-R), and Nursing Care Behavior Scale (NCBS) for End-of-Life (EOL) patients and families. Convenience sampling was used to recruit nurses. There are 318 nurses from four hospitals in Taiwan. Descriptive statistics and linear regression were used to analyze the data. Cronbach alpha was 0.91, 0.83, and 0.95 for DAP-R, FATCOD, and NCBS respectively.

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

Results: The more years nurses work in the hospital, the more positive nursing care behaviors are for EOL patients and families. The NCBS score was positively associated with the DAP-R score. The nurses’ FATCOD score was positively associated with the DAP-R score. Conclusions: Regardless of how the nurses felt about death, the provision of professional and quality care to EOL patients and their family is necessary to promote healthy dying. Developing continuing education programs that teach nurses how to take care of the EOL patients and their families may be required to improve the care quality at the end of life and have healthy death. CORRESPONDING AUTHOR: Y.L. Tsai, Tri-Service General Hospital, Taiwan, [email protected] P524 AN ONLINE PSYCHOLOGICAL INTERVENTION FOR PARTNERS OF CANCER PATIENTS: INTEREST, INFLUENCING FACTORS AND PREFERENCES N. Köhle1, C.H.C. Drossaert1, C.F. Van Uden-Kraan2, I.M. Verdonck- de Leeuw3, E.T. Bohlmeijer1 1 University of Twente, ENSCHEDE, Nederland 2 VU University, AMSTERDAM, Nederland 3 VU University and VU University Medical Center, AMSTERDAM, Nederland Background: Partners of cancer patients may suffer from psychological distress. However, the availability of evidence-based, easily accessible supportive interventions for this group is limited. Aim of this study was (1) to examine partners’ intention to use an online intervention and their wishes regarding such an intervention and (2) to identify variables that are related to this intention. Methods: A questionnaire was spread via various channels, and was filled in by 168 partners of cancer patients. Questions included: sociodemographics, disease-related characteristics (type, stage and duration of cancer), partners’ health (physical and mental health) and the psychological impact of their role as caregiver (caregiver burden and posttraumatic growth). Additionally, we asked partners about their intention to use an online intervention, and their wishes regarding such an intervention. Findings: 48.6% of the partners were not, 37.7% were maybe and 13.7% were definitively interested in an online intervention. Partners who intended to participate were significantly younger, more often employed, perceived more posttraumatic growth and more caregiver burden. Remarkably, the intention to use the intervention was neither associated with disease-related characteristics, nor with partners’ psychological distress. Most respondents felt the intervention should take less than 1 hour a week, should contain information and peer support. Respondents differed in their preferences about the maximum duration and the need for having contact with a psychologist. Discussion: Based on these results we are developing an online intervention for partners of cancer patients. Partners’ wishes will be addressed and they are actively involved in the development of the intervention. CORRESPONDING AUTHOR: MSc N. Köhle, University of Twente, Nederland, [email protected] P525 MANAGING FEAR OF CANCER RECURRENCE: A CASE STUDY M.A. van de Wal, P. Servaes, A.S. Speckens, H. Knoop, M.F.M. Gielissen, J.B. Prins Radboudumc, NIJMEGEN, Nederland Introduction: While high levels of fear of cancer recurrence (FCR) are present in almost half of all cancer survivors few FCR-specific interventions have been developed. Our aim is to describe the course and content of a psychotherapeutic intervention specifically targeted at high levels of

S157

FCR. The case of a 63-year old breast cancer survivor with high levels of FCR is presented. Methods: The intervention comprised 6 individual sessions of cognitive behavior therapy (CBT) and two telephone consults. Assessments took place before, during, and after the 12-week intervention to provide us with a better understanding of the process of therapeutic change during CBT. Both pre- and post-therapy measures assessed the severity of FCR, triggers and consequences of fear as well as psychosocial outcomes (e.g. quality of life, distress). Perceived control over FCR was assessed at 6 times during the course of therapy to monitor individual treatment progress. After each session a report form addressing the content of that session was completed by the patient. Overall therapy satisfaction was evaluated after the last session. Findings/Discussion: This intervention tackles FCR with one of the most frequently used psychotherapeutic interventions nowadays (CBT). At this moment, the effectiveness of the intervention is evaluated in a randomized controlled trial carried out in the Netherlands. CORRESPONDING AUTHOR: M.A. van de Wal, Radboudumc, Nederland, [email protected] P526 THE ROLE OF THE MEANING SYSTEM IN THE ADJUSMENT TO CANCER A. Visser1, N.F. Uwland1, A. Garssen1, G. Westerhof2 1 Helen Dowling Institute, BILTHOVEN, Nederland 2 University of Twente, ENSCHEDE, Nederland Introduction: This qualitative study is part of a mixed methods study into the role of spirituality in the adjustment to cancer. Three questions were examined: 1) Which elements of the meaning system are involved in the experience of cancer? 2) What is the role of those elements in the ways people deal with their illness? 3) Do people with a spiritual meaning system adjust to their illness differently than people with a non-spiritual meaning system? Method: We conducted in-depth interviews with 20 persons one year after they had been diagnosed with cancer. Participants were selected through purposive sampling with the intent to maximize the variance in spirituality and well-being among the interviewees. The Grounded Theory Method was applied as developed in Dutch by Wester (1995). Results: It was found that the elements ‘beliefs about control’, ‘ (transcendent) meaning in life’, and ‘self-image’ were mainly involved in the experience of cancer. The beliefs within these elements were sometimes discrepant with the experience of cancer. The participants described that they tried to resolve the negative emotions resulting from this discrepancy by finding meaning in the experience of cancer (legitimation) or trying to continue or regain their way of life (continuation). Persons with a very spiritual meaning system seemed to adjust to cancer more often by legitimation, while those with less spiritual meaning systems were more often focused on continuation. Conclusion: These results can help health care professionals to identify (spiritual) needs of their patients during the process of adjustment to cancer. CORRESPONDING AUTHOR: A. Visser, Helen Dowling Institute, Nederland, [email protected] P526 FACTORS ASSOCIATED WITH MISSED OPPORTUNITIES FOR THE DIAGNOSIS OF COLORECTAL CANCER H. Rogers1, S. Harris-Haywood2, L.A. Siminoff3 1 University of Deusto, BILBAO, Spain 2 Case Western Reserve University, CLEVELAND, OH, United States of America 3 Virginia Commonwealth University, RICHMOND, VA, United States of America

S158

Introduction: Early diagnosis of colorectal cancer (CRC) not detected during screening colonoscopies requires prompt recognition and provision of medical attention for those who are experiencing symptoms associated with the disease. The aim of this study was to examine patient socio-demographic characteristics, type of symptoms experienced, and doctor-patient communication associated with missed diagnostic opportunities (MDOs). Methods: A mixed methods study was conducted in Virginia and Ohio in which 242 patients diagnosed with Stage 1-4 CRC who experienced symptoms prior to diagnosis and who were not diagnosed as a result of routine screening were recruited. They participated in a semi-structured cognitive interview focused on recognition and appraisal of all symptoms and communication with health care providers about them. Patient selfreport data about physician-patient encounters was verified by systematic review of medical records prior to diagnosis. Validated scales were administered. Results: Independent risk factors for MDOs included: being female [OR=2.35(1.10-5.00), p=0.03], less than 50 years old [OR=2.60(1.096.20), p=0.03], having no insurance or being underinsured [OR=3.05(1.02-9.12), p=0.04], the number of physicians reported symptoms to [OR=2.06(1.48-2.89), p<0.001], and having a physician who does not fully explore presenting symptoms [OR=5.22(1.91-14.31), p=0.001]. Stomach pain, anemia, and constipation were the most common symptoms of the MDO group. More than 75% of misdiagnoses were alternate GI-GU diseases, including hemorrhoids and diverticulitis. Conclusions: Continued increased training of physicians to dispel stereotypes about who is vulnerable to CRC is needed. Communication, including staying on task to explore presenting symptoms, also seems critical to the diagnostic process. CORRESPONDING AUTHOR: L. Rogers, University of Deusto, Spain, [email protected] P527 EXPERIENCE OF ILLNESS AMONG GYNAECOLOGICAL CANCER SURVIVORS: THREE CASE STORIES C. Laranjeira Piaget Institute, VISEU, Portugal Introduction: The aim of this qualitative case study is to obtain a deeper and more profound understanding of the reality of women living with gynaecological cancer focusing particularly on changes in life perspective. Method: The study is based on a series of interviews carried out within the space of one year and involving three women with gynaecological cancer; each woman was interviewed two times. The participants were between 33 and 56 years of age; they were diagnosed with various gynaecological cancers in different stages. The participants were Portuguese, came from the same community, and had a homogeneous cultural background. Results: There was an increased consciousness of the relationship between life and death, which constituted a disclosure rather than an actual change in life perspective. The three women were ‘opening up’ to the beauty and the essentials in life and experienced an increased desire to live their life in accordance with their own values. Their revitalised view of life increased their desire for authenticity. When it proved impossible to live in accordance with new insights the women were particularly frustrated. Conclusion: From a compassionate perspective our findings suggest that an awareness of patients’ increased openness to their own needs and desires is an important resource in the healing and rehabilitative process of cancer patients. The paradoxes and the struggles involved disguise a hidden potential for health. CORRESPONDING AUTHOR: Prof. C. Laranjeira, Piaget Institute, Portugal, [email protected]

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

P528 A QUALITATIVE STUDY OF WOMENS' VISUALISATION OR CONCEPTUALISATION OF THEIR FETUS DURING PREGNANCY M. Steele, B. Williams, H. Cheyne NMAHP Research Unit, University of Stirling, STIRLING, United Kingdom Introduction: A range of health behaviours during pregnancy can have substantive impact on the health of the fetus, lasting into later life. Despite this, little is known about the way in which pregnant women visualise or conceptualise their fetus and how this may or may not influence behaviour. We sought to explore women’s views in this area. Methods: Semi-structured qualitative interviews were carried out with a purposive sample of 9 pregnant women. Framework analysis was used to analyse the data. Three researchers were involved in the analysis. Results: The women were found to employ a ‘protective visualisation’ in early pregnancy where they avoided imagining the fetus or the future baby. However, women who were later in pregnancy appeared to have constructed a visualisation of the future baby and reported feeling more connected to, and protective of, the fetus. They reported the visualisation as neither abstract nor concrete, but somewhere on a continuum of the two. Language used imposed ‘babyhood’ on the fetus to give it more human qualities. The ultrasound scan was found to have a profound impact on the emotions and attitude of the pregnant woman. Descriptions typically involved language of movement and childhood. Physical reactions of pregnancy, such as negative reactions to foods, were described in terms of the baby’s likes or dislikes rather than the woman’s own reaction. Conclusion: We predict that encouraging women to develop a concrete visualisation of the future baby in early pregnancy would make them more likely to engage in protective health behaviours. CORRESPONDING AUTHOR: M. Steele, NMAHP Research Unit, University of Stirling, United Kingdom, [email protected] P529 UNDERSTANDING AND REDUCING BINGE DRINKING: THE EFFECTIVENESS AND CONSEQUENCES OF A MINIMUM PRICE POLICY D. Keatley, M.S. Hagger Curtin University, PERTH, Australia Introduction: Excessive alcohol consumption, and binge drinking in particular, has direct negative consequences in terms of economic, social, and especially health outcomes. Health-related consequences include increased risk of chronic problems such as liver cirrhosis and heart disease. Price policies, such as setting a minimum price per unit or standard drink, have been proposed to reduce binge drinking. The current study investigated the introduction of a price policy in Western Australia (WA). Method: The current study used 14 focus groups (N = 106) to gain indepth data on attitudes, knowledge, and beliefs regarding the introduction of a minimum price policy. Data were transcribed and analysed for emergent themes using inductive thematic analysis. Results: Analyses indicated that participants largely objected to the introduction of a minimum price policy. Major themes were (1) dislike for the policy and (2) doubt over its effectiveness. There were, however, several suggestions regarding how the policy could be improved, such as (1) transparency over generated revenue and (2) educational policies. Conclusion: There was a negative reaction regarding the introduction of a minimum price policy. Most participants suggested it would be ineffective, mostly due to people finding other ways to get alcohol, regardless of cost. This has direct consequences for policy-makers. The current results support previous research (e.g., introduction of price policies in UK); however, there were several notable differences, such as the outcome in WA. Drug taking as a replacement for binge drinking was suggested by every group as a most likely consequence.

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

CORRESPONDING AUTHOR: Dr D. Keatley, Curtin University, Australia, [email protected] P530 HOW WELL PARENTAL BEHAVIORAL INTENTION PREDICTS FUTURE INFLUENZA VACCINATION FOR YOUNG CHILDREN A COHORT STUDY J.T.F. Lau The Chinese University of Hong Kong, HONG KONG, Hongkong Purpose: We investigated if behavioral intention predicts future behavior in the case of parents letting 2-5 years old child to take up seasonal/H1N1 vaccination, and whether it mediates constructs of the Theory of Planned Behavior (TPB). Sample: Hong Kong Chinese parents of 2-5 year old children. Methods: In the 12-month cohort study, population-based random telephone surveys were conducted at baseline (n=540) and 12-month afterwards (follow-up rate=81.5%). Results: At baseline, 65.5% of the parents intended to (moderate to very likely) let their child get vaccinated in the next 12 months while at 12month, 24.6% of all those being followed up had their child actually vaccinated (42.2% and 7.5% for those with higher or lower baseline intention). Using a cut-off of “very likely”in baseline intention, predictive positive and negative values were 63.3% and 73.6%. Variables assessing positive attitudes (OR=4.45), negative attitudes (OR=0.510) and subjective norm (OR=6.06) related to influenza vaccination were associated with baseline behavioral intention and significantly predicted the child’s actual influenza vaccination prospectively. The predictions became nonsignificant in the presence of baseline behavioral intention (OR=23.77) to take up influenza vaccination, which hence mediated the relationships between the three constructs of the TPB and future behavior, supporting the theory. In contrast, baseline perceived behavioral control did not predict future influenza vaccination. Conclusion: In this case, behavioral intention is a good predictor of future vaccination behavior. It mediated the predictions made by attitudes and subjective norm fully. Perceived behavioral control was non-significant; TPB was hence partially supported. CORRESPONDING AUTHOR: Prof. J.T.F. Lau, The Chinese University of Hong Kong, Hongkong, [email protected] P531 DEVELOPMENT OF EXERCISE BEHAVIOR MODEL BASED ON HAPA FOR JAPANESE HIGH SCHOOL STUDENTS Y.S. SHIMIZU International Christian University, TOKYO, Japan Introduction: Many of the Japanese high school students’ levels of physical activities are drastically lower than those in previous decades. In order to enhance the physical activity level and prevent chronic diseases, effective intervention methods with a behavior change model need to be developed. The health action process approach (HAPA) is a psychological theory of health behavior change (Schwarzer, 1992). Methods: The purpose of this study was to develop an exercise behavior model based on HAPA that can be used to predict exercise behavior in Japanese high school students. 733 Japanese high school students(Male: 291, Female: 442, mean age: 16.32, SD: 0.52) were asked to answer a questionnaire, which was composed of socio-demographic backgrounds, health behavior (the Japanese translated version of HAPA model items) and some additional question items developed for Japanese high school students in particular. Exploratory factor analyses, reliability analyses, confirmatory factor analyses and covariance structure analyses were conducted. Results: Showed that each of the five scales (Self-efficacy, Outcome Expectancy, Risk Perception, Intention, Planning) had one factor structure and acceptable structural validities and reliabilities. Covariance

S159

structure analyses found that the exercise behavior model was an acceptable model. Conclusion: The results have indicated that the hypothetical HAPA model can be used to predict Japanese high school students’ exercise behavior from the observable variable. It is also implied that exercise intervention programs with a focus on enhancement of self-efficacy for exercise and outcome expectancy can be highly effective in promoting Japanese high school students’ exercise behavior. CORRESPONDING AUTHOR: Dr. Y.S. SHIMIZU, International Christian University, Japan, [email protected] P532 MUSICAL DISTRACTION AND ITS EFFECT ON DENTAL ANXIETY AND ADHERENCE TO TREATMENT IN SCHOOL CHILDREN M. Rios Erazo1, V. Torres1, A. Herrera1, P. Barahona1, Y. Molina1, P. Cadenasso2, V. Zambrano3, G. Rojas1 1 Faculty of Dentistry, University of Chile, SANTIAGO, Chile 2 Alfonso Leng Dental Care Center, Municipality of Providencia, SANTIAGO, Chile 3 Recreo Care Center, Municipality of San Miguel, SANTIAGO, Chile Introduction: Dental anxiety is a condition that affects a large percentage of the world population and has a negative impact on oral health and quality of life. There are different forms of management of dental anxiety, and musical distraction is one of the techniques used during dental care, which could have a positive effect by decreasing the levels of dental anxiety and increasing adherence to treatment in 6 year-old children. Methods: Randomized clinical trial study. Sample of 92 children from primary care centers. The experimental group received standard dental care with musical distraction and the control group received only the standard dental care. FIS scale was used for the measurement of selfreported dental anxiety of the child. Clinical and microbiological indicators were used to assess the adherence to treatment, together with the index of Oral Hygiene Behavior. Measurements were performed at baseline and at the end of treatment. Results: 6,6% of the sample showed high levels of dental anxiety at the beginning of the treatment, and then declined to 2,2% at the end of the treatment. The clinical and microbiological indicators improved, but there are no differences in the rate of oral hygiene behavior. There are no differences in levels of dental anxiety and adherence to dental treatment between control and experimental group. Conclusions: The levels of dental anxiety in the sample are low. At the end of the treatment the dental anxiety decreases and the adherence to treatment increases, but this is not attributable to the musical distraction. CORRESPONDING AUTHOR: Prof. M. Rios Erazo, Faculty of Dentistry, University of Chile, Chile, [email protected] P533 WHAT MESSAGES DO PEOPLE WITH PSORIASIS TAKE AWAY FROM CARDIOVASCULAR DISEASE SCREENING? A QUALITATIVE STUDY A. Chisholm1, P.A. dr Nelson1, C.J. Pearce1, K Kane1, C. Keyworth1, C Chew-Graham2, C.E.M Griffiths3, L. Cordingley4 1 Dermatology Research Centre, University of Manchester, MANCHESTER, United Kingdom 2 Primary Care and Health Sciences, KEELE UNIVERSITY, United Kingdom 3 Dermatology Centre, Salford Royal NHS Foundation Trust and MAHSC, MANCHESTER, United Kingdom 4 Dermatology Research Centre, Health Psychology,University of Manchester, MANCHESTER, United Kingdom

S160

Introduction. Psoriasis is associated with unhealthy behaviours which increase the risk of cardiovascular disease (CVD). The screening for CVD risk and its management in the general population occurs in primary care. However, the effectiveness of screening is debated and it remains unknown how people with psoriasis experience such consultations. The study aimed to explore factors which influence the understanding and management of CVD risk in people with psoriasis. Method. 29 people with psoriasis who had taken part in an audiorecorded CVD risk assessment were interviewed in-depth about their perceptions of screening using ‘tape-assisted-recall’ to inform questioning. Data were analysed using Framework Analysis principles. Results. Patients reported three core perceptions as a result of taking part in screening: 1) little new or personally meaningful ‘knowledge and understanding about CVD risk’ was gained; 2) ‘strong emotional reactions to risk discussions’, were often experienced, including feeling burdened by the ‘work’ required to reduce risk particularly regarding weight loss; and 3) ‘minimal discussion with practitioners of ways to reduce CVD risk’during consultations. Patients expected practitioners to prompt risk management discussions and were passive in raising their concerns. Conclusions. The findings suggest that risk discussions during CVD screening consultations with people with psoriasis may be limited by practitioners not assessing patients’ pre-existing understanding of risk, addressing emotional reactions to risk discussions, or tailoring discussion about CVD risk and ways to reduce risk according to patients’ individual concerns. Facilitating practitioners to address these issues could improve the effectiveness of CVD screening consultations. CORRESPONDING AUTHOR: Dr. A.C. Chisholm, Dermatology Research Centre, University of Manchester, United Kingdom, [email protected] P534 EARLY DENTAL EXPERIENCES IN CHILDREN AND ITS EFFECT ON DENTAL ANXIETY AND ORAL HEALTH G. Rojas-Alcayaga1, L. Uribe1, A. Herrera1, M. Rios1, V. Zambrano2, A. Lipari1, Y. Molina1, P. Barahona1 1 University of Chile, SANTIAGO, Chile 2 Recreo Health Family Center, SANTIAGO, Chile INTRODUCTION: Children’s caries prevalence in Chile is 16, 8% in two-year-old children and 70, 4% in six-year-old children. The American Academy of Pediatric Dentistry recommends a first dental visit before 12 months of age, in order to enhance a suitable oral health status. The aim of this study was to evaluate the relationship between early dental experiences with oral health clinical outcomes and dental anxiety status when entering the sixyear-old oral health program. METHOD: 129 six-year-old children of both sex, were selected from the “Recreo”Health Center, Santiago, Chile. Oral health status was evaluated through the dmf-t index (dental caries experience), simplified oral hygiene index (OHI-S) and microbial score(S mutans). FIS scale and Frankl scale were used to assess dental anxiety. The early dental experience was informed by the child caretaker and classified as “without dental experience”, “preventive control”, “planned treatment”and “emergency for pain”. RESULTS: Children that had “planned treatment”and “emergency for pain”showed higher caries score than those with “preventive control”or “without dental experience”(Kruskal-Wallis, p<0.01) and only in “emergency for pain”the S mutans score was higher. No relationship was observed between type of experience and dental anxiety, nor between type of experience and OHI. CONCLUSIONS: Previous invasive dental experiences are related to higher caries susceptibility; however they have no effect on dental anxiety. On the other hand, early preventive dental sessions enhance a better oral health status in six-year-old children.

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

CORRESPONDING AUTHOR: Prof. G. Rojas, University of Chile, Chile, [email protected] P535 DOES INHIBITORY CONTROL TRAINING IMPROVE THE REGULATION OF HEALTH RISK BEHAVIOUR? A META-ANALYSIS V. Allom1, B. A. Mullan2, M. S. Hagger2 1 University of Sydney, SYDNEY, Australia 2 Curtin University, PERTH, WESTERN AUSTRALIA, Australia Introduction. Inhibitory control training refers to a behaviour change technique in which the ability to overrule impulsive reactions is improved in order to regulate behaviour in line with long term goals. A metaanalysis of inhibitory control training and health behaviour outcomes was conducted to determine the size and consistency of the effect of training on reducing harmful behaviours and to determine whether there were moderators of the effect; in order to facilitate the development of effective intervention strategies. Methods. A meta-analysis of 16 studies tested the effect of inhibitory control training on health behaviour outcomes. Moderators tested included type of behaviour, type of task, training duration and measurement of behaviour. Results. A small-to-medium effect of training on behavioural outcomes was found, however, the effect demonstrated significant heterogeneity. Moderation analyses revealed that type of task, type of behaviour and measurement of behaviour influenced the size of the training effect, while training duration did not. Conclusions. The present meta-analysis provides evidence that inhibitory control training results in differences in behavioural outcomes and provides elucidation of the optimal task parameters that result in behaviour change. While it appears that training duration does not influence the size of the effect and that the effects do not persist long after the training session, these elements need to be systematically examined in order to reach any firm conclusions. Determining the optimal length of training and whether these effects translate into everyday behaviour, could provide the basis for cost-effective and efficacious health promoting interventions. CORRESPONDING AUTHOR: V. Allom, University of Sydney, Australia, [email protected] P536 INSIGHTS INTO HEALTH DECISION-MAKING FROM A STUDY OF WHEAT AVOIDANCE P. Mohr1, S. Golley2, N. Corsini2 1 University of Adelaide, ADELAIDE, Australia 2 CSIRO Food Futures Flagship, ADELAIDE, Australia Introduction. This study was motivated by our previously reported findings that, coeliac cases excluded, 7% of Australian adults surveyed were avoiding wheat products on physiological grounds, without allergenic symptoms and predominantly without a formal diagnosis. The behavior was not predicted by neuroticism, illness worry, or emotional thinking. We report an intensive follow-up study designed to cast light on the clinical and psychosocial processes underlying the decision to avoid eating wheat and other foods. Method. We conducted qualitative interviews exploring the experiences and dietary practices of thirty-five self-identified wheat avoiders (31 female) recruited through newspaper advertisements. Results. Self-prescribed avoidance of wheat, frequently in association with other foods, was commonly a response to dissatisfaction with traditional medical diagnoses of symptoms or treatment options. Use of supplements was also common, and was for reasons other than replacement of nutrients lost. Conclusions. Although physiological explanations for adverse reactions to wheat or gluten are current subjects of research, the present findings

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

have broader implications for understanding of health decision-making. They suggest that, when definitive medical answers are unavailable, the propensity exists for a self-managed approach to health, seen here in elimination of suspect foods and self-administration of unregulated or natural medications. We discuss the possible role of simple heuristics in decisions by health-motivated individuals and the potential for paradoxical increases in risk of nutritional deficiencies or delayed diagnosis and treatment of any underlying medical conditions. We also consider theoretical implications of self-management of health for individuals’ receptiveness to public health interventions. CORRESPONDING AUTHOR: Prof. P. Mohr, University of Adelaide, Australia, [email protected] P537 PSYCHOSOCIAL PREDICTORS OF EXCLUSIVE BREASTFEEDING DURATION TO SIX MONTHS POSTPARTUM E. De Jager, J. Broadbent, M. Fuller-Tyszkiewicz, C. Nagle, S. McPhie, H. Skouteris Deakin University, MELBOURNE, Australia Introduction. The World Health Organization recommends that for optimal growth and development, all infants worldwide should be exclusively breastfed for the first six months of life. Previously, studies have shown that psychosocial factors such as self efficacy are important for a woman’s ability to maintain exclusive breastfeeding However to date, there has been limited research examining these factors, specifically for exclusive breastfeeding and for a duration beyond three month postpartum. Method. In this longitudinal study, 125 pregnant women completed questionnaires at three time points; 32 weeks pregnancy, two and six months postpartum. The questionnaire measured psychosocial variables such as breastfeeding self-efficacy, body attitude, psychological adjustment, attitude towards pregnancy, intention, confidence and motivation to exclusively breastfeed and importance of exclusive breastfeeding. Exclusive breastfeeding behaviour up to six months postpartum was also measured. Results. The results of the path analyses showed that at 32 weeks gestation a woman’s confidence to achieve exclusive breastfeeding was a direct predictor of exclusive breastfeeding duration to six months. At two months postpartum, psychological adjustment and breastfeeding selfefficacy were predictive of exclusive breastfeeding duration. Finally, at six months postpartum, psychological adjustment, breastfeeding selfefficacy, confidence to maintain and feeling fat were directly predictive of exclusive breastfeeding duration. Conclusion. Psychosocial factors self-efficacy, psychological adjustment, body image, motivation and confidence are important psychosocial factors for the maintenance of exclusive breastfeeding to six months. Therefore, antenatal breastfeeding education and postpartum support may be strengthened by strategies that build a woman’s confidence to exclusively breastfeed. CORRESPONDING AUTHOR: Miss E. De Jager, Deakin University, Australia, [email protected] P538 INCIDENTAL SUN EXPOSURE: USING THE PROTOTYPEWILLINGNESS (PW) MODEL TO EXPLAIN THE UNEXPLORED K. Morris, A.L. Swinbourne, S.L. Harrison James Cook University, TOWNSVILLE, Australia Introduction: Compared to deliberate tanning and sun protection, the psychosocial factors surrounding incidental sun exposure are relatively unexplored. Unlike other sun-related behaviours, incidental sun exposure is not reasoned or planned thus novel theoretical frameworks are required for its exploration. The current study aimed to determine whether the Prototype Willingness (PW) model provides a useful framework for the

S161

investigation of incidental sun exposure. Method: An experimental 1factor design with three levels aimed to manipulate the favourability of the incidental exposer prototype. Community participants (n = 96) were exposed to either a positively or negatively framed fabricated newspaper article about the prototypical person who incidentally exposes themself to the sun. Those in the control condition read an article about an unrelated health behaviour. Participants also responded to items assessing PW model variables and sun-related behaviour. Skin reflectance spectrophotometry was used to measure skin colour at point of recruitment and again at approximately 1-month follow up. Results: Preliminary findings after wave 1 recruitment suggest that the manipulation was effective. Withingroup differences were detected for prototype favourability ratings. Conclusions: Current findings suggest that the PW model, particularly the social reactive pathway provides a useful framework for the exploration of incidental sun exposure behaviour. CORRESPONDING AUTHOR: Ms K. Morris, James Cook University, Australia, [email protected] P539 LONG-TERM EFFECT OF MUSICAL DISTRACTION ON DENTAL ANXIETY IN CHILDREN AND DENTAL CHAIR BEHAVIOR P.E. Espinoza1, A. Herrera2, M. Ríos2, P. Barahona2, Y. Molina2, P. Cadenasso3, V. Zambrano4, G. Rojas2 1 Universidad de Chile, SANTIAGO DE CHILE, Chile 2 Behavioral science area, Faculty of Dentistry, Universidad de Chile, SANTIAGO DE CHILE, Chile 3 Alfonso Leng Dental Care Center, SANTIAGO DE CHILE, Chile 4 Recreo Care Center, SANTIAGO DE CHILE, Chile Introduction: Patient’s behavior is primordial in the development and prevention of oral diseases. Dental anxiety is a multifactorial process that can be a barrier to receive care. Researchers support that musical distraction could be an effective mechanism on dental anxiety and pain reduction. The goal of this research was to study the long-term effect of musical distraction on dental anxiety and dental child behavior in children from Santiago, Chile. Method: 109, 6 year-old participants, were randomized into two groups: the experimental group received standard dental care with musical distraction and the control group received only standard dental care. FIS scale was used for the measurement of dental anxiety; while Frankl Behavior Scale was used for dental chair behavior. Measurements were taken at baseline, end of treatment and six months later. Analysis was performed using Mann-Whitney U test. Results: In General children presented low level of dental anxiety. There is no effect of musical distraction on child dental anxiety neither at the end of the treatment (N = 109; U = 1475; p = 0.949) nor six months later (N = 77; U = 701; p = 0.638). There is no effect of musical distraction on child behavior in dental chair neither at the end of the treatment (N = 109; U = 654; p = 0.506) nor six months later (N = 77; U = 654; p = 0.506). Conclusions: Musical distraction is not an effective method for dental anxiety management in children with low level of dental anxiety. CORRESPONDING AUTHOR: Dentist P.E. Espinoza, Universidad de Chile, Chile, [email protected] P540 THE PREDICTIVE VALUE OF MINDFULNESS AND SELFCOMPASSION FOR PSYCHOLOGICAL WELL-BEING OVER ONE YEAR TIME A. López Angarita, R. Sanderman, M.J. Schroevers University Medical Center Groningen, GRONINGEN, Nederland Introduction: The benefits of mindfulness and self-compassion for psychological well-being are increasingly being examined, in observational as well as intervention studies. Prior research, mainly cross-sectional,

S162

usually examined these concepts separately. Therefore, little is still known about their interrelationships and unique predictive value for psychological well-being. Given that the definitions and self-report assessments of mindfulness and self-compassion partly show similar aspects, we aimed to examine the predictive value of mindfulness and self-compassion for psychological well-being over one year time. Method: A large community-based sample of 758 adults completed a series of self-report questionnaires at two time points with a year in between. Correlation and regression analyzes were conducted to compare the predictive role of self-reported mindfulness (total score of the FFMQ, Baer et al. 2006) and self-compassion (total score of the SCS, Neff, 2003) for distinct indicators of psychological functioning (i.e. depressive symptoms, CES-D, Radloff, 1977; negative affect and positive affect, PANAS, Watson, Clark & Tellegen 1988). Results: Correlation analyses showed that mindfulness and selfcompassion were both significantly correlated with all three indicators of psychological functioning over a one year time. In regression analyses, hereby controlling for baseline levels of psychological functioning, mindfulness predicted greater variance than self-compassion for positive affect, whereas self-compassion predicted greater variance for depressive symptoms and negative affect. Conclusions: Our findings suggest that a total score of self-compassion better predicts negative indicators of long-term psychological functioning, whereas a total score mindfulness better predicts long term positive affect. CORRESPONDING AUTHOR: A. López Angarita, University Medical Center Groningen, Nederland, [email protected] P541 THE LONG TERM EFFECTIVENESS OF SELF-REGULATION INTERVENTIONS ON WELL-BEING OF INDIVIDUAL CLIENTS H. Koelewijn, N. Kuipers PCC Health Promotion, AMERSFOORT, Nederland

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

have a great impact on health care, economy and society alike. Leisure, an important element of life-style, could be the easiest way to improve well being, the mediator between health and lifestyle, and to promote health. We examined whether people are aware of what they themselves can do about their own health and what their attitudes toward leisure activity are. Methods: In two consecutive studies, 388 randomly selected laymen, and 512 sport-recreation students, respectively, were evaluated by a selfcreated questionnaire about their attitudes and motivations toward leisure activity with an emphasis on physical fitness. Results: Participants conceptualized well-being as harmony, health/ fitness and quiet life with happiness. Of these, health and fitness might be the primary “target”of leisure. In fact, the majority of the participants preferred sport related activities as a tool for improving well-being, whereas traditional values - e.g. health related activities, spiritual life, successful working - remained unnoticed. Conclusion: Participants were definitive about the necessity of regular leisure activity and looked at it as an enjoyable form of improving wellbeing. Reality, however, is different. Physical activity seems to be startlingly inferior: only 9% of the population walks or hikes, and only 7% practices physical exercise regularly. CORRESPONDING AUTHOR: Prof. G. Bárdos Eötvös, Loránd University, faculty of Education and Psychology, Hungary, [email protected] P543 TIME TERNDS IN FOOD HABITS AND THEIR RELATION TO SOSCIOECONOMIC STATUS AMONG ADOLESCENTS IN nORWAY A.S. Fismen, R.S. Smith, T.T. Torsheim, O.S. Samdal Univeristy of Bergen, BERGEN, Norway

Background: The aim of this study is to evaluate the long term effects of self-regulation interventions (SRI), including the principles of Maes & Karoly (2005), on well-being of individual clients in a psychology practice. The intervention group received a care program of five meetings. The evaluation took place directly after the intervention and on a longer term. Methods: Socio-demographics and psychological well-being were assessed at baseline (T1), after the intervention (T2; N=400) and 6 months after the intervention (T3; N=250). Repeated measures ANOVA was used to evaluate the effects. Results: Based on the analyses, we found that psychological complaints decreased after the intervention. This effect was found between T1 and T2, and also between T1 and T3 . Furthermore, there was no significant difference found in psychological complaints between T2 and T3, which means complaints remained stable over time. Conclusion:Previous studies showed that SRI can positively influence psychological well-being in individual clients. Long term effects on SRI have not been studied before. Therefore, this study can shed new light on SRI and give directions for future research.

Background: In recent years, adolescents’ food habits have become a major source of concern. Substantial policy and intervention efforts have been made to influence young people to consume more fruit and vegetables and less sweets and soft drinks. Particular attention has been devoted to the social gradient in food habits, aiming to reduce dietary inequality. However, few internationally published studies have evaluated trends in teenagers’ food habits, or investigated how dietary inequalities develop. Methods: We used Norwegian cross-sectional data from the international Health Behaviour in School-Aged Children (HBSC) study, collected via three nationally representative and comparable questionnaire surveys in 2001, 2005 and 2009. Results: The analyses indicated an overall positive trend in food habits among adolescents in Norway. Further, our analyses indicated that the socio-economic differences were stable in the period 2002 - 2010, with uniform improvement in fruit and vegetable consumption across all SES levels. No significant associations between SES and intake of sweets and sugar-added soft drink were found. Conclusion: The study identifies an overall improvement in diet among adolescents over a period characterized by onset of as well as ongoing initiatives targeting young people’s food habits. However, the observed socioeconomic gradient in fruit and vegetable consumption remained unchanged.

CORRESPONDING AUTHOR: Dr. H. Koelewijn, PCC Health Promotion, Nederland, [email protected]

CORRESPONDING AUTHOR: PhD student AS Fismen, Univeristy of Bergen, Norway, [email protected]

P542 MANAGING RISK FACTORS: WELL BEING AND LEISURE G. Bárdos, J. Ábrahám Eötvös Loránd University, faculty of Education and Psychology, BUDAPEST, Hungary

P544 ACTIVE COMMUTING AND CARDIOVASCULAR RISK MARKERS IN CHILDREN A.M. Machado-Rodrigues1, M.M. Ferrão2, A. Gama2, I. Mourão3, H. Nogueira2, V. Rosado2, C. Padez2 1 University of Coimbra, VISEU, Portugal 2 Research Centre for Anthropology and Health, University of Coimbra, COIMBRA, Portugal 3 Department of Sport and Physical Education, UTAD, VILA REAL, Portugal

Introduction: It has long been established that risk factors of psychosomatic diseases are elements of lifestyle. Smoking, alcohol consumption, nutrition, obesity, immobility - just to mention but a few - contribute to the formation of many diseases. It is clear that managing risk factors may

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

Background: The positive impacts of active travel on physical activity (PA) levels and cardiovascular health are well established in adults, but still require further research, especially in youth which are increasing the risk of obesity and metabolic problems. Therefore, this study aimed to analyse the associations between active travel to school and cardiovascular risk markers in children. Methods: The sample comprised 665 children (345 boys) aged 7-9 years. Height, weight, and skinfolds were collected as well as data on cardiovascular risk markers (resting heart rate, diastolic and systolic blood pressure). Information on mode and duration of travel to school was gathered by questionnaire. Outcome variables were statistically normalized and expressed as Z scores. A cardiovascular risk score was computed as the mean of the Z scores. Logistic regressions, with adjustments for age, sex, BMI, PA, and parental education were used. Results: Children walking or bicycling to school were significantly less likely to have lower clustered cardiovascular risk scores than their passive commuting counterparts. The final regression model also indicated that those walking or bicycling to school were significantly more likely to have higher levels of PA than those who usually travel by motor transports; level of education of parents was also associated with active commuting in children. Conclusion: The present study showed an independent association between active commuting and the clustering of cardiovascular risk in children aged 7-9 yrs. These findings may be useful for policy makers and city planners when designing neighborhoods that promote physical activity. CORRESPONDING AUTHOR: Dr M. Machado-Rodrigues, University of Coimbra, Portugal, [email protected] P545 EFFECTIVENESS OF GROUP COGNITIVE BEHAVIORAL THERAPY FOR THE PREVENTION OF DEPRESSION: PILOT STUDY J. Tayama Nagasaki University, NAGASAKI, Japan Background & Aims: Harm Avoidance (HA) is a personality dimension in Cloninger's Temperament and Character Inventory (TCI) that is characterized by pessimism, anxiety, and shyness. It is a risk factor for depression. In this pilot study, we conducted group cognitive-behavioral therapy (GCBT) with people who scored high in HA, and examined the impact of the intervention in terms of preventing depression. Methods: GCBT consisted of six sessions of 60-90 minutes each, and was carried out over a period of six weeks for each of four groups. GCBT content included relaxation exercises, cognitive behavioral theory, cognitive reframing, attribution therapy, and assertion training. Data analysis was performed using the data from the 12 participants with no missing data. Results: A period effect was observed for scores of positive self-image (F (4, 15) = 2.83, p < 0.05). Post hoc tests revealed higher scores for positive self-image immediately following the intervention and at six months after the intervention. However, positive self-image scores did not remain elevated at the one-year follow-up point. Conclusion: The reason why participants maintained improvement in cognition even six months after the intervention may be that participants did not have severe depressive symptoms at baseline. Results suggest that this intervention may promote functional improvement of the prefrontal cortex, which is responsible for cognitive function. CORRESPONDING AUTHOR: J. Tayama, Nagasaki University, Japan, [email protected]

S163

P546 'TIME TO NUDGE?' USING CHOICE ARCHITECTURE TO UNDERSTAND HEALTH PROMOTION SIGNPOSTING FOR PEOPLE WITH PSORIASIS C. Keyworth1, P.A. Nelson1, L. Cordingley2, C.E.M. Griffiths3, C. Bundy1 1 Dermatology Research Centre, University of Manchester, MANCHESTER, United Kingdom 2 Dermatology Research Centre, Health Psychology,University of Manchester, MANCHESTER, United Kingdom 3 Dermatology Centre, Salford Royal NHS Foundation Trust and MAHSC, MANCHESTER, United Kingdom Introduction. Lifestyle behaviour change (LBC) is an important aspect of psoriasis management. Traditional approaches to LBC aim to improve attitudes and knowledge about LBC through written and visual information; with often modest effects on behaviour. Recent studies show healthrelated behaviour can be triggered by environmental cues. If and how LBC signposting within health care centres is conforming to such recommendations is unclear. We performed an observational study to investigate how general and psoriasis-specific LBC information is made available to patients in primary and secondary care health centre waiting areas. Methods. An observation schedule was designed to record prevalence and quality of leaflets and posters signposting LBC. Content analysis was used to analyse the frequency, characteristics and standard of LBC materials in health centres in NW England. Results. A total of 262 sources of LBC information were recorded (median per site = 10, range= 0-40) across 24 health centres. These were mainly: generic posters/displays of LBC support (n = 113; 43.1%); and generic materials available in waiting areas (n = 98; 37.4%). Information was of poor quality and poorly displayed. No high quality psoriasisspecific patient materials were identified. Conclusion. There is an urgent need to provide better LBC signposting to patients with psoriasis in the health centre environment. Evidence about the use of environmental cues to prompt behavioural change should inform the design and display of LBC information. Future work should examine whether this can induce sustained changes to improve long-term health. CORRESPONDING AUTHOR: Mr C Keyworth, Dermatology Research Centre, University of Manchester, United Kingdom, [email protected] P547 ARE DERMATOLOGY HEALTH PROFESSIONALS TRAINED TO DELIVER HEALTH BEHAVIOUR CHANGE SUPPORT FOR PATIENTS WITH PSORIASIS? C. Keyworth1, P.A. Nelson1, A. Chisholm1, C.E.M. Griffiths2, L. Cordingley3, C. Bundy1 1 Dermatology Research Centre, University of Manchester, MANCHESTER, United Kingdom 2 Dermatology Centre, Salford Royal NHS Foundation Trust and MAHSC, MANCHESTER, United Kingdom 3 Dermatology Research Centre, Health Psychology,University of Manchester, MANCHESTER, United Kingdom Introduction. Psoriasis is a chronic inflammatory skin condition. Unhealthy lifestyle behaviours contribute both to psoriasis onset and severity. Health professionals managing patients with psoriasis are well placed to support lifestyle behaviour change (LBC), but few feel confident to do so. The extent to which health promotion and LBC skills are included within training curricula in the UK is unclear. We systematically examined post-graduate training competencies across general and dermatology specialist health professionals for evidence of LBC skills. Methods. Core curricula documents from professional organisations were content analysed to examine the extent to which curricula: a) mentioned

S164

health promotion and LBC as part of the professional role; and / or b) included health promotion and LBC as explicit training competencies or requirements for qualification. Results. Of 11 core curricula documents analysed, 67 occurrences of terms related to LBC and health promotion were found. Most appeared in the General Practitioner curriculum (n=42; 62.7%), followed by the Dermatology Specialist Nurse curriculum (n=14; 20.9%) and Dermatologist curriculum (n=11; 16.4%). None were found in the General Practitioner with a Special Interest in Dermatology curriculum. Recommendations relating to LBC knowledge, skills and attitudes were not clearly specified in any documents and only described basic level competencies. Conclusion. Core curricula must provide health professionals with the necessary evidence-based knowledge, skills and attitudes to support LBC in patients with psoriasis. This has particular resonance given the evidence linking unhealthy lifestyles with psoriasis outcomes. CORRESPONDING AUTHOR: Mr C. Keyworth, Dermatology Research Centre, University of Manchester, United Kingdom, [email protected] P548 FACTORS ASSOCIATED WITH THE IMPLEMENTATION OF OBESITY PREVENTION PRACTICES IN AUSTRALIAN CHILDCARE SERVICES L. Wolfenden1, M. Finch2, S.L. Yoong1, N. Nathan2, N. Waever1, J. Jones1, P. Dodds1, R. Wyse1, R. Sutherland2, K. Gillhan2, J. Wiggers1 1 The University of Newcastle, NEWCASTLE, Australia 2 Hunter New England Population Health, NEWCASTLE, Australia Introduction: Internationally, research suggest that many childcare services do not implement recommended policies and practices supportive of obesity prevention. The study assessed whether a number of theoretically-based factors were associated with implementation of healthy eating and physical activity policies and practices in this setting. Methods: A telephone-based survey was conducted with all Service Managers of centre based childcare services in New South Wales, Australia. The survey assessed implementation of a range of healthy eating and physical activity practices recommended by best practice guidelines for the sector as well 13 factors suggested by Damschroder’s Consolidated Framework for Implementation Research to impede or promote implementation. Results: Service Managers who agreed that the physical activity policies and practices of their service needed to be improved were significantly more likely to report practice implementation (OR=2.3; 95% CI 1.2-4.3; p=0.01), as were those agreeing that their management committee (OR=3.2; 95% CI 1.2-8.4; p=0.02) or parents (OR=3.1; 95% CI 1.3-7.4; p=0.01) were supportive of implementing program initiatives. Furthermore, those that agreed that external resources to help implement the program initiatives were accessible (OR= 2.4; 95% CI1.0-6.3; p=0.04) were more likely to be implementing all program policies and practices.Conclusions: The findings highlight opportunities for improving implementation of obesity prevention interventions in this setting by developing interventions that address such factors. CORRESPONDING AUTHOR: Dr L. Wolfenden, The University of Newcastle, Australia, [email protected] P549 EFFICACY OF AN ONLINE PLANNING INTERVENTION TO REDUCE CARDIOVASCULAR RISK BEHAVIOR D.A. Reinwand1, H. De Vries1, J. Wienert2, V. Storm2, T. Kuhlmann2, S. Lippke2 1 Maastricht University, MAASTRICHT, Nederland 2 Jacobs University JCLL, BREMEN, Germany Introduction: CVD diseases are a major health problem in western countries and cause 30% of the global death. A healthy nutrition and sufficient

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

physical activity can reduce CVD risk. Computer tailored (CT) interventions have been shown to be effective in increasing physical activity as well as fruit and vegetable consumption. The goal of this presentation is to report the effects of a CT program aimed at fruit and vegetable consumption and physical activity after two months. Methods: Based on the HAPA Modelan eHealth planning intervention was developed and implemented with the aim to increase fruit and vegetable consumption and physical activity. The intervention encouraged participants to define individual health behavior goals as well as action, and coping plans to reach these self-determined goals. The effectiveness of the program was compared between the intervention condition and the waiting control group in terms of behavior change, antecedents of behavior change e.g., self-efficacy and motivation. Results: At baseline, of all participants (n=337) 51.1% did not eat 5 portions fruit and vegetable a day and 64% did not meet the recommendation to be physical active for 30min/day at 5 days. We expect that the intervention group will increase the target behaviors after completing the intervention. Further effect outcomes will be presented. Conclusion: The results will give insight in the efficacy of an online planning intervention. CORRESPONDING AUTHOR: D.A. Reinwand, Maastricht University, Nederland, [email protected] P550 EFFECTS OF A VIDEO-BASED EDUCATIONAL PROGRAM ON OSTEOPROROSIS-PREVENTING BEHAVIORS AND BONE DENSITY OF SEDENTARY WOMEN Jen-Jiuan Liaw, H.C.H. Hsiao National Defense Medical Center, TAIPEI, Taiwan Aim: The study was to investigate the effects of a video-based educational program on preventive behaviors and bone density of sedentary women, and to explore the factors related to osteoporosis-preventive behaviors and bone density. Methods: This study used a repeated measures experimental design, and convenience sampling to recruit women with office work in Taipei. We screened 324 women with an ultrasound bone density scanner and selected 138 high-risk women, who were randomly assigned to the control (n=76) and experimental group (n=62). The women in the experimental group received an osteoporosis-prevention education. After the education, we measured the women’s osteoporosis-preventive behaviors at the 6th and 12th weeks, and measured their bone density at the 12th weeks. Data was analyzed using descriptive and inferential statistics. Results: The results showed that the experimental group had better osteoporosis-preventive behaviors and ultrasound bone density than the control group after the osteoporosis-prevention video-based education, but without statistical significance. Changes in osteoporosis-preventive behaviors were negatively related to sedentary behavior. Changes in ultrasound bone density were negatively related to the drinks of carbonated beverage after the osteoporosis-prevention video-based educational program. Conclusion: Our osteoporosis-prevention video-based educational program is helpful to the sedentary women in improving osteoporosispreventive behaviors and ultrasound bone density. CORRESPONDING AUTHOR: Professor Jen-Jiuan Liaw, National Defense Medical Center, Taiwan, [email protected] P551 LONG-TERM REHABILITATION GAINS IN ORTHOPEDIC PATIENTS TO RETURN TO WORK: THE ROLE OF PHYSICAL EXERCISE V. Storm1, J. Paech1, J. Ziegelmann2, S. Lippke1 1 Jacobs University Bremen, BREMEN, Germany 2 German Centre of Gerontology, BERLIN, Germany

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

Introduction: Return to work is one of the primary goals after rehabilitation. However, many patients do not resume work after discharge or intend to retire early. Many studies show that individuals benefit from physical exercise recommendations given to them during rehabilitation. Still, little is known about long-term behavior maintenance, recovery and rehabilitation benefit on healthy aging and employability. Methods: Outpatients with different orthopedic ailments were recruited during medical rehabilitation. Patient characteristics, exercise frequency, social cognitive variables and occupational outcomes were investigated via self-report questionnaires and interviews over a period of 8 years. The final longitudinal sample consisted of 367 people, aged 18 to 83. Correlation and regression analyses were used to identify predictors of employability after discharge. Results: More frequent exercise was associated with better overall health status on all measurement points (r=.31, p<.001). More frequent exercise was also related to the feeling of being able to return to work eight years after rehabilitation (r=.25, p=.001). Regression analysis results confirmed the predictive power, even when controlling for the intention to return directly after rehabilitation (beta=.21, p=.006). However, exercise was significantly contributing to the perceived ability to return to work in men only. Conclusion: Orthopedic patients can benefit from medical rehabilitation also in the long run. Those who exercise more frequently as compared to before rehabilitation are more likely to feel able to work even eight years after rehabilitation. Exercise seems to play an important role in employability in the long run and needs to be promoted in a gender-specific way. CORRESPONDING AUTHOR: V. Storm Jacobs, University Bremen, Germany, [email protected] P552 THE INFLUENCE OF TAILORING ELEMENTS IN AUDITORY HEALTH PERSUASION AIMED AT FRUIT AND VEGETABLE INTAKE S.P. Elbert, A. Dijkstra University of Groningen, GRONINGEN, Nederland Introduction. Tailoring information to specific individual characteristics has been proven to be more effective compared to non-tailored information (e.g., Hawkins, Kreuter, Resnicow, Fishbein, & Dijkstra, 2008). In the current online study, tailoring will be applied via the auditory mode of communication, as technological developments facilitate the use of auditory channels, also in health promotion. Elements that will be tested are personalization (include one’s first name), feedback (present information on personal state) and adaptation (e.g., to a specific value). Methods. The influence of three distinct tailoring elements on persuasion (both intention and self-reported fruit and vegetable intake) was investigated in an experimental between-participants design. A fourth condition with a generic auditory health message was included. The study consisted of a pre-test, the auditory health message with an immediate posttest measuring intention (after > 7 days) and a behavioral follow-up measurement two weeks after exposure to the message. Results. 112 respondents completed the study (80 % women, Mage = 24). A significant main effect of tailoring was found on intention, showing high intentions after receiving feedback and decreased intentions after the personalized message. This pattern was mostly present in people who perceived their own health as relatively poor. Furthermore, tailoring has an effect on self-reported behavior in people who perceived the behavior as relatively difficult. Conclusion. Tailoring elements can influence health behavior in auditory forms of health communication as well. Insights from the current study can be used to develop tailored health interventions as applied via the auditory mode of communication. CORRESPONDING AUTHOR: S.P. Elbert, University of Groningen, Nederland, [email protected]

S165

P553 THE NURSES' DILEMMAS ABOUT OBESITYAND ITS TREATMENT: RESULTS FROM A QUALITATIVE STUDY A. Maia1, Filipa Teixeira2, José Luis Pais-Ribeiro2 1 University of Minho, BRAGA, Portugal 2 University of Porto, PORTO, Portugal Background: Although nurses play a key role in providing support and care to obese patients, there are few studies, with no conclusive results, about their beliefs, attitudes and practices in primary care setting and how they perceive their role in obesity management. Quantitative research has been criticized for not being able to clarify how these perceptions influence the practices and behavior of healthcare providers. This qualitative study was aimed to understand what nurses in primary care setting think about obesity and obese people and how they perceived their role in the change process. Method: Semi-structured interviews concerning beliefs, attitudes and practices about obesity were conducted with Portuguese nurses and data was analyzed according to Thematic Analysis’s procedures. Results: The main themes indicate a set of dilemmas in nurses speech. They hold negative beliefs and attitudes towards obese, who are described as unmotivated, noncompliant and demonstrating a passive coping. This is viewed as an intervention barrier out of nurses control which raises feelings of frustration and lower expectations of positive outcomes. However, the nurses practices are characterized by persistency. They demonstrate an active role, perceived themselves as efficient and being able to positively modify obese motivation and believe in the success of the interventions, which, however, are described as a constant struggle between them and the patients. Conclusion: Nurses experienced a dilemma concerning treatment responsibility: obese are recognized as central in disease management but also a potential barrier to treatment. Education and training in motivation strategies should be improved. CORRESPONDING AUTHOR: A. Maia, University of Minho, Portugal, [email protected] P554 THE EVALUATION OF A PATIENT DECISION AID FOR ANTIRHEUMATIC DRUGS I. Nota, C.H.C. Drossaert, E. Taal, M.A.F.J. Van de Laar University of Twente, ENSCHEDE, Nederland Introduction: Research among Dutch arthritis patients has shown that quality improvement is needed in the field of ‘information about medication and treatment’ and that they want to be more involved in treatment decision-making. To these aims a Patient Decision Aid (PtDA) for antirheumatic drugs was developed. Method: The PtDA was evaluated in a quasi-experimental questionnaire study among 281 arthritis patients who were facing the decision to start (a different) drug. The questionnaire included questions on, among others, patient’s perceived role, knowledge, and satisfaction with the decision process.Patients from the intervention group were also questioned about the use of the PtDA, its perceived usefulness, its perceived ease of use, and its perceived effectiveness. Results: The PtDA was used by 57% of respondents in the intervention group. Younger (p=0.003) and higher educated (p=0.004) patients were more likely to have used the PtDA. Users were on average very positive about the PtDA: they rated the PtDA as useful, easy to use and easy to understand. Compared to the control group, patients in the intervention group perceived a more active role in decision-making (p=0.01) and perceived the final choice to be more consistent with their personal values (p=0.02). A significant increased knowledge was revealed among users of the PtDA (p=0.03). No effects were found on satisfaction, trust in physician, beliefs about medication, or adherence to medication.

S166

Conclusion: Our PtDA can be a valuable aid in improving patient participation in medical decision-making about anti-rheumatic drugs. CORRESPONDING AUTHOR: I. Nota, University of Twente, Nederland, [email protected] P555 IMPACT OF PATIENT-ACCESSIBLE ELECTRONIC MEDICAL RECORDS IN RHEUMATOLOGY: USE, SATISFACTION AND EFFECTS ON EMPOWERMENT R. van der Vaart1, C.H.C. Drossaert1, E. Taal1, M.A.F.J. Van de Laar2 1 University of Twente, ENSCHEDE, Nederland 2 Arthritis Centre Twente, ENSCHEDE, Nederland Objectives: To measure the use, satisfaction and impact on empowerment of a web portal which provides patients with rheumatoid arthritis home access to their electronic medical records (EMR). Methods: A pretest-posttest study was conducted among 360 patients. Questionnaires assessed patients’ socio-demographics, health literacy, Internet use, disease characteristics, and empowerment before and after launching a hospital-based patient web portal. To measure empowerment, patients’ satisfaction with care, trust in their rheumatologist, self-efficacy in patient-provider communication, illness perceptions, and medication adherence were assessed. The post-test included questions on portal use, satisfaction, and self-perceived impact due to portal use. Results: 54% of respondents with Internet access had viewed their EMR. Respondents were positive about the ease of use and usefulness of the portal and reported very few problems. Age (P=.03), amount of Internet use (P=.01) and self-perceived Internet skills (P=.03) significantly predicted portal use. Of the respondents who had logged in, 44% reported feeling more involved in their treatment and 37% felt they had more knowledge about their treatment. Significant differences over time were not found on the empowerment-related instruments. Conclusions: The current portal succeeded in offering patients access to their EMR in a usable and understandable way. While its true impact is difficult to grasp, a relevant portion of the patients felt more empowered due to the web portal. Offering patients home EMR access appears to be a valuable addition to the care process, with the potential of involving patients more in their own treatment. CORRESPONDING AUTHOR: Dr R. van der Vaart, University of Twente, Nederland, [email protected] P556 VERBAL MEDIATION OF WELL-BEING EFFECTS OF THE EXPRESSIVE WRITING EXPERIENCE L. Sibilia SIMPS, ROMA, Italy Lucio Sibilia, Sapienza University of Rome, Dept. of Psychology, Rome, Italy. Introduction. Albeit it is known that Pennebaker's Emotional Writing Therapy (EWT) has a lasting effect in improving well-being, the mechanism with which this occurs is still unclear. In this study, we tested the hypothesis that emotional expression mediates such effect. Method. 120 undergraduate students were enrolled and randomly divided into 3 groups of 40 SS each. The first underwent three EWT sessions about own traumas, the second underwent three sessions of mood enhancing writing exercise, and the third was invited to write about neutral experiences. Both at baseline and 40 days after the end, the scores of well-being were gathered with the Ryff WB Scale (RWBS) and of depression with the Beck Depression Inventory (BDI). All the written reports were also assessed with a linguistic analysis software for the frequency of emotional words (positive, negative) as measures of emotional expression, and total word count as a measure of descriptive detail. Results. Only SS in the first

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

group exhibited a reduction in BDI and a marked improvement in RWBS scores. In contrast with the hypothesis, the number of both positive and negative emotional expressions were not correlated with favourable changes in well-being and depression, rather it was the length of the report (total n. of words). Conclusions. Other hypotheses are therefore necessary, which will be discussed. References. Cirillo M., Colasanti A.R., Sibilia L. (2012) Promuovere il benessere attraverso l'espressione delle emozioni in forma scritta. Orientamenti pedagogici, vol.59, n.3, pagg. 481-492. CORRESPONDING AUTHOR: Prof. L. Sibilia, SIMPS, Italy, [email protected] P557 HIV/AIDS PRIMARY PREVENTION PROGRAM WITH CARIBBEAN YOUTH J.S. Wodarski The University of Tennessee, KNOXVILLE, United States of America Introduction. The project consists of a program of outreach, pretreatment, and treatment services designed to reduce high-risk HIV-related behaviors among substance using young adults (ages 18-24) in the U.S. Virgin Islands (USVI), and ultimately to reduce HIV among this very vulnerable population. The need for HIV prevention programming in the USVI is tremendous. The incidence of HIV/AIDS in the Caribbean region with an HIV prevalence of 1.6% is the second highest in the world, second only to sub-Saharan Africa (UNAIDS, 2006). The Caribbean’s status as the second-most affected region in the world masks substantial differences in the extent and intensity of its epidemics (UNAIDS, 2005). Within the U.S. Virgin Islands, HIV incidence has continued to increase, and for the third year in a row the USVI leads the nation with the highest number of confirmed HIV cases at 272.7 per 100,000 US citizens (CDC, 2007). Methods. The program is based on a package of evidence-based interventions, including outreach, motivational interviewing and the Community Reinforcement Approach (CRA). The program was implemented specifically on the island of St. Thomas and expanded to St. John and St. Croix in the United States Virgin Islands. Service expansions provided culturally and age relevant outreach services to over 5,000 young adults, pretreatment services to 1,000 young adults, and 356 young adults received early intervention enhanced substance abuse treatment services over the five-year span of the program. Summary of Results and Conclusion. Study was successful providing outreach, motivational interviewing and the Community Reinforcement Approach (CRA). CORRESPONDING AUTHOR: Professor J.S. Wodarski, The University of Tennessee, United States of America, [email protected] P558 FERTILITY PLANNING AND SAFER CONCEPTION WITH HIV: PATIENT AND PROVIDER PERSPECTIVES D. Jones, R. Cook, J.E. Potter, N. Chakhtoura, A. Coll, A. Spence University of Miami Miller School of Medicine, MIAMI, United States of America Background: Conception without risking HIV transmission to partners requires significant planning, yet, patients and providers may fail to discuss fertility desires or practices during visits. Understanding patient and provider perspectives on fertility planning can help patients make informed decisions about safer conception methods. This pilot study describes knowledge, attitudes and practices regarding fertility and conception among women, partners and providers in Florida. Methods: Women (n = 32) were assessed regarding the relative importance of finances, opinions of partners and family, the risk of HIV to partners, and their doctor’s opinion on fertility desires using a conjoint questionnaire and a 10 item scale. A subset of women (n = 19), partners (n = 11)

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

and providers (n = 14) completed qualitative interviews. Results: Women (n=32) were aged 18-45 (m=36±7) and most were African American (72%). The majority of participants had children (69%) and planned to have more (78%). Using conjoint analysis, provider’s opinion, partner’s opinion, and family’s opinion emerged as the most influential factors in fertility desires (mean “importance scores”24.0%, 23.8%, 21.4%); transmission was less influential. Women expressed unwillingness to discuss fertility desires with providers, partners and family members; providers relied on patient initiation of the topic. Both women and partners had little information on safer conception.Conclusion: Results suggest women value doctor’s and partner’s opinions on fertility planning, but discussions are infrequent. Providers should incorporate fertility planning in care, and discuss strategies to prevent transmission. Opening the dialogue between patients and providers regarding conception may reduce transmission during conception. CORRESPONDING AUTHOR: Prof. D. Jones, University of Miami Miller School of Medicine, United States of America, [email protected] P559 MINDFULNESS INTERVENTION FOR PEOPLE WITH MULTIPLE SCLEROSIS: MIMS TRIAL A. Bogosian1, Rona Moss-Morris2, Paul Chadwick2 1 King's College London, LONDON, United Kingdom 2 Institute of Psychiatry, King's College London, LONDON, United Kingdom Introduction: Multiple sclerosis (MS) is a chronic disease of the central nervous system. Mindfulness interventions have been shown to effectively reduce anxiety, depression and pain in patients with chronic physical illnesses. The aim of this study is to assess the effectiveness of an easy to access mindfulness based intervention, which aims to reduce distress for people affected by primary and secondary progressive MS. Methods: Forty participants were randomly assigned to the 8week intervention (n=19) or a waitlist control group (n=21). One-hour sessions were delivered to groups of 3-5 people using SKYPE videoconferencing. Participants completed standardized questionnaires to measure mood, impact of MS and symptom severity at baseline, post-therapy and 3 months follow-up. At the end of the intervention, participants were interviewed about their experiences of the mindfulness course. Results: All but one participant in the intervention group completed the mindfulness course. Health related distress scores (primary outcome) were lower in the intervention group compared to control group at posttherapy and follow-up (p<0.05), effect size -.64 post-therapy and -.94 at follow-up. Mean scores for pain, fatigue, anxiety, depression, impact of MS (secondary outcomes) were reduced for the mindfulness group compared to control group at post-therapy and follow-up; effect sizes, -.27-.99 post-therapy and -.29 - -1.12 at follow-up. Interviews revealed that participants valued the group discussions during the sessions and distance setting of the intervention. Conclusions: Accessibility and feasibility of distance-delivered mindfulness interventions and the challenges of adapting a mindfulness course for people with progressive MS will be discuss.

S167

pain in people with OA, because study participants are likely to fill in a questionnaire or receive an interviewer when they have a pain-free episode. A well-known questionnaire designed to measure pain in OA (WOMAC) asks about pain in the past 48 hours. This rather short period may not sufficiently reflect the actual pain experience in people with OA. A longer period, i.c. of 2 weeks, may better capture their experience of pain. This study presents the first findings on the 14-day pain experience in a population-based study. Methods: The baseline of the six-country European Project on Osteoarthritis (EPOSA) consisted of a face-to-face interview and a clinical OA examination. The interviewer left a so-termed pain calendar, on which the participants were asked to indicate daily on a scale from 0 to 10 if they experienced pain, for 14 consecutive days. Summary measures were number of days with any pain, and intensity of pain averaged across days. Results: In 2417 participants (82.2%) both a clinical examination and valid pain calendar data for at least 7 days were available. Both pain measures distinguished equally well participants with and without OA. Both measures showed higher scores in females, participants with a low sense of mastery, and participants living in southern Europe. Conclusion: The pain calendar is a feasible approach to measuring pain in older people. CORRESPONDING AUTHOR: Prof. D.J.H. Deeg, VU University Medical Centre / LASA, Nederland, [email protected] P561 IMPACT OF CLINICAL OSTEOARTHRITIS ON SELF-RATED HEALTH IN EUROPE N.M. Van Schoor VU University Medical Centre / LASA, AMSTERDAM, Nederland Introduction: Only few studies examined the association between osteoarthritis (OA) and self-rated health (SRH) in the general population. To our knowledge, none of these studies used a clinical definition of OA. In addition, most studies were restricted to one country only. The objectives of the current study are: (1) to examine the relation between clinical OA at different sites and SRH in the general population; (2) to examine whether country is an effect modifier. Methods: Baseline data of the European Project on OSteoArthritis (EPOSA) were used. EPOSA is an observational study including preharmonized data from six European cohort studies (GER, IT, NL, ES, SWE, UK) on older community-dwelling persons aged 65-85 years (n=2709). Clinical OA was defined according to the ACR criteria. SRH was assessed using 1 question: How is your health in general? Response categories were dichotomized in the analyses (poor versus good SRH). Results: An interaction effect with country was observed. The OR (95% CI) for poor SRH ranged between 2.1 (1.1-4.2)(IT) and 13.9 (1.8104.3)(ES) for persons having hip OA versus no hip OA; from 1.8 (1.03.1)(SWE) to 5.5 (3.1-9.9)(UK) for knee OA versus no knee OA, from 0.8 (0.4-1.6)(GER) to 2.8 (1.5-5.1)(UK) for hand OA versus no hand OA. All associations were statistically significant, except for the associations between SRH and hip and hand OA, respectively, in Germany. Conclusion: In most countries, clinical OA at different sites was related to poor SRH.

CORRESPONDING AUTHOR: Dr A. Bogosian, King's College London, United Kingdom, [email protected]

CORRESPONDING AUTHOR: Dr N.M. Van Schoor, VU University Medical Centre / LASA, Nederland, [email protected]

P560 MEASURING PAIN USING A 14-DAY PAIN CALENDAR IN OLDER PEOPLE WITH OSTEOARTHRITIS D.J.H. Deeg VU University Medical Centre / LASA, AMSTERDAM, Nederland

P562 DIFFERENTIAL PREDICTORS OF ACUTE POST-SURGICAL PAIN INTENSITY AFTER ABDOMINAL HYSTERECTOMY AND MAJOR JOINT ARTHROPLASTY P.J.R. Pinto1, T. McIntyre2, V. Araújo-Soares3, P. Costa1, A. Almeida4 1 ICVS - Minho University, GUIMARÃES, Portugal 2 Texas Institute for Measurement, Evaluation and Statistics (TIMES) and Departmen, HOUSTON, United States of America

Introduction: Osteoarthritis (OA) is a condition that is known for its fluctuating course. This poses a problem for studies that aim to measure

S168 3 Institute of Health & Society, Faculty of Medical Sciences, Newcastle University, NEW CASTLE, United Kingdom 4 Life and Health Sciences Research Institute (ICVS), School of Health Sciences, U, BRAGA, Portugal

Introduction: Acute pain is the most common, anticipated and expected problem after surgery. Psychological factors play a significant role in post-surgical pain experience and their study can inform post-surgical pain management. The aim of this study is to investigate whether psychological predictors of acute post-surgical pain intensity could vary according to different types of surgery- abdominal hysterectomy (AH) and major joint arthroplasty (MJA) -, in order to identify risk factors that could inform the development of specific pre-surgical psychological pain interventions. Method: This was a prospective cohort study.142 women undergoing AH and 110 patients undergoing MJAwere assessed 24 hours before (T1) and 48 hours after (T2) surgery. Results: A predictive postsurgical pain model was found for AH which included younger age, presurgical pain and pain catastrophizing. Pre-surgical optimism was the best predictor of acute post-surgical pain intensity after MJA, irrespectively of demographic and clinical factors. Conclusions: This preliminary evidence indicates that pre-surgical pain catastrophizing and optimism seem to be key factors for patients undergoing AH and MJA, respectively. Present findings highlight the relevance of targeting specific psychological risk factors according to type of surgery in order to more effectively manage acute post-surgical pain. CORRESPONDING AUTHOR: P.J.R. Pinto, ICVS - Minho University, Portugal, [email protected] P563 A PILOT STUDY OF THE BACK SCHOOL BY MULTIDISCIPLINARY COOPERATION WITH COGNITIVE BEHAVIORAL THERAPY R.M. Motoya, Y.O. Ono, K.O. Otani, S.Y. Yabuki, S.N. Niwa, H.Y. Yabe, S.K. Konno Fukushima Medical University, FUKUSHIMA, Japan Introduction: Back school program for chronic low back pain, consisting mainly of exercise and patient education, is effective in improvement of pain symptoms and daily function (Shirado et al., 2005). Meanwhile, there are also studies have reported some negative effects (Brox et al., 2008) and how content is not unified. Therefore, the purpose of this study is to examine the effect of the back school by multidisciplinary with cognitive-behavioral therapy as a pilot study. Methods: Participants: seven chronic low back pain patients with poor improvement in the usual treatment. Practitioners: Orthopedic surgeon, physical therapist and clinical psychologist. Procedure: Total of five times with once every other week. Group format of 90 minutes per session. Contents: patient education, self-monitoring, back exercise, relaxation, stress management, cognitive restructuring, activity pacing, exposure, Measurements: Interview sheet (duration, frequency, severity), physical function (FFD: finger floor distance, grip strength), PCS (pain catastrophizing), PASS-20 (fear of pain), PDAS (pain disability). Results: Of the seven, five people showed an improvement with respect to pain severity, grip strength and sense of control of low back pain. In addition, six has improved FFD, PCS, PASS-20 and PDAS. As a result of calculating the effect size, sense of control (d=0.57) is 'moderate', PCS (d=1.19) and PASS-20 (d=1.25) were 'large'. Conclusions: Our findings suggested that this back school was effective for pain symptoms, physical function, and psychological variables, which much related to pain management and daily disabilities. Therefore, this school may be useful for pain patients with poor effect in the usual care. CORRESPONDING AUTHOR: R.M. Motoya, Fukushima Medical University, Japan, [email protected]

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

P564 ATTENTIONAL BIAS AMONG CAREGIVERS: ITS ASSOCIATION WITH ACCURATE DETECTION OF PATIENTS' PAIN BEHAVIORS S. Mohammadi1, Mohsen Dehghani2, Robbert Sanderman1, Mariet Hagedoorn1 1 University Medical Center Groningen, GRONINGEN, The Netherlands 2 Family Research Institute, Shahid Beheshti University, TEHRAN, Iran Background: Recent literature suggests the existence of attentional bias to pain among family caregivers of pain patients. Caregivers high in attentional bias to pain might misinterpret neutral cues as pain cues, which lead to inaccurate evaluations of patients’ pain behaviors. This study aims to provide further evidence for the presence of attentional bias to pain among caregivers, furthermore to explore the role of caregivers’ attentional bias to pain in the disagreement between caregivers’ and patients’ pain behaviors report. Methods: The sample consisted of 96 caregivers and their patients who were referred to the Orthopedic clinic center of Atieh Hospital (Tehran, Iran). Caregivers completed a dot-probe task assessing attention to painful and happy stimuli. Both patients and caregivers completed a checklist assessing patients’ pain behavior. Findings: Caregivers showed significantly more attention to pain information comparing to happy information (p=.038). The association between attending to pain and detecting pain behaviors (r=.28, p=0.006) was significant. Secondly, hierarchical linear modelling did not show an association between caregivers’ attentional bias and disagreement between caregivers’ and patients’ pain behavior report. Discussion: Caregivers showed attentional bias to pain information, which was associated with detecting more pain behaviors. Our hypothesis that attentional bias might also increase the possibility of an imprecise and inaccurate detection of pain behaviors- i.e. based on disagreement between patients and caregivers- was however not confirmed. CORRESPONDING AUTHOR: S. Mohammadi, University Medical Center Groningen, The Netherlands, [email protected] P565 STRESSFUL EVENTS AND COPING IN ACUTE AND SUB-ACUTE WHIPLASH ASSOCIATED DISORDER S. Pettersson, A. Bring, P. Åsenlöf Uppsala Universitet, UPPSALA, Sweden Introduction. The purpose of the study was to describe daily stressors and related coping strategies experienced by individuals with whiplash associated disorder (WAD), 0-1 month (acute, A) and 3-4 months (sub-acute, SA) after the injury, using the WAD-Daily Coping Assessments (WAD-DCA). Method. Persons (n=53, age=19-64 years) with WAD (grade I-III) were asked to complete the WAD-DCA during one week in the A and the SA phases respectively. Thirty-one persons (median 39.5 years, 71% female) described at least one stressful event in both phases and their diaries were included in the analysis. The stressful events were categorized into ten different stressors. Individual mean scores /phase (0-1) were calculated, regarding stressors and coping strategies and compared using Paired Samples T-Test. Results. Descriptions from 311 WAD-DCAs were categorized; the most frequently reported stressors were “symptoms”, “emotions”, and “occupations/studies”. Only the stressor “Unspecific activity limitation”was less frequently reported in SA (p=0.008). A majority of responders (A=84%,SA=84%) used the strategy “Told myself that I could not let this problem stop me”. “Thought of something I like to do”(p=0.009) was less and “Relaxed”more often reported (p=0.042) in the SA. Conclusion. This unique study follow situational specific stressors and coping over time in a population with acute WAD. The WAD-DCA captures patient’s experiences of daily stressful situations and mirrors stressors of bio-psycho-social character. Cognitive and behavioral

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

strategies were used in both phases. The result renders favor for the use of WAD-DCA, to further understand coping behaviors and implement individually tailored behavioral medicine strategies. CORRESPONDING AUTHOR: PhD S. Pettersson, Uppsala Universitet, Sweden, [email protected] P566 USABILITY OF A COMPUTERIZED ADAPTIVE TEST TO MEASURE FATIGUE IN RHEUMATOID ARTHRITIS C. Bode1, S. Nikolaus1, E. Taal1, H. Vonkeman2, C. Glas1, M. Van de Laar1 1 University of Twente, ENSCHEDE, Nederland 2 Medisch Spectrum Twente, ENSCHEDE, Nederland Background: Many patients with rheumatoid arthritis (RA) experience fatigue. To measure fatigue precisely on individual level, a computerized adaptive test (CAT) has been developed. With the CAT, items are selected based on the previous answer and the number of items needed to achieve the same level of precision as obtained in a traditional questionnaire can be up to 40% less. Aim of the study was to examine the usability of the first version of the CAT. Methods: Participants were 15 patients with RA. In individual sessions, they filled in the CAT while thinking aloud and were interviewed about their experience with the new measurement instrument. Findings: Patients perceived the CAT as easy to use. They described it as clear, brief and easy to complete. They were positive about getting one question per screen, the changing response options, lay-out, progress bar and number of items. Patients perceived the CAT as useful and liked the idea of the adaptive test mechanism. It was also reported that it was pleasant to fill in the CAT. However, patients noted that not all items were applicable to everybody, e.g. questions about work. Some patients indicated that the wording of questions within the severity dimension were very similar. Discussion: Patients perceived the CAT as useful and easy to use. The second version of the CATFatigueRA has been improved according to the patients’ comments and is currently used in a validation study to examine its measurement characteristics. Finally the CATFatigueRA will be implemented in daily clinical practice. CORRESPONDING AUTHOR: Dr C. Bode, University of Twente, Nederland, [email protected] P567 FATIGUE IN RHEUMATOID ARTHRITIS PATIENTS WITH LOW DISEASE-ACTIVITY C. Bode1, L. Lenferink1, S. Nikolaus1, H. Vonkeman2 1 University of Twente, ENSCHEDE, Nederland 2 Medisch Spectrum Twente, ENSCHEDE, Nederland Background: Unless effective pharmacological treatment for controlling disease-activity in rheumatoid arthritis, the majority of patients report clinically relevant fatigue which challenges the (medical) view of diseaseactivity as main cause of elevated levels of fatigue. The primary aim of this study was to analyse the prevalence of fatigue in RA patients with very low disease-activity, by using a novel multidimensional fatigue questionnaire (BRAF-MDQ). As the Dutch version was not yet validated, the second aim was to psychometrically evaluate the Dutch BRAFMDQ.Methods: Data of 199 RA patients (69 % women, mean age 59 years, mean DAS-28 = 1.99) were selected from a multi-centre randomized clinical trial. Firstly, the prevalence rates of clinical important levels of fatigue were examined and Spearman Rho correlations were used to determine how fatigue was related to disease-activity. Secondly, the dimensional structure, internal consistency, discriminant validity, distinctiveness, reliability, construct validity and possible floor- and ceiling effects were examined to psychometrically evaluate the BRAF-

S169

MDQ.Findings: Clinically relevant fatiguewas present in 73 % of the patients. The correlation coefficients between disease-activity and dimensions of fatigue ranged from .05 to .09. The dimensional structure of the BRAF-MDQ was broadly confirmed, but major floor effects were detected.Discussion: The findings indicate that even though RA inflammation is clinically under control, the majority of the patients still reported clinically relevant fatigue. Not all items of BRAF-MDQ are equally suitable for patients with low disease-activity, pointing to possibilities for improving fatigue measurement by, for example, adaptive testing methodologies. CORRESPONDING AUTHOR: Dr C Bode, University of Twente, Nederland, [email protected] P568 LONG-TERM CHANGES OF PAIN-SITES WITH PREDICTIVE FACTORS FROM EACH BASELINE PAIN-SITE NUMBER: THE HUNT STUDY E.A. Fors1, I. Mundal2 1 Norwegian University of Science and Technology, RANHEIM, Norway 2 Department of Neuroscience, Faculty of Medicine, NTNU, TRONDHEIM, Norway Introduction/objectives: (1) Study the long-term changes of pain-site numbers in the population during 11 years. (2) Study the impact of mood, lifestyle and health-related factors on the long-term pain-site distribution originating from each pain site number (0-9) at baseline (HUNT2). Methods: Design: A prospective, population-based cohort study using the HUNT2 (baseline) and HUNT3 (follow-up) surveys. Participants: 22922 adults > 19 years, who answered both the HUNT2 and HUNT3 questionnaires. Analyses: Percentages. Multiple linear regression. Results: Pain-site number (range 0-9) distribution at 11 years follow-up (HUNT3) in % of total/HUNT2 participants (n= 22922). From 0 painsites in HUNT2: 0 (62%), 1 (12%), 3 (7%), 5 (2%), 9 (0.4%). 1 pain-site in HUNT2: 0 (45%), 1 (16%), 3 (10%), 5 (3%), 9 (0.6). 3 pain-sites in HUNT2: 0 (27%), 1 (11%), 3 (17%), 5 (8%), 9 (2%). 5 pain-sites in HUNT2: 0 (15%), 1 (6%), 3 (16%), 5 (14%), 9 (3%). 9 pain-sites in HUNT2: 0 (5%), 1 (1%), 3 (7%), 5 (13%), 9 (24%). Significant predictive factors (independent variables) associated with increasing pain-sites numbers at follow-up (HUNT3) (dependent variable),from each baseline pain-site number: From 0 pain-sites (HUNT2): Female gender, anxiety, sleeping problems, BMI, smoking, chronic disease. From 1: Female gender, anxiety, smoking, education, chronic disease. From 3: Female, sleeping problems, BMI, smoking, education, chronic disease. From 5: Age, female, anxiety. From 9: Increasing age. Conclusion: Pain site numbers appear stable 11 years ahead; vulnerability factors vary with the baseline numbers. CORRESPONDING AUTHOR: Assoc professor E.A. Fors, Norwegian University of Science and Technology, Norway, [email protected] P569 NEUROPEPTIDE Y LEV ELS ARE REDUCED AFTER COGNITIVE BEHAVIOURAL THERAPY (CBT) IN WOMEN WITH FIBROMYALGIA (FMS) B. Karlsson1, G. Burell2, P. Kristiansson2, K. Björkegren2, F. Nyberg2, K. Svärdsudd2 1 Uppsala University, UPPSALA, Sweden 2 Uppsala University,Anderberg UM Uppsala University, UPPSALA, Sweden Aim: To evaluate CBT based stress management for FMS patients on plasma levels of neuropeptide Y (NPY). Background: FMS has been increasingly attributed to stress. NPY, a neuropeptide pivotal for maintaining stress responses has been found to be elevated in FMS patients compared to controls.

S170

Study population and method: 48 women, age 27-62 years were randomly allocated to an early treatment group and a wait list group. A modified protocol for CBT originally designed for treating women after myocardial infarction was used with the key components education, self-monitoring, skills training, cognitive reconstructing, life value issues and brief relaxation. The patients were treated in groups of 5-7 participants during two hour sessions once a week during six months. Assessments by self-rating questionnaires showed increased life control and decreased stress and vital exhaustion. However the self-rated pain was not significantly reduced. Outcome evaluation was also made by measuring NPY. Results: NPY was increased to some degree in the treatment group (74.380.8) as compared to a much higher increase in the control group (74.3106.6) in the randomized clinical trial (N= 24+24). In the before and after design allowing all participant (N=48) to be evaluated, giving a higher power, a 76% decline (p=0.006) of the baseline level was seen and maintained during the follow-up period (p=0.011). Conclusions: CBT has the potential to improve well-being among pain patients. Reduced levels of NPY might be a marker of the biological effect of CBT for female patients with FMS. CORRESPONDING AUTHOR: MD B. Karlsson, Uppsala University, Sweden, [email protected] P570 CAN SOCIAL SUPPORT WORK VIRTUALLY? THE IMPACT OF AN ON-LINE TOOL ON RA PATIENTS Z. Kostova Institute of Health Communication, University of Lugano, LUGANO, Switzerland Introduction: Although much has been evidenced of the social support effects on health, there is limited information how far social support can be provided online and under which conditions such support can deliver positive outcomes for chronic pain patients. The aim of this study is to examine the impact of an online social support tool on patients’ health outcomes and to identify the determinants that impact the usage and the efficiency of such tool. Methods: we used a mixed approach method where on a first stage we conducted an experimental quantitative study with 160 RA patients interacting on a web site that we developed previously using an online questionnaire. On a second stage we conducted 20 in-depth interviews with RA patients evaluating the factors that determined the usage and the impact of the site. Results: Onlinesupport improved more concrete outcomes related on the daily management of the disease, but was less successful on deeper selfperceptions variables as empowerment and acceptance. These results could be explained by some determinants like the characteristics of the web source; the internal and external experience with the disease; illness dimensions and personality type. Conclusions: defining the factors that determine patients’ usage of online support lead to clear advice for those designing future online tools in order to maximize the effectiveness in delivering online social support. CORRESPONDING AUTHOR: Z. Kostova, Institute of Health Communication, University of Lugano, Switzerland, [email protected] P571 PHYSICAL ACTIVITY AND THE DETERMINANTS OF THE ELDERLY IN TAIWAN S. Huang1, W. Hung2, M. Shyu3, J. P. Wei4, T. Chou1 1 National Taiwan Normal University, TAIPEI, Taiwan 2 Cardinal Tien's Junior College of Nursing and Management, TAIPEI, Taiwan 3 Taipei Medical University, TAIPEI, Taiwan 4 Taiwan Sports University, TAIPEI, Taiwan

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

Introduction: This study aims to investigate the situation and the determinants of physical activity in the un-institutionalized elderly in Taiwan. Methods: Both surveys and five focus group interviews were conducted. Elderly above the age of 65 in urban and rural areas were selected as the subjects. Four hundred and eighty-one questionnaires were collected with a response rate of 90.2%. The sample were composed mainly of elderly females (60.1%), living with their children (47.9%), with the elementary school educational level of the household (28.1%) and with an average age of 73.53. Results: The findings are: (1) better perceived health status, higher perceived health benefit and self-efficacy are associated with higher intention of regular physical activity; (2) more social support and social capital are associated with more physical activity. The findings from the focus group interviews were:(1).the respondents who were physically active went through three stages, such as “the difficult beginning”, “health follows physical activity”and “Being active is life”. (2).their preference tended to be easy movement, walking and team sports. For them, the barriers of exercise were “ difficultly in limb movement “ and “ difficulty to spare time from family life “. Conclusion: Based on this study, we suggest that the elderlys' physical activities and their intention should be enhanced with well-designed programs to increase their social support and social capital and decrease the barriers to be physically active. CORRESPONDING AUTHOR: Professor S. Huang, National Taiwan Normal University, [email protected] P572 PERSISTING SOCIAL PARTICIPATION RESTRICTIONS AMONG FORMER BURULI ULCER PATIENTS IN GHANA AND BENIN J. de Zeeuw1, TF Omansen1, M Douwstra1, YT Barogui2, C Agossadou2, GE Sopoh2, RO Phillips3, C Johnson2, KM Abass4, P Saunderson5, PU Dijkstra1, T.S van der Werf1, Y Stienstra1 1 University Medical Center Groningen, GRONINGEN, Nederland 2 Programme de Lutte Contre la Lèpre et l’Ulcère de Buruli, Ministries of Health, COTONOU, Benin (Dahomey) 3 KNUST, KUMASI, Ghana 4 Presbyterian Hospital, AGOGO, Ghana 5 American Leprosy Missions, GREENVILLE, United States of America Introduction: Buruli ulcer may induce severe disabilities impacting on a person’s well-being and quality of life. Information about long-term disabilities and participation restrictions is scarce. The objective of this study was to gain insight into participation restrictions among former BU patients in Ghana and Benin. Methods: Former Buruli ulcer patients were interviewed using the Participation Scale, the Buruli Ulcer Functional Limitation Score to measure functional limitation, and the Exploratory Model Interview Catalogue to measure perceived stigma. Healthy community controls were also interviewed using the Participation Scale. Trained native interviewers conducted the interviews. Former Former Buruli ulcer patients were eligible for inclusion if they had been treated between 2005 and 2011, had ended treatment at least 3 months before the interview, and were at least 15 years of age. Results: In total 143 former Buruli ulcer patients and 106 community controls from Ghana and Benin were included in the study. Participation restrictions were experienced by 67 former patients (median score, 30, IQR; 23;43) while 76 participated in social life without problems (median score 5, IQR; 2;9). Most restrictions encountered related to employment. Linear regression showed being female, perceived stigma, functional limitation and larger lesions (category II) as predictors of more participation restrictions. Conclusion: Persisting participation restrictions are experienced by former BU patients in Ghana and Benin. The most important predictors of participation restrictions are: being female, perceived stigma, functional limitation and larger lesions.

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

CORRESPONDING AUTHOR: Msc J. de Zeeuw, University Medical Center Groningen, Nederland, [email protected] P573 SELF-EFFICACY CONTRIBUTES TO MAINTAINED PHYSICAL ACTIVITY AFTER A TRAINING INTERVENTION IN PERSONS WITH HIP OSTEOARTHRITIS T. Bieler1, N.M. Hammer1, J. Midtgaard2, N. Beyer1 1 Bispebjerg Hospital, University of Copenhagen, COPENHAGEN, Denmark 2 The University Hospitals Centre for Health Care Research, Rigshospitalet, COPENHAGEN, Denmark Background. A substantial number of participants in exercise studies are unable to maintain a physically active lifestyle. The reason for this is complex but self-efficacy (SE) appears to play an important role. Aim of study The aim of this study was to examine how SE contributes to the understanding of maintenance of physical activity during 8 months follow-up after participation in a 4 months supervised exercise intervention in persons with hip osteoarthritis. Method. A parallel mixed methods design was applied. Quantitative data (n=52) on physical activity and perceived SE were obtained through questionnaires. Qualitative data were obtained through semi-structured interviews (n=15) based on SE-theory and analyzed using directed content analysis. Results 60% (n=31) had increased their physical activity level from baseline to 12 months (p<0.001) and during the same period their SE increased (p<0.05), while there was a reduction among those who had reduced their physical activity (n=9,p<0.05). The qualitative results showed that the participants had increased their SE to exercise through personal experiences of having the skills required, inspiration by other participants to progress in training intensity, the instructors’ support and encouragement, and changes in the interpretation of exercise induced physiological response. Experienced symptoms, expectations of outcome and a sense of obligation towards the research project also influenced the choice of maintaining a physically active life style. Conclusion SE to exercise appears to greatly contribute to the understanding of maintenance of physical activity in people with hip osteoarthritis, but SE alone cannot explain the maintenance. CORRESPONDING AUTHOR: PhD student T. Bieler, Bispebjerg Hospital, University of Copenhagen, Denmark, [email protected] P574 PSYCHOLOGICAL SKILLS USAGE AND COMPETITIVE ANXIETY RESPONSE AS A FUNCTION OF SKILL LEVEL IN SQUASH A. Halawani UMM AL-QURA UNIVERSITY, MAKKAH, Saudi Arabia Introduction: This study examined the intensity and direction of competitive anxiety symptoms and psychological skill usage in Squash union players of different skill levels. Elite (n=10) and nonelite (n=10). participants gave measurements of competitive anxiety, self-confidence, and psychological skills. The elite group reported more facilitative interpretations of competitive anxiety symptoms, higher levels of selfconfidence, lower relaxation usage, and greater imagery and self-talk use than their nonelite counterparts. Methods: Participants’ scores on the measurements were examined for accuracy of data entry, missing values, and fit between their distribution and the assumptions of multivariate analysis. Results: The findings of this study suggest that elite and nonelite athletes differ in their use of psychological skills to cope with their experiences of symptoms associated with competitive anxiety. Specifically, nonelite

S171

performers primarily use relaxation strategies to reduce anxiety intensity while elite athletes appear to maintain intensity levels and adopt a combination of psychological skills to interpret symptoms as facilitative. Potential mechanisms for this process include the use of imagery and verbal persuasion efficacy-enhancement techniques. Conclusion: Nonelite performers who experience anxiety symptoms as debilitative should implement relaxation-based programs. However, this may be inappropriate for certain sports that require high levels of activation states. Performers may therefore need to reduce symptom intensity, restructure cognitions, and raise activation states once again to appropriate levels. Elite performers who are debilitated are advised to implement cognitive restructuring techniques to interpret their anxiety as facilitative to performance via a combination of goal setting, self-talk, and imagery. CORRESPONDING AUTHOR: DR A Halawani, UMM AL-QURA UNIVERSITY, Saudi Arabia, [email protected] P575 A CONTROLLED TRIAL OF BEHAVIORAL THERAPY IN REHABILITATION FOR IMMOBILITY IN PATIENTS OF POSTSTROKE DEPRESSION T. Hamaguchi1, H Takizawa1, K Sasao1, T Ishioka1, J. Tayama2, T Tomiie3, T Saigou2, TH Nakamura1, K Kayaba1 1 Saitama Prefectural University, School of Health and Social Services, Saitama, Japan 2 Center for Health and Community Medicine, Nagasaki University, NAGASAKI, Japan 3 School of Psychological Science, Health Sciences University of Hokkaido, SAPPORO, Japan Background: Post-stroke depression (PSD) is commonly observed among stroke survivors. PSD reduces participation in rehabilitation, but the effect on physical activity from acute rehabilitation is unknown. This study examined whether behavioral therapy improves physical activity in PSD. Methods: Twelve inpatients of PSD and 12 non-depressive stroke patients were assessed and assigned to either an intervention or non-intervention group depending on their motor grade. All patients were hospitalized after strokes and participated in rehabilitation programs. The Zung’s Selfrating Depression Scale (SDS), Behavioral Inhibition System/ Behavioral Activation System (BIS/BAS) and Functional Independence Measure were administered at the screening and initial phases of the study, and at the 4th and 8th week of the usual occupational therapy and physiotherapy. The activity log, breathing technique, and exercise were also provided to the intervention group. Patient's locomotor activity was monitored using pedometers during the study. This study was approved by the Ethics Committee of the Saitama Prefectural University. Results: By repeated measures analysis of variance, pedometer counts within 1st week were significantly lower in PSD than controls. At 8th week, locomotor activity was not significantly different between intervention group and non-depression group. In contrast, the PSD control had significantly lower locomotor activity than that of the control group after the 8th week. The SDS and BIS/BAS did not change throughout the study in either group. Conclusion: The study results demonstrate that behavioral therapy can improve physical activities of patients with PSD, and that the treatment effect varies depending on the degree of motor function impairment. CORRESPONDING AUTHOR: Prof. T. Hamaguchi, Saitama Prefectural University, School of Health and Social Services, Japan, [email protected] P576 IMPLEMENTATION INTENTIONS PREDICT CHANGES IN PHYSICAL ACTIVITY BUT NOT HABIT STRENGTH AMONG CHILDREN AT THREE-MONTHS E.L. Thomas University of Birmingham, BIRMINGHAM, United Kingdom

S172

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

Introduction. Numerous studies have shown positive effects of implementation intention based interventions for increasing physical activity, although few have tested their effects in children. This study aims to test the predictive power of implementation intentions for changes in physical activity, motivation and habit strength in a sample of children aged 9-11 years over a three month period. Methods. Participants (N=120) were randomly allocated to a control group or intervention group. Physical activity, intentions and habit strength were measured at baseline and follow up. Participants in the intervention group were asked to form up to three implementation intentions to be active over the next three months. Results. After controlling for age and school level clustering effects, past behaviour explained 41% of the variance in PA at three months. Intentions and implementation intentions added to this prediction by explaining a small but significant amount of incremental variance i.e. change in physical activity (2% and 4% respectively). Implementation intentions did not predict changes in habit strength or goal intentions. Conclusion. Implementation intentions can increase physical activity in children aged 9-11 years over and above goal intentions and prior performance. However, findings serve to undermine the idea that implementation intention effects can be explained by automated execution of behaviour and/or enhanced motivation.

elementary students from 10 schools in Makkah aged between 9 and 12 years old. Methods: Each child had his body composition indirectly measured by calculating his body mass index (BMI). The Bruininks-Oseretsky (BOT) was used to measure motor skill performance of all the participants. A correlation was used to determine if a relationship exists with the independent variable being BMI and the dependent variable being BOT scores. An ANOVA was utilized to identify specific significant relationships, with BMI and its three classifications as the independent variable and BOT scores as the dependent variable. In addition, an ANCOVA was performed to determine if the effect of BMI level on motor skills differs by age for fine and gross motor skills, by using BMI calculations as independent variables and BOT scores as the dependent variable. Mean and standard deviation were used to express the data. For all analyses significance was determined at the .05 level of probability. Results: The major findings in this study were 1) a relationship between BMI and fine motor skills is nonexistent, 2) a relationship between BMI and gross motor skill performance exists, 3) Conclusion: The effect of BMI level on motor skills does differ by age.

CORRESPONDING AUTHOR: Dr. E.L. Thomas, University of Birmingham, United Kingdom, [email protected]

P579 CHANGE OF QUALITY OF LIFE ON DIFFERENT EXERCISE BEHAVIOR IN HEPATITIS C PATIENTS DURING TREATMENT S.C. Shun National Taiwan University, TAIPEI, Taiwan

P577 PROMOTING PHYSICAL ACTIVITY OF THE ELDERLY THROUGH COOPERATIVE LEARNING STRATEGIES W. Hung1, S. Huang2 1 Cardinal Tien Junior College of Healthcare and Management, YILAN, Taiwan 2 National Taiwan Normal University, TAIPEI, Taiwan Introduction: The main purpose of this project was to construct a physical activity improvement program for the elderly based on social capital and social psychology theory and to evaluate the effectiveness of the program. Methods: Six neighborhoods in Taipei city were chosen and randomly allocated to the experimental group (N=119) and control group (N=81). The intervention tools included lectures on health issues and community-based cooperative learning strategies. The experimental group was given lectures of health issues, two fitness gymnastics videos and a cooperative learning protocol for four months with the theme of “Promoting physical activity of the elderly”; meanwhile, the control group was given generic health information. Results: The ANCOVA showed the experimental group and the control group were not significant different in the perceived benefits of exercise, general social support, social support, self-efficacy, behavioral intentions and physical fitness. Only perceived barriers of exercise and physical activity reached statistical significances. Conclusion: Strategies based on social capital and social psychology theory may be more effective for increasing physical activity levels and significantly reducing perceived barriers of exercise in the elderly participants in the experimental group. The experiences generated from this study would be provided to relevant institutions for their references to develop health promotion programs for the elderly. CORRESPONDING AUTHOR: Ph.D. W. Hung, Cardinal Tien Junior College of Healthcare and Management, Taiwan, [email protected]

CORRESPONDING AUTHOR: DR S. Zamzami, UMM AL-QURA UNIVERSITY, Saudi Arabia, [email protected]

Background: Hepatitis C virus (HCV) infection is one of major health problems in the world and it is recommended to receive combined therapy including interferon and Ribavirin for the patients to prevent from liver cancer. However, there is no study to compare the different change pattern of the physical and psychological distress during combined therapy based on different level of exercise behaviors before treatment. Therefore, the aim of this study was to identify any different pattern in symptom distress, fatigue, anxiety and quality of life during six months of combined therapy. Methods: A longitudinal design was used to recruit the participants and data were collected before treatment and at 8, 16, and 24 weeks during treatment. Eligible participants were patients with HCV infection who were outpatients scheduled to receive interferon combined ribavirin therapy. Disease information was collected from charts and a set of structured questionnaires were used to assess patients’ fatigue, symptom distress, anxiety, and quality of life. The descriptive and generalized estimating equations were used to examine the differences between the two groups. Results: The participants with regular exercise at baseline (n = 60) and irregular exercise (n = 54) all reported fatigue was the most distressed symptom in the past week of treatment. Compared to the patients with regular exercise group, those with irregular group had higher levels of symptom distress, fatigue, anxiety and lower levels of quality of life. Conclusions: It is recommended to perform regular exercise before and during treatment. CORRESPONDING AUTHOR: Associate Prof S.C. Shun, National Taiwan University, Taiwan, [email protected]

P578 RELATIONSHIP OF SCHOOL OBESITY AND MOTOR SKILLS OF ELEMENTARY AGE STUDENTS S. Zamzami UMM AL-QURA UNIVERSITY, MAKKAH, Saudi Arabia

P580 RELATIONSHIP OF PHYSICAL ACTIVITY AND DIETARY HABITS WITH BMI IN TRANSITION FROM CHILDHOOD TO ADOLESCENCE A. Khawandanh UMM AL-QURA UNIVERSITY, MAKKAH, Saudi Arabia

Aim: This investigation was to measure the connection among childhood obesity and gross and fine motor skill performance. Participants were 400

Introduction: The aim of the study was to explore (i) the relationship between several physical activity and dietary behaviours and BMI

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

Z-score and (ii) the relationship of changes in these PA and dietary behaviours with changes in BMI Z-score over a year period from childhood to adolescence. Method: Longitudinal study in which children were included in the fifth grade and measured for a consecutive year. Dietary and PA behaviours as well as height and weight were measured by means of self-reported validated questionnaires. Two sets of models were conducted. All explanatory variables were mean centred before they were entered in the models. Results: Significant inverse relationships with BMI Z-score were observed for frequency of breakfast consumption ([beta] = -0·033, se = 0·012) and frequency of sports participation ([beta] = -0·011, se = 0·004) across four time points. Significant inverse relationships between changes in BMI Z-score and changes in frequency of sports participation ([beta] = -0·011, se = 0·006) and hours of physical education (PE; [beta] = -0·052, se = 0·023) were observed, meaning that decreases in sports participation and hours spent in PE were associated with increases in BMI Z-score. It provides an insight into different dietary and PA behaviours related to (changes in) BMI Z-score during the transition from childhood to adolescence. Conclusion: PA occurs over four dimensions: frequency, intensity, time and type. Two were measured. The results provide evidence that frequency, but not duration, of sports participation plays a substantial role in weight management from childhood to adolescence. CORRESPONDING AUTHOR: A Khawandanh UMM AL-QURA UNIVERSITY, Saudi Arabia, [email protected] P581 ACTIVE BEHAVIORS, BMI AND PARENTAL PERCEPTIONS OF NEIGHBORHOOD ENVIRONMENTS IN CHILDREN A.M. Machado-Rodrigues1, M.M. Ferrão2, A. Gama2, I. Mourão3, H. Nogueira2, V. Rosado2, C. Padez2 1 University of Coimbra, VISEU, Portugal 2 Research Centre for Anthropology and Health, University of Coimbra, COIMBRA, Portugal 3 Department of Sport and Physical Education, UTAD, VILA REAL, Portugal

S173

P582 AUTISM SPECTRUM DISORDER FROM BIRTH TO DIAGNOSIS THROUGH PARENTS' EYES M. Bolduc, N.A. Abouzeid, N.P. Poirier University of Quebec in Montreal, MONTREAL, Canada Introduction. Autism Spectrum Disorder (ASD) affects nearly 1% of the population which greatly impacts the lives of children with the disorder, their family as well as the community. This study aims to provide, through parents experience, the significant events, from birth leading up to the diagnosis of ASD (early signs, age of first appearance, person who noticed these symptoms, proccess leading to the diagnosis). Method. 41 parents completed a self-administered questionnaire. Quantitative analysis (frequencies an means) were executed. Results. Parents are usually the first ones to notice the early signs of ASD. Their first concerns occur before the age of two. The language delay is the most obvious observation. Results show a long waiting period, in average over four years, between the first signs to the diagnosis. Generally, the diagnosis is established through a multi-disciplinary team. Conclusion. In conclusion, parents are the most involved and concerned in regards to their child development. It is crucial to adress their concerns so that children can be diagnosed earlier and benefit from early intervention. For that to happen, it is important to promote best practices in terms of detection, diagnosis and intervention. CORRESPONDING AUTHOR: M. Bolduc, University of Quebec in Montreal, Canada, [email protected] P583 PROBLEM ALCOHOL USE AND INTENTION TO QUIT AMONG PRIMARY CARE PATIENTS IN ODISHA, INDIA S. Pati1, S. Swain1, M.A. Hussain2 1 PHFI, BHUBANESWAR, India 2 School of Population Health, University of Queensland, BRISBANE, Australia

Background: Habitual PA may be influenced by a broader range of neighborhood, school, community, and family factors. Young people, particularly girls, tend to show lower habitual PA than boys and should be a target for prevention strategies aimed at healthy lifestyles. The purpose of this study was twofold: 1) to determine which neighborhood perceived attributes are related to active behaviors; 2) to analyze which neighborhood perceived attributes are related to BMI of children. Methods: The sample comprised 1825 girls aged 7-9 years. Height and weight were measured, and Body Mass Index (BMI) was calculated subsequently. Participants were classified as normal weight or overweight/obese (Cole et al., 2000). Environmental variables and physical activities were assessed by questionnaire filled by their parents - the Environmental Module of the International Physical Activity Prevalence Study (IPS, 2002) was used. Multiple linear regressions, with adjustments for parental education, were used. Results: Data revealed that neighborhoods with recreational facilities and infrastructure for walking and cycling were predictors of active behaviors; neighborhood safety was an additional related factor of habitual PA. On the other hand, neighborhoods with better access to destinations and with interesting things to look at while walking are significantly associated with lower BMI in Portuguese children. Conclusion: The study found some significant relationships between parental perceptions of the environment and BMI and habitual PA of Portuguese girls, especially related to security for walking and unpleasant neighborhood, and access to destinations.

Introduction: Primary care is ideally suited for early detection and secondary prevention of alcohol-related problems due to its high contactexposure with the population. Limited information is available on how problem alcohol use is detected in primary care setting and its prevention in India. We conducted this maiden study to explore the prevalence and pattern of problem alcohol use among patients and their quitting intentions in a rural primary care practice of Odisha, India. Methods: This cross sectional study was conducted during May-July 2013, wherein 431 adult out-patients attending a community health centre in Buguda (Odisha) were interviewed using a structured questionnaire. Information on socio-demographic characteristics, alcohol habits, and self rated physical and mental health were elicited. Data was analysed for alcohol habit, quitting intentions and their association with socio demographic characteristics using SPSS v20.0. Results: More than one third (38.0%;95% CI: 33.4-42.6) of patients were alcohol consumers and 59.1% (95% CI: 51.6-66.6) of them were hazardous drinkers. One quarter respondents were smokers and 36.1% (95% CI 31.6 - 40.7) used chewing tobacco. Duration of working hours, perceived job stress, income, and tobacco use were significantly associated with alcohol abuse. Hazardous drinkers had a lower self-rated general and mental health than teetotallers. From 165 alcoholics, 112 had thought of quitting and 88 had attempted to quit. Conclusion: The high prevalence of problem alcohol use in primary care practice indicates the need of health promotion in these settings in India. A community- primary care collaborative intervention could be helpful.

CORRESPONDING AUTHOR: Dr A.M. Machado-Rodrigues, University of Coimbra, Portugal, [email protected]

CO RRES PONDIN G AU THOR: Dr S. P ati, PHF I, India , [email protected]

S174

P584 IMPROVING MALE PARTICIPATION IN COLORECTAL CANCER SCREENING, A RANDOMISED CONTROLLED COMPARISON OF TARGETED INVITATIONAL APPROACHES A. Duncan1, I. Zajac2, I. Flight2, C. Wilson3, D. Turnbull1 1 The University of Adelaide, ADELAIDE, Australia 2 Preventative Health Flagship, CSIRO, ADELAIDE, Australia 3 Flinders Centre for Innovation in Cancer, Flinders University, BEDFORD PARK, Australia Background: Colorectal Cancer (CRC) is a major public health problem in Australia. Men participate in faecal occult blood test (FOBT) screening at a consistently lower rate than women despite their increased risk for CRC. This study compared the efficacy of targeted, mailed invitation strategies for increasing FOBT use in men. Method: Participants (N=6,816) were randomised to 1 of 4 trial arms. The control arm replicated the strategy used in Australia’s National Bowel Cancer Screening Program (advance notification letter, invitation letter and free FOBT). For the intervention, modifications were made to the content of the advance notification, invitation or both. Modifications targeted psychosocial variables associated with the precontemplation and contemplation stages of readiness to screen. Primary outcome was FOBT participation after 12 weeks. An additional 570 participants (split across intervention groups) completed surveys pre and post screening in order to measure change on the psychosocial variables targeted in the intervention. Results: Participation rates were 31.8%, 35%, 30.4% and 34.6% in the control, modified advance, modified invitation and modified both groups respectively (excluding survey participants). Compared to the control group, those who received the modified advance notification letter were significantly more likely to participate in the screening (OR= 1.15, p=0.05, 95% CI=1.00-1.33).Analysis of survey respondents’ scores on psychosocial variables over time, and between conditions, found no significant changes attributable to the intervention. Conclusions: A modified advance notification letter targeting variables associated with precontemplation improves screening compared to the current approach. Modifications beyond initial contact did not significantly improve participation compared to control. CORRESPONDING AUTHOR: A. Duncan, The University of Adelaide, Australia, [email protected] P585 WHICH INDIVIDUALS FROM VULNERABLE GROUPS REFRAIN FROM PARTICIPATION IN A CARDIOMETABOLIC HEALTH CHECK, AND WHY? I. Groenenberg1, M.R. Crone1, S. van Dijk1, J. Ben Meftah1, B.J.C. Middelkoop1, A.M. Stiggelbout1, W.J.J. Assendelft2 1 LUMC, LEIDEN, Nederland 2 Radboud Universiteit, NIJMEGEN, Nederland Introduction: Cardiometabolic health checks are increasingly implemented in primary care. Vulnerable groups -lower SES, ethnic minority groupsare underrepresented in these checks despite being at increased risk. Research questions: Which patient characteristics and which psychosocial determinants influence participation of vulnerable groups in a cardiometabolic health check? Methods: Study population (n=1690): Turkish, Moroccan, Surinamese, and lower SES native Dutch patients (45-70yrs) invited to participate in a health check. Patients were asked to answer questions on psychosocial determinants of participation. Univariate and multivariate logistic regression analyses were used to asses the influence of patient characteristics and psychosocial determinants on participation. Results: Response rate to the questionnaire was 66% (n=1120), of whom 67% (n=750) participated in the health check. Women were more inclined to participate than men (OR: 1.40 CI: 1.15-1.71). Native Dutch participated more often (54%) than Turkish (44%), Moroccans (40%),

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

Hindustani Surinamese (43%), and Creole Surinamese (30%). This response rate of native Dutch was significantly higher than all other ethnicities, but mostly so compared to the Creole Surinamese (OR: 2.63, CI: 1.67-4.16). Determinants associated with non-participation: being too old, experiencing symptoms, not feeling in control over one’s health, others finding it unimportant, not wanting to know one’s risk, and needing help. Multivariate analyses showed the latter two to be the most important predictors of non-participation (OR: 1.89 CI: 1.06-3.39 and OR: 1.52 CI: 1.02-2.25 respectively). Conclusion: Future interventions should particularly target male and (Creole) Surinamese patients, while simultaneously providing extra (practical) help and putting additional effort in increasing adequate risk perception. CORRESPONDING AUTHOR: I. Groenenberg, LUMC, Nederland, [email protected] P586 OUTCOMES OF A CRC SCREENING AMONG MEDICALLY UNDERSERVED ADULTS: A CDC-PADOH FUNDED PROJECT IN PA O.G. Dominic1, E.J.L. Eugene J. Lengerich2, T.M. Thomas McGarrity3, J.H. Jimithy Hawkins4 1 Highmark, Inc.; Penn State, CAMP HILL, United States of America 2 The Pennsylvania State University, College of Medicine, Department of PHS, HERSHEY, United States of America 3 Penn State Milton S. Hershey Medical Center,College of Medicine,Gastroenterolog, HERSHEY, United States of America 4 Pennsylvania Department of Health,Department of Health Bureau of Health Promoti, HARRISBURG, United States of America The purpose of this work was to develop, implement and evaluate a targeted CRC screening intervention to increase CRC screening uptake among nonadherent medically underserved adults residing in central PA. We measured actual CRC uptake as a method to determinate completion rates. The project was a CBPR design to determine completion of a provider-recommended, take-home FIT kit with education and social support components among a sample of average- or increased-risk adults age 50 and older not currently adherent to national CRC screening guidelines (N=250). This screening opportunity was offered at our partnered community-based sites utilizing 5 medical facilities (one academic institution, one medical center, two FQHC, and a community clinic). Methodology: Each consented participant attended a one-time, CRC screening offered at one of the participating project sites located in Central PA. Results: 164 (66%) returned the FIT kit within a one-week timeframe. 10 tests were abnormal (positive FIT results). Of these six colonoscopies were referred to medical center with four complete, 1 was a “no show’ and 1 was cancelled due to hypertension (BP reading taken prior to performing the colonoscopy- patient was sent home). The remaining four scheduled with their own provider; the other 3 were seen at our partnered FQHCs. Conclusion: this project demonstrated increased completion rate from baseline 24% to 66% at post project intervention among previously nonadherent medically underserved adults in Central PA. CBPR approach utilizing these 5 medical facilities seems to be effective strategy for improving CRC screening uptake among these medically underserved adults. CORRESPONDING AUTHOR: Conslt;Adj Prof O.G. Dominic, H ig h m ar k , I nc . ; P en n S t at e , U n it e d S t at e s o f Am er ic a , [email protected] P587 BARRIERS TO HELP-SEEKING AMONGST OBESE WOMEN WITH SYMPTOMS OF GYNAECOLOGICAL CANCERS: A HEALTHCARE PROFESSIONALS' PERSPECTIVE S. Tookey, M. Larkin, R.A. Howard, E.A. Grunfeld University of Birmingham, BIRMINGHAM, United Kingdom

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

Introduction: Obese women are reported at higher risk of developing some gynaecological cancers than non-obese women. Weight-related barriers contribute to delay in healthcare utilisation, however, specific factors that may contribute to late presentation have not been examined. The aim of this presentation is to compare the perspectives of healthcare professionals (HPs) with patients’ experiences to explore barriers and gaps in HP knowledge regarding the help-seeking behaviour of obese women presenting with potential symptoms of gynaecological cancer. Method: Interviews were undertaken with twenty UK-based specialist gynaecological and oncology service HPs and seven obese female patients. Semi-structured interviews explored perceived weight-related and cultural barriers of obese women presenting with abnormal gynaecological symptoms, and identified potential interventions to improve help-seeking behaviour amongst this population. Interviews were transcribed verbatim. The HP study was analysed using the ‘framework’ method and Interpretive Phenomenological Analysis for the patient study. Key themes were compared across the two studies. Results: Five main themes emerged as ‘HP perceived’ and ‘patient experienced’ barriers amongst this population: Low Body-Esteem, Low Awareness of Gynaecological Cancer Risks/Symptoms, Sociocultural & Economic Barriers (i.e. language, religious & cultural values), Obesity as Physical Barrier (i.e. mobility, symptom recognition/interpretation difficulties), and HP Judgmental Attitudes. Suggested interventions included community education (e.g. distribution of risk/symptom information through media), and effective doctor-patient communication addressing increased risks associated with obesity. In summary, results highlighted individual and cultural influences to help-seeking for gynaecological cancers. This study enhances understanding for potential delay in this population and the importance of HP knowledge in this area. CORRESPONDING AUTHOR: Ms. S. Tookey, University of Birmingham, United Kingdom, [email protected] P588 CAN DEPRESSION SCREENING IN CARDIAC PATIENTS ENHAN CE TREATMENT? RATIONAL AND DESIGN OF DEPSCREENINFO RCT S. Kohlmann, B.G. Gierk, B.L. Loewe University Medical Center Hamburg Eppendorf, HAMBURG, Germany Introduction. The American Heart Association recommends a standardized depression screening for cardiac patients. To date there is no randomized controlled trial (RCT) that supports this recommendation. Depscreen-INFO is the first trial that investigates the impact of depression screening on the level of depression, treatment, quality of life and health economic consequences. Methods. Outpatients with coronary heart disease are screened with the Patient Health Questionnaire-9 (PHQ-9). In terms of a suspected depression (PHQ-9>9 points) feedback and treatment recommendations are given to the cardiologist in charge. Half of the patients are randomized into the intervention group and receive feedback, patient tailored information on depression and contact addresses. Primary endpoint is level of depression after one and six months. As secondary endpoint number of treated patients, quality-adjusted years of life, direct and indirect health costs are assessed. Results. The rational, implementation, and a detailed sample description will be presented. To date, 2279 cardiac patients were screened. In total, 17.8% (n=403) patients had a suspected depression. Of these, 40.2% (n=162) were randomized into the intervention group and received a direct patient tailored feedback. Primary endpoint and secondary endpoints are currently assessed and will be evaluated. Conclusion. Design and procedure as well as research implications and limitations of Depscreen-INFO will be discussed. To our knowledge, Depscreen-INFO is the first RCT that will answer the pending question whether depression screening in cardiac settings is clinically efficient and cost-effective. So far, the successful recruitment already shows that depression screening is feasible in cardiac outpatient clinics.

S175

CORRESPONDING AUTHOR: Dr. S. Kohlmann, University Medical Center Hamburg, Eppendorf, [email protected] P589 IMPROVING PATIENT SAFETY VIA A PERSUASIVE REGISTRATION AND MONITORING SYSTEM FOR INFECTIONS IN NURSING HOMES N. de Jong1, A. Eikelenboom-Boskamp2, A. Voss2, J.E.W.C. Van Gemert-Pijnen1 1 University of Twente, ENSCHEDE, Nederland 2 Canisius Wilhelmina Hospital & Radboud University Medical Centre, NIJMEGEN, Nederland BACKGROUND Patient safety is increasingly threatened by healthcareassociated infections. Insight into the occurrence of (risk factors for) these infections is necessary for early detection and treatment. Therefore, the regional network iPrevent started to register which (risk factors for) infections occur and what antibiotics are used, via an online questionnaire. However, the questionnaire seemed prone to interpretation errors and not very user friendly. Therefore, via user-centered design a persuasive registration system was developed. METHODS For successful monitoring, it is important that the registration system fits existing work processes and needs of end-users (elderly care physicians). Therefore, we closely cooperated with end-users during development, combining multiple methods (literature search, expert discussion, questionnaire, interviews and scenario-based user-tests). Furthermore, we included elements of Persuasive Systems Design (PSD) in the system. RESULTS User-centered design allowed us to find opportunities for improvement. PSD enabled us to deal with them. Questions were simplified; Routing was added to omit redundant questions; Shown information was tailored to the individual user; Reminders/alerts (to add missing information) were added. Thus, the new registration system uses elements of Primary Task Support, Dialogue Support and System Credibility Support. The system was used during two prevalence measurements, registering a total of about 3.000 nursing home residents, living in 37 nursing homes. After some teething troubles, overall user experiences are positive. Currently, we are planning to do a summative evaluation to quantify these initial responses. Therefore, together with University of Oulu, we also started a questionnaire validation study, to validate the Perceived Persuasiveness Questionnaire. CORRESPONDING AUTHOR: MSc N. de Jong, University of Twente, Nederland, [email protected] P590 PREVALENCE RATE AND POSITIVE PREDICTIVE VALUE ON SEMI-HEALTH SYMPTOMS AMONG YOUNG ADULTS H.Y. Yamazaki1, S.S. Sakabe2, J.G.Z. Zhang3, X.Q. Xiao4 1 Tokoha University, HAMAMATSU, Japan 2 Soka University, TOKYO, Japan 3 Nanjing Normal University, NANJING, China 4 Ken-yu, TOKYO, Japan Introduction: The purpose of the present study was to identify a cutoff point on the semi-health index, which evaluated a level of condition in the susceptibility phase based on the natural history of disease, and to clarify prevalence rate and positive predictive value of semi-health symptoms among young adults in Japan. Chronic diseases, such as metabolic syndrome and circulatory diseases, are called life-style related diseases in Japan. They are not only serious causes of death but also risk factors of broken health. They have been steadily increasing. Method: The self-report questionnaire, which consisted of 53 items, was administered to 2,074 young adults in Japan between 2011 and 2012.

S176

Results: Then a principal components analysis was applied to the valid data in order to extract indices which evaluated structural characteristics on semi-health conditions. By this analysis, four principal components were extracted. Especially, the first principal component was extracted as an index which indicated a quantitative aspect of the semi-health condition. Every eigenvector of its component had a mark of plus. Therefore, this component was used as the semi-health index. Furthermore, a distribution of young adults with semi-health symptoms was determined by using this index. Then a cutoff point was calculated from the value of the index in order to determine predictive values of discriminating the semihealth from good health conditions. Conclusions: Finally, the prevalence rate (13.1%) and positive predictive value (46.0%) on the semi-health condition were derived form the distribution based on the semi-health criterion. CORRESPONDING AUTHOR: Prof. HY Yamazaki, Tokoha University, Japan, [email protected] P591 EXPLORING ATTITUDES TO HPV SELF-SAMPLING: THE ROLE OF SELF-EFFICACY D. Williams1, A. Fiander1, M. Davies2, D. Farewell1, K. Brain1 1 Cardiff Univerisity, CARDIFF, United Kingdom 2 Bangor University, BANGOR, United Kingdom Background: Cervical screening coverage is below the target of 80% across the UK. Testing for Human Papillomavirus (HPV) is being incorporated into the cervical screening programme, with the possible future introduction of HPV self-sampling. Women’s attitudes and intentions towards HPV self-sampling were examined using an extended Health Belief Model (HBM). It was expected that women with higher selfefficacy would have a higher intention to self-sample. Methods: Women were recruited opportunistically through Cervical Screening Wales, sexual health clinics and community groups. A crosssectional survey operationalising HBM constructs (perceived benefits, barriers, severity, susceptibility and self-efficacy) in relation to HPV selfsampling was used to identify determinants of HPV self-sampling intentions. Results: A total 194 out of 291 (67%) women aged 20-64yrs were recruited, of whom 69% (n=133) reported that they would be likely to self-sample. Women with higher self-sampling intentions reported significantly higher self-efficacy (p<.001, OR= 24.96, CI 6.35-92.20), rated HPV as more important in causing cervical cancer (p<.034, OR=2.33, CI 1.07-5.07), perceived fewer barriers to self-sampling (p<.001, OR=.663, CI 0.53-0.82), and more benefits to self-sampling (p<.012, OR=1.36, CI 1.07-1.74) and smear testing (p<.016, O.R.=1.43, CI 1.07-1.91) compared to women with lower self-sampling intentions. Conclusions: Greater self-efficacy was the strongest determinant of intention to self-sample, followed by causal beliefs regarding HPV and cervical cancer, and the benefits of screening outweighing the barriers. These are key concepts for inclusion in future interventions to encourage HPV self-sampling. Qualitative work is being undertaken to explore the interaction between identified factors and intentions to self-sample.

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

occurs. This negatively affects illness-related aspects (functional status, dyspnea, exacerbation risk, quality of life, treatmentadherence or survival). Therefore, routine distress-screening is advocated. We investigated the uptake of screening, distresslevels of COPD-patients, level of health status, and need for psychosocial care. Methods: Out-clinic patients (UMCGroningen, the Netherlands) received a screening questionnaire two weeks prior to hospital routine-checkup. Distress was assessed with the HSCL-25, Health status with the Clinical COPD Questionnaire (CCQ). Need for psychosocial care with one question (options: Yes’, ‘Maybe’, ‘No, I don’t need it’ or ‘No, I already receive treatment from a healthcare provider). Results. Of 420 patients, n=327 (78%) completed the HSCL-25 and CCQ; 57% (n=186) scored above the HSCL-25-cut-off (high distress), 28% (n=92) scored above the CCQ-cut-off (poor health status). Concerning ‘need for psychosocial care’ (n=323), a majority (232, 72%) reported no need for psychosocial care, 50 (15%) already received help and 41(13%) indicated an unmet need. The HSCL-25 and the CCQ were highly correlated (r=.58, p<.01); patients with high distress-levels showed more need for psychosocial care, independent of their health status. Conclusions. Although the uptake of our procedure was satisfying , and a vast majority of patients reported high distress-levels, most patients reported no need for help, despite being seriously distressed. Based on these findings the question remains: is it cost-effective to implement routine distress-screening in patients with COPD, as only few patients report an unmet need for care. CORRESPONDING AUTHOR: Dr G. Pool, UMCGroningen, Nederland, [email protected] P593 NEURAL PATHWAYS IN EMOTION REGULATION: AN EXPLORATORY DTI- STUDY M. Vandekerckhove1, Anja Waegeman2, Wim Van Hecke2, Caroline Vogels3, Frank De Belder2, Paul Parizel2, Frank Van Overwalle1 1 Vrije Universiteit Brussel, DRONGEN, Belgium 2 University of Antwerp, ANTWERP, Belgium 3 Sint Camillus, SINT DENIJS WESTREM, Belgium

P592 THE EFFECTIVENESS OF SCREENING FOR DISTRESS AND HEALTH STATUS IN PATIENTS WITH COPD G. Pool, G.A. Maters, R. Sanderman, J.B. Wempe, J.B. Fleer UMCGroningen, BEDUM, Nederland

Actual research on emotion show an increasing interest in how emotion regulation play an important role in mental health and well-being. While previous research on emotion regulation has focused on gray matter correlates, this represents the first exploratory study on white matter integrity and brain networks of emotion regulation. Quantitative diffusion tensor magnetic resonance imaging (DTI-MRI) measures of fractional anisotropy (FA) and mean diffusivity (MD) evaluated individual disposition of emotion regulation in a group of women (age range: 18-55 years, M = 36, SD = 8.87). In order to assess emotion regulation, two dispositional emotion regulation modes were distinguished: (1) a ‘high emotional approach’ (HEA) (N = 11) and (2) a ‘low emotional approach’ (LEA) (N= 10) defined by a tendency to affectively acknowledge, understand and express emotional experience of a situation and how emotional experience unfolds (cf. Stanton, Kirk, Cameron et al., 2000). Those with an HEA exhibited more structural connectivity of the cingulum, a limbic pathway important in affective processing and emotion regulation. Those with LEA exhibited a higher FA in motor and cognitive frontal pathways as the corticospinal tracts, the inferior and superior longitudinal fasciculus (involved in motor and cognitive processing and control), together with less structural connectivity in the body and splenium of the corpus callosum which is important in interhemispheric communication, affective processing and integration of affect.

Background: Although research shows that patients with COPD show relatively high distress-levels, under-treatment of distress frequently

CORRESPONDING AUTHOR: Professor M. Vandekerckhove, Vrije Universiteit Brussel, Belgium, [email protected]

CORRESPONDING AUTHOR: Mrs D. Williams, Cardiff Univerisity, United Kingdom, [email protected]

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

P594 SENSE OF COHERENCE AMONG RELIGIOUS AND NONRELIGIOUS STUDENTS FROM GERMANY AND POLAND M.S. Schonder TU Braunschweig Educational Psychology, BRAUNSCHWEIG, Germany Introduction. The research framework of this study is based on the theory of sense of coherence (SOC), defined as as a global perceptual predisposition in responding to life stress (Antonovsky, 1979). General Resistance Resources (GRRs) are important for a SOC level. Religion is one of them. However, empirical evidence on the association are scarce (Gibson 2003, Delgado 2007). Additionally, little is known about how sociodemographic factors (nation) contribute to a SOC value (Bowman 1996). The current study attempted to fill the gap in the neglected line of research and investigated how cultural and socio-demographic factors are connected with SOC. Thus, the differences in a SOC level and varied aspects of religion (strength of religious faith, church attendance, private religious practice) among religious and non-religious students from Germany and Poland have been researched. Methods. Sample consisted of 2266 students (1623 Germany, 643 Poland), 1370 religious (Protestants 493, Catholics 735, Buddhists 142) and 896 non-religious respondents. Sense of coherence was assessed on Orentation to Life Questionnaire. The strength of religious faith was evaluated on Santa Clara Strength of Religious Faith Questionnaire. Organisational (church attendance) and non-organisational (private religious practice) religious activities were valued on Duke religion index. Results. More devout students enjoy better SOC than less devout students. However, those who go to church, visit other religious meetings, pray, meditate or study the Bible less frequently, show higher SOC than those who do it more often. Significant dissimilarities in SOC across the different religious affiliations, but not across countries have been revealed. CORRESPONDING AUTHOR: MA M.S. Schonder, TU Braunschweig Educational Psychology, G P595 SEDENTARY BEHAVIOR, PHYSICAL ACTIVITY AND BMI AMONG U.S. LATINOS: DIFFERENCES BY ACCULTURATION FACTORS M.C. Rosal, Stephenie Lemon University of Massachusetts Medical School, WORCESTER, United States of America Introduction: We examined associations between acculturation-related factors, physical activity and sedentary behaviors, and body mass index (BMI) in a sample of Latinos residing in the U.S. mainland who were born elsewhere. Method: Baseline data from a longitudinal study of Latinos in Massachusetts were used. Participants were randomly selected patients from a community health center. Interviewer-administered surveys assessed self-reported ability to speak English, years in the U.S., age first moved to the U.S., number of returns to place of origin, and desire and plans to return to place of origin. Multivariable linear regression models assessed the association of each acculturation-related variable with number of hours of sedentary behavior per week (measured by the Sedentary Behavior Questionnaire), number of minutes of recreational physical activity per week (measured by the Brief Physical Activity Questionnaire) and measured BMI, adjusting for age, gender and education. Results: The sample (n=554) was 52% female with mean age of 48 years. Mean length of time in the U.S. was 18 years, 21% reported being able to speak English well or very well, and 79% were overweight or obese. Mean hours of sedentary behaviors per week was 49.4. Mean minutes of recreational physical activity per week was 127.4. In multivariable models, ability to speak English was associated with more sedentary

S177

behavior (β=17.8; SE=3.5; p=<.0001) and greater BMI (β=1.5; SE=.68; p=.03). No other associations were observed. Conclusion: Results suggest that English-speaking skills increase risk for sedentary behaviors and higher BMI among Latinos who move to the U.S. mainland. CORRESPONDING AUTHOR: Ph.D. M.C. Rosal, University of Massachusetts Medical School, United States of America, [email protected] P596 ASSOCIATION BETWEEN ANXIETY IN FIRST TRIMESTER OF PREGNANCY AND BIRTH PROCESS J.M. Koelewijn1, A.M. Sluijs2, T.G.M. Vrijjkotte1 1 Academic Medical Centre, University of Amsterdam, AMSTERDAM, Nederland 2 Leiden University Medical Centre, LEIDEN, Nederland Objective To identify associations between first trimester anxiety and birth process and interventions; to explore associations by ethnicity and parity. Methods Amsterdam pregnant women participated in the multi-ethnic ABCD-study (inclusion 2003-2004,n=8,266,response 67%). A questionnaire covered socio-demographic data, life-style, anxiety symptoms (STAI state). 6,372 women met the inclusion criteria: singleton pregnancy, child alive at labour start >= 24 weeks. Outcomes were tested with regression analyses. Subgroup analyses were performed for parity and ethnicity (Dutch, Black, Turkish, Moroccan). Results 30.9% showed high anxiety (STAI >=43). Outcome prevalences: elective caesarean section 5.5%, emergency caesarean 9.2%, labour induction 10.5%, augmentation 23.1%, transfer during labour 22.6%, pain relief/sedation 14.6%, delayed first stage 19.6%, delayed second stage 25.9%, instrumental delivery 10.0%. After adjustment for age, parity, education, ethnicity, pre-pregnancy BMI, smoking, alcohol, hypertension, diabetes, gestational age, birth weight, high anxiety was associated with increased risk for pain relief/sedation (OR 1.27; 95%-CI 1.07-1.50). Only in multiparae anxiety was associated with induction (OR 1.55; 95%CI 1.16-2.05) and possibly with primary caesarean (OR 1.32;95%-CI 0.93-1.87). Only anxious Dutch women were at decreased risk for delayed second stage (OR 0.75;95%-CI 0.59-0.95). Anxious Moroccon women might be at lower risk for primary caesarean (OR 0.32;95%-CI 0.10-1.09) and at higher risk for instrumental delivery (OR 2.11;95%-CI 0.89-4.98). Conclusions High anxiety, early in pregnancy, was only associated with increased risk for pain relief/sedation. Anxious Dutch women were at decreased risk for delayed second stage. Anxious multiparae were more often induced. Anxiety is not entirely neutral in its effects on the process of labour. CORRESPONDING AUTHOR: Dr J.M. Koelewijn, Academic Medical Centre, University of Amsterdam, Nederland, [email protected] P597 THE ASSOCIATION OF INVOLUNTARY NON-REGULAR EMPLOYMENT STATUS AND PSYCHOLOGICAL DISTRESS IN JAPAN E.A. Ando, N. Kawakami The University of Tokyo, TOKYO, Japan Purpose:While non-regular emoloyees are known to have poorer mental health, it is not clear if their mental health is affected by selecting this employment status voluntary or involuntary. The object of this study is to examine the cross-sectional association of these involuntary-non regular employment status and psychological distress among Japanese workers. Methods: A total of 2,201 respondents of the 2ndwave from the Japanese Study on Stratification, Health, Income, and Neighborhood (J-SHINE)

S178

study were analyzed. Taking regular employment as a reference, the odds ratio and its 95% confidence intervals (95%CIs) of having psychological distress (K6 score: 5+) were estimated by using multiple logistic regression adjusting for gender, age, marital status, education level, and chronic diseases. Results: Involuntary non-regular employees had a significantly higher prevalence of psychological distress compared to regular employees (OR 1.95 [95%CI: 1.27-2.98]).This tendency remained significant after all covariates were adjusted. Other non-regular employees had a slightly and non-significantly higher prevalence of psychological distress than regular employees. Conclusion: This study suggested that involuntary non-regular employment status is associated with psychological distress, while other non-regular employment may not be less associated with psychological distress. CORRESPONDING AUTHOR: E.A. Ando, The University of Tokyo, Japan, [email protected] P598 EXPERIENCING OTHERNESS: THE EVOLUTION AND DEVELOPMENT OF INTERCULTURAL COMPETENCE OF MEDICAL STUDENTS STUDYING ABROAD N. Faubl, Frédéric Zuhorn, Zsuzsanna Füzesi University of Pécs (Hungary), Medical School, PÉCS, Hungary Introduction. Intercultural competence is an essential part of knowledge of the present and future generations. Our goal is to conduct long-term research of the integration of medical students at the University of Pécs Medical School with empiric investigations, by exploring the elements of development of intercultural communication and competence. Methods. Since research of the development of intercultural competence requires complex research methods, therefore we have applied several methods simultaneously for our long-term follow-up research. Firstly, we have used closed-ended questions, in our self-developed questionnaires distributed several times among students, and also semi-structured personal and group interviews, secondly, we have gained clarifying information due to the participant observant status of our researchers. Results. It is apparent from the findings so far (beginning in: German Program - 2009, English Program - 2011, Hungarian Program - 2012), that regarding their residence in Hungary, the foreign students are motivated primarily by their university studies, rather than by the opportunity to learn about a foreign culture. Although most of them arrive in Hungary with previous foreign experience, therefore despite the lack of preparation, the initial difficulties do not affect them severely on the whole. Also, the students coming for a limited period focussing primarily on studies remain in larger numbers at the ethnocentric phase of development, while the students studying in Hungary for a longer period will more easily rather move to ethnorelativity. CORRESPONDING AUTHOR: Ph.D. student N. Faubl, University of Pécs (Hungary), Medical School, Hungary, [email protected] P599 DEVELOPING INTERCULTURAL COMPETENCE IN MIGRANT HEALTH ENVIRONMENT N. Faubl, Frédéric Zuhorn University of Pécs (Hungary), Medical School, PÉCS, Hungary Introduction. Due to globalization and labour mobility, intercultural competence is a determining element of knowledge for health care professionals, who often contact immigrant patients, also being migrant employees themselves in certain cases. In the medical institution of the Westphalian university town of the research, with a 12% immigrant background, it is important to gain the intercultural competence for both the patients and for the foreign medical professionals.

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

Method. In 2012 a relocation program to relieve doctor shortage was implemented in the hospital. To effectively deal with the experienced difficulties, a complex program was designed, in which - besides the programs to facilitate the soft landing of the medical professionals intercultural trainings were also suggested. After consultations and interviews with the affected groups, we suggested designing trainings for the following groups: health care workers arriving from abroad, medical professionals from the host culture working with immigrant colleagues and partly treating immigrant patients, as well as volunteers of the relocation program. Results. Following the decision regarding the launch of the intercultural trainings, we worked out trainings for the target groups. The leaders made it possible for the migrant health care workers - while ensuring the continuity of their work - to participate in a one-and-a-half-day training of two occasions during their working hours. Although the necessity of team trainings for the host departments was acknowledged, the feasibility was rated low in view of the work overload. CORRESPONDING AUTHOR: Ph.D. student N. Faubl, University of Pécs (Hungary), Medical School, Hungary, [email protected] P600 THE IMPACT OF PRISON OF HEALTH: DO WOMEN PRISONERS HEALTH IMPROVE OR DETEORATE? A. Maia, Joana Alves University of Minho, BRAGA, Portugal Introduction: Women prisoners have been described as vulnerable in terms of health and with high prevalence of diseases. In Portugal little is known about the health of inmates at the moment of incarceration and the impact of imprisonment on health. Method: Health records of 100 women prisoners were used as a source of information to characterize health status, number of contacts with clinical services, and type of health complaints in two moments: the first 30 days of imprisonment and in the 30 days before the study. Results: At T1 participants had on average 1.97 (SD = 1.34) diagnoses of a health problem, 23.7% were hospitalized, and the average number of clinical appointments was 13.01 (SD = 12.36). Somatic complaints were more reported (M = 4.21, SD = 3.39) than psychological complaints (M = 1,8; DP=1.62). At T2 the average number of diagnoses was 1.37 (SD = 1.41), only one inmate was hospitalized, and the average of clinical appointments was 6.28 (SD = 6.34). Complaints with somatic origins were higher (M = 3.68 ;SD = 3.42) than psychological complaints (M=0.74; SD = 1.04 ). Conclusion: These women have poor health status at time of imprisonment observed in the number of diseases and high needs of health care, including hospitalization. These complaints and use of medical services decrease over time. These data reinforce the idea that prison is a place of excellence to intervene and promote health in individuals with weakened health conditions. CORRESPONDING AUTHOR: Prof A. Maia, University of Minho, Portugal, [email protected] P601 GUIDELINES FOR EHEALTH AND SOCIAL MEDIA IN SEXUAL HEALTH PROMOTION FOR YOUNG ETHNIC MINORITIES O. Kulyk1, Chantal den Daas2, Cristel Boom,3, Silke David2, Lisette van Gemert-Pijnen1 1 University of Twente, ENSCHEDE, Nederland 2 National Institute for Public Health and the Environment (RIVM), BILTHOVEN, Nederland 3 Public Health Service Twente, ENSCHEDE, Nederland Introduction: Recently a rapid growth of modern technologies addressing sexuality and health has taken place. Young ethnic minorities could

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

especially benefit from these eHealth initiatives, but they have to meet their specific needs. Sexual health is a sensitive subject in many cultures and eHealth interventions could therefore be useful. We aim to identify what qualities interventions should possess to facilitate qualitatively well-designed and tailored eHealth interventions in the future and to evaluate currently available ones. Method: We investigated what features are important in three focus groups with young ethnic minorities. In total, 28 ethnic minorities participated in the study. We focused on social media, serious games, and mobile applications.Results: Five factors were especially important. The most important factor was anonymity, which has important implications for the use of social networks for sexual health promotion. Additionally, interactivity facilitates engagement and insures better uptake of eHealth technology. Mobility was another crucial factor, as it insures privacy and effortless use. The source of information has to visible and reliable. Specific to young ethnic minorities, eHealth interventions for health promotion also have to be more visually aided and easily worded, if possible in different languages.Conclusions: Applying these factors in the design and implementation of eHealth technologies should increase their uptake and usefulness for enhancing sexual well-being of young ethnic minorities. CORRESPONDING AUTHOR: Dr. O. Kulyk, University of Twente, Nederland, [email protected] P602 STRESS REDUCTION THROUGH SELF-DIRECTED PHYSICAL ACTIVITY, MINDFULNESS MEDITATION, OR HEART RATE VARIABILITY BIOFEEDBACK: AN RCT J.E. van der Zwan1, W. de Vente2, A.C. Huizink1, S.M. Bögels2, E.I. de Bruin2 1 VU University Amsterdam, AMSTERDAM, Nederland 2 University of Amsterdam, AMSTERDAM, Nederland Introduction: The need for easy accessible stress-reducing methods is high. This study compared the efficacy of three stress-reducing self-help interventions”'physical activity (PA), mindfulness meditation (MM), and heart rate variability biofeedback (BF)”'on stress and its related symptoms. Method: Seventy-six participants (20 males; mean age 25.8, range 18-40) were randomly allocated to PA, MM, or BF. Each of the interventions consisted of 2 hours psycho-education and introduction to its specific intervention techniques, followed by 5 weeks of daily exercises at home. For the PA exercises, participants chose a vigorous intensity activity of their liking; the MM exercises consisted of guided mindful meditations recorded on a CD; the BF exercises consisted of slow breathing (~6 breaths per minute) while using a heart rate variability biofeedback device. Participants received daily reminders for their exercises, and they were contacted weekly by phone to monitor progress. Psychological well-being, stress, anxiety, depression, and sleep quality were assessed prior to and after the training (pre-test and post-test), and 6 weeks later (follow-up), using questionnaires. Results: Results indicate an overall beneficial effect consisting of reduced stress, anxiety and depressive symptoms, and improved psychological well-being and sleep quality. No significant between-intervention effect was found on any of the outcome variables. Conclusions: Results suggest that PA, MM, and BF are equally effective in reducing stress and its related symptoms. The best intervention for someone may thus be the intervention that one finds easiest to commit to. These self-help interventions provide easy accessible help for people with stress complaints. CORRESPONDING AUTHOR: J.E. van der Zwan, VU University Amsterdam, Nederland, [email protected] P603 IMPLICIT AFFECT AND CARDIOVASCULAR RESPONSES TO MENTAL ARITHMETIC WITH AND WITHOUT ANGER HARASSMENT M.M. van der Ploeg, J.F. Brosschot, B. Verkuil Leiden University, LEIDEN, Nederland

S179

Introduction: Non-conscious stress might partially explain stress-induced increased cardiovascular (CV) activity and eventually contribute to the development of CV disease. Non-conscious stress can be assessed by implicit negative affect (INA) tests. INA, measured with the Implicit Positive and Negative Affect Test (IPANAT), has been shown to predict cortisol response to acute stress. The present study examined whether INA, in addition to explicit negative affect tests (ENA), explained differential CV responses to mental arithmetic with and without harassment. Method: Students were randomly assigned to perform a mental arithmetic task with (n=14) or without (n=16) anger harassment. Afterwards, the INA and an ENA test (Visual Analogue Scale) were administered. Systolic (SBP) and Diastolic (DBP) Blood Pressure and Heart Rate (HR) were measured continuously. Results: In the harassment condition SBP and DBP reactivity was higher (t(27) = 3.38, p <.01 and t(27)= 2.13, p <.05, respectively), and SBP recovery was slower (t(28)= 2.12, p <.05). No effects of condition on INA were apparent, but ENA was higher after harassment (t(28) = -2.91, p <.01). Additional analyses showed that increases in HR during both tasks were associated with higher INA (r2 change = 0.13, p = 0.05) but not with higher ENA. However during recovery, only in the harassment condition higher INA was associated with higher HR (r = 0.17, p <.05). Conclusion: The results suggest that INA as measured with the IPANAT does not differentiate between stressors, but high INA seems to augment CV responses, especially during recovery after harassment. CORRESPONDING AUTHOR: M.M. van der Ploeg, Leiden University, Nederland, [email protected] P604 STRESS IS ASSOCIATED WITH DIMINISHED SEXUAL INTEREST IN A SWISS STUDENT POPULATION C. Markert, S. Fischer, U.M. Nater University of Marburg, MARBURG, Germany Introduction: There is a widespread notion that stress and sexual desire are closely linked to each other. However, there is a fundamental lack of studies testing the assumption that stress may lead to sexual dysfunction. The objective of this study was therefore to multi-dimensionally measure stress and examine its relation to sexual desire in an apparently healthy sample. Methods: A total of 3054 Swiss students (mean age 24.57 years) took part in a study about physical and mental well-being. Participants were asked whether during the past month they had been bothered by “little or no sexual desire or pleasure during sex”. Chronic stress during the past 3 months, dispositional stress reactivity, stress burden within last month, and childhood trauma were assessed. Results: 2241 participants were women (73.4%). Three groups were formed based on whether they were “not bothered”(2168 subjects),”bothered a lot”(235 subjects), or “bothered a little”(651 subjects) by diminished sexual desire or pleasure. There were significant differences in all dimensions of stress between the three groups (all comparisons p<.001), with highest stress values in the group being “bothered a lot”. Conclusion: Diminished sexual desire or pleasure seems to be relatively common in our university population, particularly in those reporting high stress levels. Longitudinal studies are needed to examine the causal relationship between these factors. Healthcare providers should address stress-related issues when dealing with sexual problems. CORRESPONDING AUTHOR: C. Markert, University of Marburg, Germany, [email protected] P605 LINKING SELF-REPORTED STRESS TO AGGREGATE CORTISOL MEASURES AND RECURRENT PAIN IN MIDADOLESCENT GIRLS AND BOYS P. Lindfors1, L. Folkesson1, V. Östberg2 1 Stockholm University, STOCKHOLM, Sweden 2 Centre for Health Equity Studies, Stockholm University/Karolinska Institutet, STOCKHOLM, Sweden

S180

Introduction: Previous research has developed an 11-item self-report measure assessing activation and pressure stress among adolescents. However, the biological correlates of this measure are unclear. Considering this, the present study investigated how perceived stress relates to aggregate cortisol measures and recurrent pain in mid-adolescent girls and boys. Methods: Mid-adolescents (119 girls and 56 boys) provided self-ratings of activation and pressure stress and recurrent pain (headache, stomachache, neck/shoulder and back pain) in questionnaires. Additionally, adolescents sampled saliva during an ordinary school day: 1) immediately at awakening, 2) 30 minutes after waking up, 3) 60 minutes after waking up, and 4) at 8 p.m. Aggregate cortisol measures including ground and increase measures of the area under the curve and a diurnal slope measure were computed. Hierarchical regression analyses were performed for girls and boys respectively. Results: Activation and pressure were significantly associated with recurrent pain in girls but not in boys. However, there were no significant associations between self-ratings of stress and salivary cortisol, neither for girls nor for boys. Conclusion: While self-rated activation and pressure stress were related to recurrent pain in girls, but not in boys, neither activation nor pressure was linked to any of the aggregate cortisol measures. The differences between subjective and objective measures perhaps relate to these measures reflecting distinct and unrelated aspects of functioning. However, the findings may also result from the participants being mid-adolescents whose bodily systems are flexible and still unaffected by daily activation and pressure stress. CORRESPONDING AUTHOR: Professor. Lindfors, Stockholm University, Sweden, [email protected] P606 FROM PHYCHOSOMATIC INTEGRATION TO PHYCHOSOMATIC HEALTH: PHYCHOSOMATIC PRINCIPAL,TECHNIQUE AND HEALTH IFFECTS H. Wu1, J.Z. Tian1, Q.S. Geng2, Li Chen3, Ai-hua. OU4, Qi. Wang4, S.M. Zhang1, J.J. HE1 1 The second people's hospital of Guangdong province, GUANGZHOU, China 2 The people's hospital of Guangdong province, GUANGZHOU, China 3 The Health and Family Planning Commission of Guangdong province, GUANGZHOU, China 4 Traditional Chinese medicine university of Guangzhou, Guangzhou, China Purpose: To get the psychosomatic unification principle, technique and its effects. Method: Analyzed the research results of system science, modern medicine, traditional Chinese medicine, psychology, qigong, Internal Martial Arts, yoga, meditation, Sinology, philosophy, etc, with integrated application of the Deductive and Inductive methods. Result: Under the condition of reasonable body posture and psychosomatic relaxation, we can establish psychosomatic biofeedback by paying attention to our own inner body, which leads to better processing of physiological and pathological information, and improves the ability of self-organization to maintain psychosomatic health. In history, humans have consciously or unconsciously developed many kinds of health techniques such as qigong, yoga, Internal Martial Arts and meditation according to this principle, and it is also the common nature of these techniques. Under the guidance of psychosomatic integration principles, we integrated the essence of Internal Martial Arts, qigong meditation, yoga, and so on to establish the psychosomatic integration technique. Psychosomatic integration emphasizes psychosomatic participation and psychosomatic holistic health, and it will greatly enhance the level of psychosomatic health and moral cultivation. Conclusion: Psychosomatic integration is a behavioral intervention technique under the guidance of system science, and it is the embodiment of

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

system medicine and health medicine. Psychosomatic integration will greatly promote the development of system science, system medicine, health medicine, and psychosomatic medicine, etc. CORRESPONDING AUTHOR: H. Wu, The second people's hospital of Guangdong province, China, [email protected] P607 WHAT IS QUANTITATIVE WORKLOAD FOR JAPANESE SCHOOL TEACHERS? M. MATSUOKA1, Y.O. OTSUKA2 1 Hiroshima University / Matsuoka Counseling Office, HIROSHIMA, Japan 2 Hiroshima University, HIGASHI-HIROSHIMA, Japan Introduction: This study aimed at finding the concrete contents of quantitative workload as stressor for Japanese school teachers. Methods: The questionnaire was distributed to 132 Japanese elementary, junior-high, middle, high, and special support school teachers. A hundred and ten (42 male, 65 female, and 3 missing values) responded to the question, “To what jobs, do you feel quantitative workloads as burden (for example, clerical work, extracurricular club activities, allotment school management duties, helping students)?”in August and September, 2012. The KJ method (a qualitative research method classified the answers into several categories) was utilized for the analysis. Eleven people including 3 undergraduates, 7 graduates, and 1 research student specializing in clinical psychology engaged in the task of card sorting. Results: One hundred and one items raised as quantitative workloads were classified into 45 categories. Thirty six teachers mentioned clerical work as quantitative workload. Thirty five cited allotment work of managing schools, and 34 raised preparation for classes as quantitative workload. Nineteen perceived helping students and extracurricular club activities, 10 referred to student guidance, and 9 named corresponding with students’ guardians as quantitative workload. Five raised the documentation of teaching plans, marking, and job spill and holiday work as quantitative workload. Conclusions: Clerical work, allotment work of managing schools, and preparation for classes were mainly perceived as quantitative workload among Japanese school teachers. In addition, many items might have reflected the Japanese teachers’ working situation that they are expected to engage in many tasks other than teaching. CORRESPONDING AUTHOR: Ms. M. MATSUOKA, Hiroshima University / Matsuoka Counseling Office, Japan, [email protected] P608 STRESSFUL EVENTS: IMPACT ON ASTHMA MEDICATION USE R. Lietzén1, P. Virtanen2, M. Koskenvuo3, J. Vahtera1 1 University of Turku, TURUN YLIOPISTO, Finland 2 University of Tampere, TAMPERE, Finland 3 University of Helsinki, HELSINKI, Finland Introduction. Stressful life events increase the risk of asthma onset in adults, but the extent to which they play a role in in exacerbation of asthma is unknown. Methods. This is a prospective study of 1,102 asthmatic adults. Participants were linked to records of filled prescriptions for inhaled bronchodilators from national registers for seven years. For the purpose of this study, lifeevent exposure (14.7% high exposed; 66.6% unexposed) was assessed. Results. After exposure to highly stressful events, the rate of filled prescriptions for bronchodilators was 54% higher (95%CI 1.32-1.80 adjusted for demographics) than before the events occurred (mean 1.26). As unexposed, the corresponding rate ratio was 1.14. Conclusion. Our results suggest that bronchodilator usage increases following exposure to highly stressful life events and this increase may last years.

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

CORRESPONDING AUTHOR: R.L. Lietzén, University of Turku, [email protected] P608 PERCEIVES STRESS AND CORRELATES IN A REPRESENTATIVE HUNGARIAN POPULATION SAMPLE A. Stauder Semmelweis University Budapest, BUDAPEST, Hungary Aims: This is the first study providing normative Hungarian population data for the widely used Perceived Stress scale, allowing national and international comparative studies. Methods: In the Hungarostudy 2013 survey 2000, individuals completed structured interviews. This sample was representative of the Hungarian adult population according to gender, age groups, education and type of settlement. Besides Cohen’s Perceived Stress Scale (PSS10), the short Beck Depression Inventory (BDI), the Patients Health Questionnaire (PHQ15) for subjective somatic symptoms, and demographic and socioeconomic variables were included in the analysis. Results: PSS10 mean was significantly higher in women (16,3±7,0; N=1059) than in men (14,9±6,9; N=927). PSS10 was not in linear correlation with age, the 50-59 age group had the highest stress level in both gender. Perceived stress was negatively correlated with level of education (pearson r=-0,177) and with relative economic situation (pearson r=-0,372). PSS10 was correlated with depressive symptoms (r=0,560) and subjective somatic complaints (r=0,411). Concerning employment status unemployment and disability retirement were associated with the highest perceived stress level in both gender (PSS10 mean > 18). Hours of work and shift work did not correlate significantly with PSS10. The lack of control at work predicted slightly higher stress levels with the fear of losing the job more significant. Discussion: When compared with international data, we found similar relationships between PSS scores and basic demographic or socioeconomic factors as other studies; however the Hungarian PSS10 scores were significantly higher than those found eg. in US or Danish populations. CORRESPONDING AUTHOR: A. Stauder, Semmelweis University Budapest, Hungary, [email protected] P609 STRESS REDUCTION IN YOGIC RELAXATION DURING UPPER GI ENDOSCOPY M.R. Kotwal Government of Sikkim, GANGTOK, India Introduction: Many patients fear upper gastrointestinal endoscopies. Shavasan is probably the best known relaxation exercise. Through relaxation of muscle tension, an anxiety free sate is reached. Objective: To examine effects of relaxation in Shavasan posture in patients undergoing GI endoscopies. Methods: Study was conducted on 63 consecutive patients 16 to 86 years of age. Patients were randomly assigned to two groups regardless of sex, age and underlying disease. Thirty one patients relaxed in shavasan, before and during procedure, while 32 patients did not. No patients were given sedation or local anesthetic spray. Blood pressure, heart and respiratory rate were recorded. Perception of procedure using a 5- point attitude scale was assessed. Results: Paired T-test was used to analyze the data. Test group results showed significant difference in three parameters i.e. blood pressure, systolic, heart and respiratory rate in the difference of values at the beginning and end of procedure. Control group did not show statistically significant difference in any parameters. However, data comparison in two groups showed statistically significantly difference in the heart and respiratory rate. More patients in test group reported decrease in distress. Conclusion: This preliminary study shows the effect of relaxation in Shavasan posture. Repeated practice is necessary to attain quick de-

S181

stressed state in distress. We suggest that this exercise could be applied to any other medical situations which tend to generate undue psychological stress and anxiety as well. CORRESPONDING AUTHOR: Prof M.R. professor Kotwal, Government of Sikkim, India, [email protected] P610 STRESS REDUCTION BY LISTENING INDIAN CLASSICAL MUSIC DURING UPPER GASTROINTESTINAL ENDOSCOPY M.R. Kotwal Government of Sikkim, GANGTOK, India Background: Many patients fear GI Endoscopy. Natural anxiety may be aggravated by horror stories from friends or inappropriate remarks by endoscopy staff. Music serves on familiar conjunctures, such as in waiting rooms, and air travel, helping us to relax or increase our patience. Objective: To evaluate scientific and therapeutic possibilities. Method: Study conducted on 110 consecutive patients undergoing GI endoscopy. Fifty-five patients listened to the recorded music, while the other group of 55 did not. BP, heart and respiration recorded at beginning and procedure end. Procedure perception attitude scale analyzed. Results: Using paired T-test, no statistically significant difference in the four parameters BP-S, BP-D, Heart and Respiratory rate. However data analysis between two groups had statistically significant differences in three parameters i.e. Bp-S, BP-D, R-Rate. Conclusion: Selective music is efficacious in reducing psychological distress during gastroscopic examination. Music could be applied to other medical situations as well, which tend to generate undue stress and anxiety. Music as a familiar personal and culture medium could be used to ease anxiety, to act as a distracter, to increase discomfort and pain threshold. CORRESPONDING AUTHOR: Prof M.R. professor Kotwal, Government of Sikkim, India, [email protected] P611 BIOFEEDBACK TREATMENT FOR GENERALIZED ANXIETY DISORDER (GAD) E. Salinas-Rivera1, P. Campos-Coy2, J.J. Sánchez-Sosa1 1 National University of Mexico, MEXICO CITY, Mexico 2 National Institue of Psychiatry (INPRF), MEXICO CITY, Mexico Introduction: Generalized Anxiety Disorder (GAD) is one of the most prevalent anxiety disorders worldwide. In Mexico, its twelve-month prevalence stands around 0.4% in the general population representing over 400,000 people. GAD involves excessive, uncontrollable and chronic worrying, nervousness, irritability and increased muscle tension which deteriorate the individual's quality of life. Suffering increases considerably when GAD is accompanied by other symptoms, such as depression. Thus, it becomes necessary to develop effective treatment interventions aimed at reducing anxiety and depression symptoms, and increasing quality of life. Biofeedback-based procedures have shown efficacy at reducing somatic symptoms associated to GAD. However, it is still necessary to examine whether these positive effects can be potentiated through simultaneous recording and training of multiple physiological responses in the context of several cognitive-behavioral techniques. The purpose of the present study was to test the efficacy of a multipleresponse biofeedback training protocol in reducing anxiety and depression symptoms and increasing quality of life. Method: Twelve patients (4 men, 8 women) with GAD primary diagnosis and secondary depression received nine biofeedback training sessions individually, comprising six anxiety-associated physiological responses. Results showed clear positive changes in all psychological measures at post treatment and at 2 month follow-up evaluations. Psychological improvement was supported by a reduction in somatic activation of some

S182

of the physiological responses evaluated (respiration amplitude, respiration frequency and muscle tension). Conclusion: Biofeedback procedures appear to be an alternative intervention option for GAD patients. CORRESPONDING AUTHOR: E. Salinas-Rivera, National University of Mexico, Mexico, [email protected] P612 PSYCHOLOGICAL MORBIDITY AND IMPAIRED QUALITY OF LIFE IN PATIENTS WITH PRIMARY ADRENAL INSUFFICIENCY C.D. Andela LUMC, LEIDEN, Nederland Introduction: A high prevalence of psychological morbidity, maladaptive personality and impaired quality of life is observed in patients following Cushing's syndrome. It is currently unclear whether a similar pattern is present in patients with primary adrenal insufficiency (PAI). Therefore, the aim of this study was to evaluate psychological functioning, personality traits, and QoL in patients with PAI. Methods: A cross-sectional study including 54 patients with PAI and 54 healthy matched controls. Both patients and controls completed questionnaires on psychological functioning (Apathy Scale, Irritability Scale, Mood and Anxiety Symptoms Questionnaire short-form, and Hospital Anxiety and Depression Scale), personality traits (Dimensional Assessment of Personality Pathology short-form), and QoL (Multidimensional Fatigue Inventory, Short-Form 36, EuroQoL-5D, Nottingham Health Profile, and Physical Symptom Checklist). Results: Patients with PAI suffered from more psychological morbidity (i.e. irritability and somatic arousal) and QoL impairments compared with controls (all P<0.01). There were no differences regarding maladaptive personality traits between patients and controls. However, there was a strong and consistent positive association between the daily hydrocortisone dose and prevalence of maladaptive personality traits (i.e. identity problems, cognitive distortion, compulsivity, restricted expression, callousness, oppositionality, rejection, conduct problems, social avoidance, narcissism, insecure attachment (P<0.05)). There was also a strong relation between the mean daily hydrocortisone dose and both psychological morbidity (i.e. depression, P<0.05) and QoL impairments (i.e. general health perception, several measures of physical functioning, vitality (P<0.05)). Conclusion: Patients with PAI report psychological morbidity and impaired QoL. Psychological morbidity, impaired QoL, and maladaptive personality traits were all associated with higher hydrocortisone dosages. CORRESPONDING AUTHOR: MSc C.D. Andela, LUMC, Nederland, [email protected] P613 DANCE FOR HEALTH AND COPING: A CROSS-SECTIONAL PILOT STUDY IN HUNGARY A. Nagy1, V. Premusz1, L. Lampek2, T. Tigyi2, H. Pusztafalvi2, V. Bobály1 1 University of Pécs, Faculty of Health Sciences, Doctoral School Health Sciences, PÉCS, Hungary 2 University of Pécs, Faculty of Health Sciences,Institute of Health Insurance, PÉCS, Hungary Introduction: The majority of young people do not develop the ability to cope with difficulties, they are not exposed to enough challenges, especially in the physical sense. Only regular physical exercise can replace this. The aim of this study was to assess dance as a unique form of exercise that has a different effect on coping. Methods: A cross-sectional, quantitative pilot study with convenience sampling was performed in the city of Pécs, south-west Hungary. A total of 102 adolescent girls and young adult women, 16-37 years old participated in the study.

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

Three groups were compared in the trial: participants of a modern and contemporary dance course, athletes of the University of Pécs, and students of the University of Pécs who don't engage in regular exercise. Self-filled standardized questionnaires were used on the topic of the stress and coping - Beck and Weismann DAS questionnaire, Rahe life event questionnaire and Lazarus and Folkman resolution of conflict questionnaire. Results: It became possible to determine the individuals' degree of coping with, and their chosen coping strategies. A group of athletes and nonathletes were significantly more likely (p <0.05), to demonstrate dysfunctional attitudes causing negative affectivity, while 24% of a group consisting of dancers did not display dysfunctional attitudes. The selfadmitted standard of health of the dancer group was significantly better than the other two groups. Conclusion: Dance seems to have a special significance compared to other forms of exercise. CORRESPONDING AUTHOR: Dr. A. Nagy, University of Pécs, Faculty of Health Sciences, Doctoral School Health Sciences, Hungary, [email protected] P614 SALIVARY STRESS HORMONES, EMOTIONAL RESPONSES AND PUBERTAL STATUS AFTER THE PSYCHOSOCIAL STRESS IN JAPANESE TWINS M. D. Deno1, T. Tachikawa1, S. Izawa2, K. K. Fujisawa3, M. Tanaka4 1 Musashino University, TOKYO, Japan 2 National Institute of Occupational Safety and Health, TOKYO, Japan 3 Keio University, TOKYO, Japan 4 Center for Forensic Mental Health, Chiba University, CHIBA, Japan Introduction:We examined the association of the physical and the emotional responses after psychosocial stress burden and the pubertal status among the monozygotic twin girls. This twin study examined these relationships that were not confounded by genetic and family environmental factors. Method:Twenty-six pairs of healthy girl monozygotic twins were exposed to the Trier Social Stress Test for Children (TSST-C). Saliva samples were collected for cortisol and dehydroepiandrosterone (DHEA) levels and a subjective symptoms test and a visual analogue scale were used to collect subjective emotional response before and after the TSST-C. They also answered their puberty status and the efficacy of controlling and expressing their feelings. The correlation analysis was examined on the phenotypic scores and the monozygotic twin difference. Results:With controlling of genetic and common environment, the pubertal status did not significantly correlate with AUCg of DHEA (r = .083), that was an entire reverse result to the phenotypic correlation (r = .445, p < .01). The emotional responses toward psychosocial stress significantly correlated with the efficacy of controlling their feelings (r = -.640, p < .01), whereas they did not correlated with the efficacy of expressing their feelings (r = -.173) with the monozygotic twin difference scores. Conclusions:The phenotypic association of the pubertal status and DHEA might be explained by genetic and common environment. In addition, the relationships between the emotional responses toward psychosocial stress and the efficacy of controlling and expressing their feelings might be influenced by the different factors; genetic and common environment or non-shared environment. CORRESPONDING AUTHOR: M. D. Deno, Musashino University, Japan, [email protected] P615 WHAT MARS POLICIES TO MODIFY RISK BEHAVIORS IN LMICs? M.B. Aghi Advocacy Forum for Tobacco Control, NEW DELHI, India

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

Introduction. Risk behaviors like tobacco and alcohol use leads to many deaths globally ( tobacco smoking kills nearly 6 million people each year, 5 due to active smoking and the rest due to secondhand smoke exposure. This does not include deaths due to smokeless tobacco use. 320 000 die annually due to alcohol use). Methodology. In order to mitigate the adverse effects of these two governments all over the world introduce alcohol and tobacco control policies to regulate their use. Extreme policies like prohibition and bans have been often found to counter-productive. Many countries have espoused a nuanced approach involving the regulation, availability and legislation with rigorous enforcement.But while formulating them, much too often the governments especially in Low-income countries neglect to involve the populations who are target of the policies Results. Consequently those who need to follow the policies do not understand the rationale why these policies are made and this becomes a road block in their implementation. Therefore wisdom tells us that we hold consultations with populations both users and non users. This will enable us to understand their felt information needs, their attitudes and profiles, their misconceptions, useful beliefs, their sources of motivation and their thought on tactics to make their environments enabling and friendly and fit polices into these. Conclusions. Policies so formulated have been successful like banning smoking in public places in Ireland, Thailand and in the city of New york. CORRESPONDING AUTHOR: Dr M.B. Aghi, Advocacy Forum for Tobacco Control, India, [email protected] P616 SCALING UP SMOKING CESSATION INTERVENTIONS FOR HOSPITAL PATIENTS: FROM CLINIC TO A NETWORK OF HOSPITALS J.H. Wiggers1, L. Wolfenden1, M. Freund1, E. Campbell2, P. Wye1, K. Gillham2, C. Slattery2, J. Knight2, J. Bowman1 1 University of Newcastle, WALLSEND, Australia 2 Hunter New England Local Health District, WALLSEND, Australia Introduction. Despite the public health and clinical benefits of smoking cessation, most tobacco users do not receive smoking cessation support recommended by clinical practice guidelines when they attend hospital. The aim of this paper is to report on the impact of an approach to improving the provision of smoking cessation care across an entire health service region geographically the size of England. Method. We conducted randomised, quasi-experimental and longitudinal studies to determine the impact of clinical practice change interventions in improving the provision of smoking cessation support provided to hospital patients in New South Wales, Australia.Multi-component clinical practice change interventions involving staff training, point of care prompts, resources, performance feedback and executive level support and endorsement were implemented at a clinic, whole of hospital (2 hospitals) and whole of hospital network (38 hospitals) levels. Results. Significant improvements in the provision of smoking cessation care, such as assessment of smoking status, provision of brief advice and nicotine replacement therapy was found at each level. Despite the significant effects, variability in the size of effect in improving the provision of smoking cessation care was evident. Conclusion. The findings demonstrate that the delivery of smoking cessation care in hospitals can be increased and that such improvements can be achieved at scale. Variability in effect suggests further research is required to optimise benefits for patients. CORRESPONDING AUTHOR: Professor H. Wiggers, University of Newcastle, Australia, [email protected]

S183

P617 ADVANCING PATIENT-CLINICIAN INTERACTIONS THROUGH SPECIFIC GUIDANCE: A SECONDARY ANALYSIS TO IDENTIFY CLINICIANS' BEHAVIOURS FOR IMPLEMENTATION J. McAnuff1, N. Kolehmainen2 1 Leeds Community Healthcare NHS Trust, LEEDS, United Kingdom 2 Newcastle University, NEWCASTLE, United Kingdom Introduction: The behaviours clinicians take to interact with patients affect health outcomes. Two broad types of interactions are recommended: relational (e.g. listening, empathy, respect) and participatory (e.g. shared decision making and problem solving); but specific guidance on how to do these in practice is scarce. We aimed to identify clinicians’ specific, real-life behaviours that illustrate these interactions in order to provide guidance to clinicians. Methods: A qualitative secondary analysis in one domain of healthcare (child and family health). The data consisted of a list of clinicians’ selfreported practice behaviours (n=217) and direct quotes describing their content and context. The behaviours were categorised into relational and participatory family-centred interactions by using a modified framework analysis. Results: Of the 217 behaviours, 19 (8.76%) illustrated ‘relational’ interactions (e.g. ‘Listen to parent’, ‘Gather perspectives from others’); 47 (21.66%) illustrated ‘participatory’ interactions (e.g. ‘Identify the family’s goals and priorities’, ‘Allow the family to choose’); and two (0.92%) illustrated both (‘Take guidance from parents’). The majority (121 [55.76%]) did not illustrate interactions but other behaviours (e.g. administer a test) and 28 (12.9%) were excluded from the analysis. Conclusions: Illustrations of both participatory and relational behaviours were identified, although both were described less frequently than other clinician behaviours. The real-life, practice-based illustrations of clinicians’ behaviours are now used to provide specific guidance to clinicians on the actions thet can take to implement good patient-clinician interactions. CORRESPONDING AUTHOR: J. McAnuff, Leeds Community Healthcare NHS Trust, United Kingdom, [email protected] P618 WORKING WITH EDUCATORS TO MOBILISE BEHAVIOURAL SCIENCE IN HEALTH PROFESSIONAL TRAINING: UK AND AFRICA L. Byrne-Davis1, J.K. Hart1, C.A. Armitage1, M. Johnston2, G.J. Byrne3 1 University of Manchester, MANCHESTER, United Kingdom 2 University of Aberdeen, ABERDEEN, United Kingdom 3 Health Education North West, MANCHESTER, United Kingdom Introduction: Educational courses are delivered across the globe with an (often implicit) aim to change health professional behaviour. Despite this, the literature suggests that educators are largely unaware of behavioural science, and evaluate their courses in terms of learner satisfaction rather than behaviour change. This paper presents data about the collaborative challenges and outcomes of engaging with educators. Methods: We evaluated the introduction of the COM-B model (Michie et al., 2011) of behaviour change to 3 different groups of educators designing and delivering training in acute illness management in the UK and Uganda. We consulted with our University Ethics Committee about requirements of international research on education and training. Results: Respondents all agreed that the behaviour change concepts were appropriate for evaluation of training and were keen to use them. Educators in Uganda have begun to adapt their course based on the findings of behaviour change evaluations. Educators in the UK have piloted behaviour change evaluations and found them to be acceptable. Guidance on research governance has resulted in approved methods.

S184

Conclusions: Collaboration between educators and behavioural researchers can yield mutual benefits. Ultimately, this streamlines the mobilisation of knowledge and makes knowledge creation more timely and cost-effective. We have begun to create a database of behaviour change concepts and changes over time, related to a growing number of courses. Behavioural researchers in public health and behavioural science would benefit from creating solid collaborative relationships with the stakeholders whose core business is the design and delivery of interventions. CORRESPONDING AUTHOR: Dr L. Byrne-Davis, University of Manchester, United Kingdom, [email protected] P619 OBESITY PREVENTION POLICIES AND PRACTICES IN AUST RALIAN PRIMARY SCHOOLS 2006-2013: HOW FAR HAVE WE COME? J.H. Wiggers1, N. Nathan2, L. Wolfenden1, C.M. Williams1, S.Z. Yoong1, C. Lecathelinais2, A.C. Bell3, R. Wyse1, R. Sutherland2, K. Gillham2 1 University of Newcastle, WALLSEND, Australia 2 Hunter New England Local Health District, WALLSEND, Australia 3 Deakin University, GEELONG, Australia Purpose: Despite significant investment over the last decade, the extent to which schools have implemented healthy eating and physical activity policies and practices has not been routinely reported. The aims of this paper are to; describe the prevalence of healthy eating and physical activity policies and practices in Australian schools over time and; to determine if school size, rurality or socio-economic location influences practice adoption. Methods: Over an eight year period, a representative randomly selected cohort of primary schools in NSW, Australia (n=476) participated in four telephone surveys regarding their implementation of 4 healthy eating and 5 physical activity policies and practices. Repeated measures logistic regression in a GEE framework was undertaken to examine trend over time. Results: Between 2006 and 2013 there was a significant increase in the number of schools adopting all healthy eating practices; vegetable and fruit breaks was the largest (42.6% to 88.5%, p<0.0001). Of the five physical activity practices, only one significantly improved over time i.e. schools provision of playground markings and sports equipment during recess and lunch (52.3% to 70.2%, p<0.0001). Adoption of practices did not significantly differ by school characteristics. Conclusion: Government policy/ investment can equitably improve school adoption of obesity prevention policies and practices on a jurisdiction-wide basis. Additional and/or different dissemination strategies may be required to increase adoption of physical activity practices. Routine monitoring of such outcomes over time is needed to ensure the intended benefits of government policy and investment are achieved, and the need for remedial action identified. CORRESPONDING AUTHOR: Professor J.H. Wiggers, University of Newcastle, Australia, [email protected] P620 UNDERSTANDING PUBLIC ATTITUDES TO ORGAN DONATION: A MEDIA ANALYSIS P.B. Hutchings, P. M. Grey, S. J. Manchipp, C. Phelps University of Wales Trinity Saint David Swansea, SWANSEA, United Kingdom Introduction. Following the passing of the Human Transplantation (Wales) Bill, a ‘deemed consent’ (‘opt out’) system for organ donation will be brought into effect in Wales (UK) in December 2015. This research report was commissioned by the Welsh Government to explore whether differences between regions in relation to both knowledge of

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

proposed changes in legislation and attitudes toward the new legislation were due to media influence. Method. A retrospective content analysis of media coverage of organ donation legislation from Television, Newspapers and Social Media sources in Wales was conducted. A message-framing approach (Tversky and Kahneman, 1981) was employed, utilising an algorithm to examine whether information was framed positively or negatively; argument strength; and source credibility. Results. Regional differences in attitudes towards the legislation did not reflect differences in presentation of information across different types of media in Wales. Political themes featured in more negative television stories whilst statistical health themes were found in more positive television stories, with a clear pattern of presenting factually objective stories relating to organ donation legislation rather than religious, moral or emotionally-laden themes. In newspapers positive themes reflected emotional, statistical and political information whilst negative themes reflected political, moral and religious arguments. Conclusions. Research into the way media reports key political campaigns and debates can identify how information is depicted to target audiences. Whilst there is clearly a need to explore wider psychosocial factors that account for differences in attitudes, such analyses can provide valuable information for future targeted government communication campaigns and policies. CORRESPONDING AUTHOR: Dr. P.B. Hutchings, University of Wales Trinity Saint David Swansea, United Kingdom, paul.hutchings sm.uwtsd.ac.uk P621 INCONSISTENT VIEWS ON MAIN PROVOKING FACTORS OF SELF MEDICATION IN IRAN R. Shaghaghi, Marzieh Asadi, Hamid Allahverdipour Tabriz University of Medical Sciences, TABRIZ, Iran Introduction: Self-medication could pose serious consequences for communities. The degree of consensus among physicians and pharmacists about precipitating factors of the behavior have not yet been investigated in Iran. Consistency of their prospects on the contributing factors of selfmedication was scrutinized. Methods: In this cross-sectional study, a random sampling method was used to recruit 140 pharmacists and 364 physicians practicing in Tabriz, capital city of the East Azarbaidjan province in the North West of Iran. Researcher structured questionnaires were used to collect data through face to face interviews. Indices of central tendency and variability and Phi test were applied to analyze and report the findings. Results: Availability of non-over the counter (OTC) drugs without prescription (98%), prescription drugs by pharmacists (90%) (from the perspective of the physicians), lack of knowledge about the drugs' side effects (81%), and the high costs of visiting physicians (75%) (from pharmacists' point of views) were the main reasons for self-medication. No consensus was observed regarding the effect of the costs of a physician visit ('=0.13) and the availability of non-OTC drugs ('=0.11) on selfmedication. Conclusions: The required consensus about the contributing factors of self-medication, as a main requisite in attempts to eliminate the behavior, does not exist between two important players in the Iranian health care network. This may explain the failure to cut down self-medication in Iran and other developing countries. CORRESPONDING AUTHOR: Dr. R. Shaghaghi, Tabriz University of Medical Sciences, Irana, [email protected] P622 ELECTROMAGNETIC HYPERSENSITIVITY THROUGH MULTIPLE LENSES - A PILOT STUDY R. Szemerszky1, P. Varsányi2, Á. Révész3, G.y. Thuróczy4, F. Köteles1

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216 1 Institute for Health Promotion and Sport Sciences, Eötvös Loránd University, BUDAPEST, Hungary 2 Hieronymus Bock School for Traditional Herbal Knowledge, BUDAPEST, Hungary 3 Department of Personality and Health Psychology, Eötvös Loránd University, BUDAPEST, Hungary 4 Department of Non-Ionizing Radiation, National Research Inst for Radiobiology, BUDAPEST, Hungary

Background: Electromagnetic hypersensitivity (EHS) is a condition in which patients report symptoms following perceived exposure to weak electromagnetic fields (EMFs). Our current knowledge on the development (aetiology) and treatment of EHS is scarce. This study is a first attempt to develop an interdisciplinary approach including assessment of possible biophysical and psychological background and evaluating a psychological intervention. Method: In this case study, first medical, psychological, and physicalsocial environmental profiles of a patient with serious EHS were assessed. Causal relationship between EMF exposure and EHS symptoms was investigated in a double-blind provocation study and in an ecological momentary assessment (EMA) study. In the latter, actual EMF exposure, physiological (ECG, respiration rate) and subjective variables (symptoms, mood, perceived exposure, etc.) were registered simultaneously for a week before and after the intervention, and six months later. Treatment accepted by the patient was psychotherapy with personally suited combination of Bach flower remedies (a complementary method). Results: Neither provocation study nor EMA measurements supported the existence of a causal relationship between symptoms and EMF exposure. Assessment of the patient’s medical-psychosocialenvironmental background enabled us to understand the aetiology of her EHS. Already after four therapeutic consultations, the patient's wellbeing and life quality were improved, however, the perceived symptoms did not attenuate. The patient accepted that not the EMF exposure but her reaction to the exposure should be changed. Conclusion: This complex approach shed more light on the aetiology of EHS and enhanced the life quality of a patient with serious EHS. CORRESPONDING AUTHOR: Dr R. Szemerszky, Institute for Health Promotion and Sport Sciences, Eötvös Loránd University, Hungary, [email protected] P623 POSITIVE AND NEGATIVE EMOTIONS IN WELL-BEING: A SELF-OBSERVATIONAL STUDY L. Sibilia SIMPS, ROMA, Italy Stefania Borgo, MD, Centre for Research in Psychotherapy, Rome, Italy. Introduction. In recent years the association between emotional states and health has been widely investigated suggesting that negative states and traits are linked to poor health. On the contrary positive emotions are supposed to play an important role in salutogenesis. The study here presented aimed at investigating the role of positive and negative emotions in subjective wellbeing. Method. More than 1000 self-observational forms were collected in 7 emotional states: 5 negative (anxiety, anger, disgust, sadness, pain) and 2 positive (joy, pleasure). 4 variables were extracted: 1) Antecedents, 2) Body sensations, 3) Behaviours, 4)Mental images. Results. The matching between the two groups of emotions have shown a remarkable difference in the distributions of all variables, as if negative and positive emotions were mutually self-excluding. Conclusions. Implications of our findings are discussed, as raising a basic question: whether it's correct to assume, apart from perceived well-being, that negative emotions always lead to illness and positive ones to health. CORRESPONDING AUTHOR: Prof. L. Sibilia, SIMPS, Italy, [email protected]

S185

P624 NEUROSCIENCE-BASED BEHAVIOURAL MEDICINE. NEW METHODS FOR ASSESSMENT, TREATMENT AND SELF REGULATION L. Sibilia SIMPS, ROMA, Italy Tullio Scrimali, Department of Psychiatry, University of Catania, Italy Introduction. Neuroscience is one of the most powerful emerging forces in the contem porary landscape of the Mind’s Sciences. In this presentation Tullio Scrimali explores how recent advances in neuroscience can be harnessed in practice to improve the effective ness of assessment and treatment with clients affected some psychosomatic disorders such as hypertension. Methods. Two new methods that can be easily applied in a clinical setting, such as Quantitative Electrodermal Activity (QEDA) and Quantitative Heart Activ ity Profile (QHRP) are presented. Results. After introducing and explaining the theoreti cal basis of such a new approach, in Behavioural Medicine, which the Author developed and named “Neuroscience-Based Behavioural Medicine“some experimental data on as sessing (by using QEDA and QRAP) and treating (with Biofeedback) hypertension will be presented. Conclusion. These techniques have become widely available thanks to some new devices and software that can, after a basic training, be easily used in practice by any clinician involved in the field of Behavioural Medicine. CORRESPONDING AUTHOR: Prof. L. Sibilia, SIMPS, Italy, [email protected] P625 TRAJECTORIES OF CHANGE IN MENTAL HEALTH FROM PRE TO POST EARTHQUAKE EXPOSURE R.G. Kuijer, E.M. Marshall University of Canterbury, CHRISTCHURCH, New Zealand Introduction: In 2010 and 2011, Canterbury, New Zealand, was struck by two major earthquakes. This study examined changes in mental health from pre- to post-earthquakes. Because assessing average-level responses may mask significant individual differences in response trajectories, we examined trajectories of change using group based trajectory modelling (GBTM). We also investigated whether age, sex, and trauma impact distinguished between different trajectories. Method: Participants (community sample) completed questionnaires preearthquakes (T1-T3), 1 month post-2010 earthquake (T4: N = 185), 3 (T5: N = 156) and 12 months (T6: N = 142) post-2011 earthquake. Mental health was assessed with the Short Form Health Survey (SF-12) at every time point. Trauma impact was assessed at T4 and T5. Results: Three trajectories of change were found. Most participants entered the study with high (50.4%) or moderately high (32.4%) levels of mental health, remained stable over time, with no changes occurring after the earthquakes. A minority (17.2%) entered the study with lower levels of mental health, experienced a significant drop in mental health after the earthquakes and then recovered to near study entry levels at T6. This “lower mental health”group was younger, more likely to have feared for their life during the 2010 earthquake, and reported more earthquake related hassles post-2011 earthquake. The groups did not differ in terms of objective earthquake impact. Conclusions: Although most people did not experience a decrease in mental health post-earthquakes, a minority of participants with lower levels of mental health at study entry did. Practical implications will be discussed. CORRESPONDING AUTHOR: Dr. R.G. Kuijer, University of Canterbury, New Zealand, [email protected] P626 WHAT IS NEEDED TO LEAVE A VIOLENT PARTNER RELATIONSHIP? C. Mercado, C.A. Santos National University of Mexico, MÉXICO, D.F., Mexico

S186

Introduction: It takes beaten women 10 to 15 years to get rid themselves of the relationship. This study tried to indentify some variables involved in the difficulty of leaving their violent relationship. Method: Twenty seven beaten women were interviewed inside an institution for battered women. Ages were between 24 to 63 years old. Twenty had already abandoned the violent relationship, of which fourteen took the decision of leaving and six were kicked out by their partner. Seven were living with them. A semi-structured interview involving 30 questions was built looking for description of violence, perception and thoughts about it. Results: Twenty two out of 27, 81% had left the couple at some time. The main reasons to remain were: economic dependence and fear of coping with life by themselves. Half of them were threatened, mostly about taking away their children, and committing suicide. The main reasons to finish the relationship were the violence increase against them or their children, infidelity, and tiredness of the situation. Women who remain in the relationship thought they were able to manage the violent situation. The institutional support played an important role in the decision of leaving. Conclusion: Leaving a violent relationship involves strength, fear defeat and social support. It is important to know more about the women who are subject to violent situations in order to be able to develop support programs. CORRESPONDING AUTHOR: Dr. C. Mercado, National University of Mexico, Mexico, [email protected] P627 REDUCING ALCOHOL RELATED VIOLENCE IN A CITY ENTERTAINMENT PRECINCT: IMPACT OF A REGULATORY INTERVENTION J.H. Wiggers1, L. Wolfenden1, J. Tindall2, K. Gillham2, C. Lecathelinais2 1 University of Newcastle, WALLSEND, Australia 2 Hunter New England Local Health District, WALLSEND, Australia

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

P628 INJURIES, COPING STRATEGIES AND PSYCHOPATHOLOGY IN A WAR HANDICAPPED VETERANS SAMPLE A. Maia1, José Mendes2, Luísa Sales2, Pedro Araujo2, Aida Dias2, Rafaela Lopes2 1 University of Minho, BRAGA, Portugal 2 Centro de Estudos Sociais, COIMBRA, Portugal Introduction: The purpose of this study is to describe coping strategies used by handicapped veterans and their relation to PTSD Symptoms and PTSD. We also intend to analyze if percentage of recognized disability and length of hospitalization have different impact on current PTSD symptomatology and Psychopathology. Methods: Members of the Portuguese Handicapped of the Armed Forces Association (ADFA) were invited to participate and report coping strategies they used to deal with the injuries and the traumatic situations during their Military Experience in the war in Africa. We measured coping strategies (Brief COPE), PTSD symptoms (PCL), global psychopathology (BSI) and global satisfaction in various areas of life. Results: Participants have a high percentage of injuries, with a mean of 53,98 % of disability (SD=26,57) and long periods of hospitalization after war (M=22,90 months, SD=19,33). Preliminary analysis showed negative correlations between degree of disability, PTSD and psychopathology. Also veterans who had more hospitalization time reported less PTSD and less Psychopathology. Avoidant emotional coping strategies explains 28% of psychopathology and 5% is explain by percentage of disability (negative beta). 25% of PTSD symptomatology is also explained by avoidant emotional coping strategies. Conclusion: Handicapped Veterans with more injuries have also longer periods of hospitalization. Qualitative studies suggests that during hospitalization periods the victims built a sense of identity, which can explain a bigger perception of support and consequently higher levels of adaptation. Those preliminary results show that the use of avoidant emotional coping strategies predicts higher levels of PTSD symptomatology and Psychopathology. CORRESPONDING AUTHOR: Prof A. Maia, University of Minho, Portugal, [email protected]

Introduction. Licensed premises in entertainment precincts are associated with higher levels of alcohol-related violence. In 2008, an intervention involving additional licensing conditions on licensed premises was introduced in the entertainment precinct in the city of Newcastle, Australia. The conditions included reduced trading hours, curfews and additional responsible service of alcohol strategies.The aim of this study is to describe the impact of the conditions on night-time assaults, assaultrelated injury ED presentations, community acceptability and the nighttime economy. Methods:Police and hospital emergency department data were analysed to assess change in police attended night-time nondomestic assaults and assault-related injury hospital ED presentations between 2008 and 2013. A random household survey assessed community acceptability, whilst change in the number of licensed premises was used as an indicator of impact on the night-time economy.Results In the 5 years following the additional licensing conditions a 33% reduction was found in the number of night-time non-domestic assaults, a reduction that was greater than elsewhere in the state. In addition a 26% reduction in assault-related injury ED presentations was observed. In excess of 82% of community members supported the additional conditions. The number of licensed premises in the precinct increased by 25%. Conclusion. The finding of a large and sustained positive effect on community attitudes and alcohol related violence occurring at the same time as an increase in the scale of the night time economy provides a unique basis for policy makers to improve the safety, amenity and vibrancy of city entertainment precincts.

Introduction: A series of natural disasters happened in several locations in Indonesia. One of them happened in Pangalengan district, West Java. A tectonic earthquake as strong as 7,3 richter scale hit the area on Sept 2th in 2009, which caused significant impacts on both mental and physical conditions of the people. The objective of this research is to measure the change of the psychological and PTSD symptoms among survivor housewives before and after the treatment of EMDR. Methods: This research was designed using one group pretest-posttest design. Four respondents were involved in this research. Pretest assessment was taken to get initial PTSD symptoms of the subjects. Respondents in this research were four housewives who survived the earthquake . Before and after therapy assessment was taken to find change of PTSD symptoms of the subjects by using PSS (PTSD Symptom Scale) and CAPS (Clinician Administered PTSD Scale). EMDR treatment was taken in four repetitive sessions. Each session was done in 60 - 90 minutes. Results: All of subjects had reduced PTSD symptoms. After EMDR treatment , Subject A, B, C had changed from moderate to mild category of PTSD symptoms and subject D hadn’t changed moderate category but had changed on PSS score.

CORRESPONDING AUTHOR: Professor J.H. Wiggers, University of Newcastle, Australia, [email protected]

CORRESPONDING AUTHOR: Eka Susanty, Jenderal Achmad Yani university, Indonesia, [email protected]

P629 EYE MOVEMENT DESENSITIZATION AND REPROCESSING THERAPY FOR THE HOUSEWIVES SURVIVOR EARTHQUAKE Eka Susanty Jenderal Achmad Yani university, BANDUNG, Indonesia

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

P630 MENTAL HEALTH STATUS AMONG FEMALE 'VISUAL DISPLAY TERMINALS' OPERATORS A.E. Fahim Faculty of Medicine, Suez Canal University, ISMAILIA, Egypt Introduction: The use of Visual display terminals (VDTs) can have psychological and physiological effects. The present study assessed the effect of work conditions and personal factors on the mental health status of visual display terminals' (VDTs) female operators. Methods: The mental health status of 127 female VDT operators at the administrative office of the university and the IT centers affiliated with the university was investigated using the 12-item General Health Questionnaire (GHQ-12) - validated Arabic version. Other workplace and personal factors were investigated. Results and Conclusion: Seventy one percent of VDT operators were classified as GHQ-12 high score group (≤4). Among the studied subjects, 35.7% of those with high GHQ score had rest during VDT work while 46.5% of low GHQ score received breaks during VDT work. The subjects who reported physical symptoms (eyestrain/musculoskeletal pain) were 29.6% and 55.7%, among low and high GHQ score respectively. Among high GHQ score group, 62.7% worked with VDTs more than 4 hours daily, while only 33.7% of low score had the same time of VDT work exposure with statistically significant difference between the two groups (p<0.05). Logistic regression analysis revealed that, not receiving breaks during VDT use, and the presence of physical symptoms were significantly associated with poor mental health status (high GHQ-12 scores). Also using VDTs for more than 4 h/day were also associated with high GHQ scores (p<0.05). The management of physical health as well as work conditions is important for good mental health status among VDTs operators. CORRESPONDING AUTHOR: Asst. Prof. A.E. Fahim, Faculty of Medicine, Suez Canal University, Egypt, [email protected] P631 ORGANIZATIONAL JUSTICE AND CORONARY RISK FACTORS IN JAPANESE WORKERS: J-HOPE A. Inoue1, N. Kawakami2, H. Eguchi3, K. Miyaki4, A. Tsutsumi3 1 University of Occupational and Environmental Health, KITAKYUSHU, Japan 2 The University of Tokyo, TOKYO, Japan 3 Kitasato University School of Medicine, SAGAMIHARA, Japan 4 National Center for Global Health and Medicine, TOKYO, Japan Introduction: Growing evidence has shown that lack of organizational justice (i.e., procedural justice and interactional justice) is associated with coronary heart disease (CHD) while biological mechanisms underlying the association have not yet been fully clarified. The purpose of the present study was to investigate the association of organizational justice with physiological CHD risk factors (i.e., blood pressure, high-density lipoprotein [HDL] cholesterol, low-density lipoprotein [LDL] cholesterol, and triglyceride) in Japanese employees. Methods: A total of 3,617 male and 910 female employees from two manufacturing companies in Japan completed self-administered questionnaires measuring organizational justice, demographic characteristics, and lifestyle factors. They also received health check-up and blood pressure and serum lipids were measured. Multiple logistic regression analyses and linear trend tests were conducted. Results: Among male employees, multiple logistic regression analyses and linear trend tests showed significant association of low procedural justice (OR=1.29 [95% CI: 1.08-1.55], p for trend=0.006) and low interactional justice (OR=1.37 [95% CI: 1.07-1.74], p for trend=0.022) with high triglyceride (≥150 [mg/dL]), after adjusting for demographic characteristics and lifestyle factors. Among female employees, linear

S187

trend tests showed significant dose-response relationship between low interactional justice and high LDL cholesterol (≥140 [mg/dL]) (p for trend=0.027) while multiple logistic regression analyses showed insignificant odds ratios of high LDL cholesterol among moderate or low interactional justice group. Neither procedural justice nor interactional justice was associated with blood pressure or HDL cholesterol. Conclusion: Organizational justice may be an important psychosocial factor associated with increased triglyceride at least among Japanese male employees. CORRESPONDING AUTHOR: A. Inoue, University of Occupational and Environmental Health, Japan, [email protected] P632 ORGANIZATIONAL JUSTICE AND JOB PERFORMANCE IN JAPANESE WORKERS: MEDIATION EFFECT OF WORK ENGAGEMENT Y.N. Nakagawa1, A. Inoue1, N. Kawakami2, K. Tsuno3, K. Tomioka4, M. Nakanishi5, K. Mafune1, H. Hiro1 1 University of Occupational and Environmental Health, Japan, KITAKYUSYU, Japan 2 The University of Tokyo, TOKYO, Japan 3 Wakayama Medical University, WAKAYAMA, Japan 4 Nara Medical University, KASHIHARA, Japan 5 Nakanishi Healthcare Office, HIGASHIOSAKA, Japan Objective: A previous study reported that organizational justice (i.e., procedural justice and interactional justice) was associated with increased work engagement (Inoue et al., 2010). Furthermore, another previous study reported that work engagement enhances job performance (Shimazu et al., 2012). However, no study tested the mediation effect of work engagement linking organizational justice with job performance. The purpose of the present study was to investigate whether work engagement mediates between organizational justice and job performance in Japanese employees. Methods: A total of 466 men and 756 women from a manufacturing company in Japan were surveyed using a self-administered questionnaire including the Organizational Justice Questionnaire, Utrecht Work Engagement Scale, World Health Organization Heath and Work Performance Questionnaire, and demographic characteristics. Multiple mediation analyses with the bootstrap technique (5,000 bootstrap samples) were conducted. Results: In the bivariate analysis, after adjusting for demographic characteristics, procedural justice and interactional justice were significantly and positively associated with job performance. In the mediation analysis, work engagement significantly mediated between procedural justice or interactional justice and job performance, while the associations of procedural justice and interactional justice with job performance were no longer significant. When we included each component of work engagement (i.e., vigor, dedication, and absorption) as a mediator, dedication significantly mediated between procedural justice or interactional justice and job performance, whereas the mediation effect of vigor or absorption was not significant. Conclusions: The effect of organizational justice on job performance seems to be mediated by dedication component of work engagement to a large extent. CORRESPONDING AUTHOR: Y.N. Nakagawa, University of Occupational and Environmental Health, Japan, Japan, [email protected] P634 DO CHANGES IN JOB DEMANDS MODIFY ASSOCIATIONS BETWEEN CHANGES IN TESTOSTERONE AND ANDROPAUSE SYMPTOMS? K. Hirokawa1, T. Taniguchi2, Y. Fujii2, J. Takaki3, A. Tsutsumi4

S188 1

Baika Women's University, IBARAKI, Japan Okayama Prefectural University, SOJA, Japan 3 Mie University Graduate School of Medicine, TSU, Japan 4 Kitasato University School of Medicine, SAGAMIHARA, Japan 2

Introduction: In a previous cross-sectional study, we showed that job stress modified the association between testosterone deficiency and symptoms of andropause. Aim: In this study we investigated whether changes in job demands modified associations between changes in testosterone and symptoms of andropause in Japanese male workers. Method: In 2007, we conducted a baseline survey of job demands and lifestyle factors and administered the Aging Males’ Symptoms scale and drew blood samples to assess levels of testosterone. Of 256 men (mean age = 50.4 years, SD = 9.1) who completed all requirements in 2007, 102 (52.8 years, SD = 7.3) were again evaluated in 2009. We calculated changes in measured variables over the two years. Results: Testosterone levels increased significantly, whereas job demands and physical symptoms decreased significantly over the 2 years. Changes in testosterone were negatively associated with changes in the total andropause and sexual symptoms (Standardized-beta = '0.27 and -0.34, p < 0.05, respectively) after adjusting for confounding factors. Changes in job demands were positively associated with changes in somatic and sexual symptoms (Standardized-beta = 0.21 and 0.19, p < 0.05, respectively). When stratified by changes in job demands, changes in testosterone showed a negative association with changes in sexual symptoms in the reduced job demands (Standardized-beta = -0.30, p < 0.05). Conclusions:Symptoms of andropause may be affected by changes in levels of testosterone and changes in job demands. Changes in job demands may modify the association between change in testosterone and symptoms of andropause. CORRESPONDING AUTHOR: Ms. K. Hirokawa, Baika Women's University, Japan, [email protected] P635 CONCENTRATION REQUIREMENTS MODIFY THE EFFECT OF O F F I C E T Y P E O N I N D I C AT O R S O F H E A LT H A N D PERFORMANCE A. Seddigh1, E.B. Berntson2, C.B.D. Bodin Danielsson1, H.W. Westerlund1 1 Stress research institute, STOCKHOLM, Sweden 2 Department of Psychology, STOCKHOLM, Sweden Studies of the impact on health and productivity of different types of office environments usually compare cell offices with open-plan offices. This paper investigates the interaction between need for concentration on the job and six more specific office types (cell office, shared-room office, small open-plan office, medium-sized open-plan office, large open-plan office and flex office) in relation to indicators of health and productivity. A 2x6 between-subject MANCOVA was conducted on the dependent variables distraction, cognitive stress, emotional exhaustion, depersonalization, personal efficiency and general health while controlling for age, educational level, gender and sector of the labor market. The independent variables were need for concentration (two categories) and office type (six categories). 1241 employees from five organizations participated in the study. Cell offices were associated with lower reported levels of distraction and cognitive stress, and flex offices with lower distraction, among the employees compared with all other open-plan office types. However, there were no significant differences in the outcome variables between different types of open-plan offices. There was an interaction between office type and the need for concentration for the job; employees in the high need for concentration group reported more distraction in all office types except in cell offices and also more cognitive stress in all office types except cell offices and flex offices. This study suggests the presence of negative health and productivity aspects in individuals working in open-plan offices. This

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

seems especially true for those who have tasks that require a high need for concentration. CORRESPONDING AUTHOR: Mr A. Seddigh, Stress research institute, Sweden, [email protected] P636 WORK AND COMMON MENTAL DISORDERS: ARE EXPECTAN CIES MORE IMPORTANT THAN WORK SATISFACTION AND WORK STRAIN? T.L. Johnsen, T. H. Tveito, A. Indahl Vestfold Hospital Trust, TØNSBERG, Norway Background.Common mental disorders (CMD) are one of the most frequent reasons for sick leave and disability pensions. Thus, it is important to explore early predictors for CMD to be able to develop effective interventions. According to The Cognitive Activation Theory of Stress learned response outcome expectancies are important contributors to health. Individual differences in the expectancy and ability to cope with workplace and general life demands may be important for how the work strain influence the health of the employees. Thus, high levels of positive response outcome expectancies, and low levels of negative response outcome expectancies and no response outcome expectancies, might make it more likely for employees to manage the consequences of an adverse work environment. Objective.Investigate the relative effect of work satisfaction, work strain and response outcome expectancies on CMD. Method.A cross-sectional study with 1,746 municipal employees (mean age 44.2 SD=11.5, 81% female). Multiple logistic regression with anxiety and depression as outcome were conducted. Job satisfaction, work strain, and response outcome expectancies were independent variables. Results.1721 employees was included in the analyses. Fifteen percent reported anxiety during the last month, 24% reported depression, and 11% reported both anxiety and depression. Of the employees with anxiety sixty-seven percent reported a little, 26% some, and 7% severe anxiety. Seventy percent of the employees with depression reported a little, 21% some, and 9% severe depression. No and negative response outcome expectancies and perceived high mental work strain were associated with anxiety and depression. The associations were small, although statistically significant. CORRESPONDING AUTHOR: PhD-candidate T.L. Johnsen, Vestfold Hospital Trust, Norway, [email protected] P637 SELF-RATED HEALTH PREDICTS DISABILITY RETIREMENT - A TEN-YEAR PROSPECTIVE STUDY IN GERMAN WORKING POPULATION J. Li, A. Angerer University of Düsseldorf, DÜSSELDORF, Germany Introduction: Several studies have indicated that poor health is associated with disability retirement, the evidence from Germany, however, is still sparse. The aim of this study was to examine the prospective associations between self-rated health and disability retirement in Germany. Method: 10035 workers aged from 18-64 without any disability (5499 men and 4536 women) from a national representative sample of Germany were followed up from 2001 to 2011. At baseline, the participants answered a questionnaire regarding their self-rated health, demographic, socioeconomic, and behavioral information. The disability retirement was based on self-reports during follow-up. Cox regression was used to calculate hazard ratios (HRs) and 95% confidence intervals (CIs). Results: Over 10 years, 189 workers (1.88%) were identified to receive disability pension. After adjustment for potential confounders, workers with poor self-rated health at baseline had a more than three times higher risk (HR = 3.21, 95% CI = 2.31-4.46) for disability retirement. The

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

similar pattern was observed in both men and women (in men, HR = 3.08, 95% CI = 1.98-4.79; in women HR = 3.31, 95% CI = 2.01-5.45). Conclusions: Self-rated health is a strong predictor of disability retirement in German working population. Our findings suggest that improving health status would be helpful to prevent disability retirement. CORRESPONDING AUTHOR: J. Li, University of Düsseldorf, Germany, [email protected] P638 THE TIME IS READY TO STARTING TAKING CARE OF MEDICAL DOCTORS! J.K. Sluiter, M.H.W. Frings-Dresen Academic Medical Center, AMSTERDAM, Nederland Introduction: Being a medical doctor is known to with several risk factors and high prevalence of different health complaints that may interfere with work functioning. Workers’ Health Surveillance (WHS) is the preventative occupational health strategy that may be offered periodically to specific worker groups in order to monitor and impact work-related health and workability. This was recently shown to provide a good business case in other health care workers and should be started for medical doctors as target. What happens when the scientific development is finished? Methods: After developing a job-specific WHS for the profession of medical doctors and a successful feasibility study in one academic medical center, implementation was continued together with an occupational health service. Adapting to the Dutch WHS guideline for occupational physicians (2013), preparations for use were continued. Results: The job-specific WHS protocol was the starting point. Building an e-system by an organization was guided by researchers and occupational health service professionals. It allows workers, physician assistant and occupational physician to add data in a save platform. It also includes a professional supporting part to guide providing advices and the start of interventions when decreased workability or impaired health requirements is detected. Professionals were trained to apply the WHS. A database is coupled to the system and allows historical results to be retrieved and used to better guide the doctors. Conclusions: Preventative occupational health guidance with medical doctors as target population is globally new and now prepared and ready for use in practice in The Netherlands. CORRESPONDING AUTHOR: Professor J.K. Sluiter, Academic Medical Center, Nederland, [email protected] P639 RELIABILITY, VALIDITY, AND SIGNIFICANCE OF ASSESSMENT OF SENSE OF CONTRIBUTION IN THE WORKPLACE J. Takaki1, T. Taniguchi2, Y. Fujii2 1 Mie University Graduate School of Medicine, TSU, Japan 2 Okayama Prefectural University, SOJA, Japan Introduction: The purpose of this study was to assess the validity and reliability of the Sense of Contribution Scale (SCS), a newly developed, 7-item questionnaire to measure sense of contribution in the workplace. Methods: Workers at 272 organizations answered questionnaires that included the SCS. Because of non-participation or missing data, the number of subjects included in the analyses for internal consistency and validity varied from 1,675 to 2,462 (response rates 54.6%-80.2%). Fiftyfour workers were included in the analysis of test-retest reliability (response rate, 77.1%). Results: The SCS showed high internal consistency (Cronbach’s α coefficients in men and women were 0.85 and 0.86, respectively) and test-retest reliability (intraclass correlation coefficient = 0.91). Significant (p < 0.001), positive, moderate correlations were found between the SCS score and scores for organization-based selfesteem and work engagement in both genders, which support the SCS’s convergent and discriminant validity. The criterion validity of the SCS

S189

was supported by the finding that in both genders, the SCS scores were significantly (p < 0.05) and inversely associated with psychological distress and sleep disturbance in crude and in multivariable analyses that adjusted for demographics, organization-based self-esteem, work engagement, effort-reward ratio, workplace bullying, and procedural and interactional justice. Conclusions: The SCS is a psychometrically satisfactory measure of sense of contribution in the workplace. The SCS provides a new and useful instrument to measure sense of contribution, which is independently associated with mental health in workers, for studies in organizational science, occupational health psychology and occupational medicine. CORRESPONDING AUTHOR: J. Takaki Mie, University Graduate School of Medicine, Japan, [email protected] P640 ASSOCIATIONS OF LEADERSHIP STYLES WITH SELF-RATED HEALTH - FINDINGS FROM A CROSS-SECTIONAL STUDY B. Schmidt1, M. Seegel1, R.M. Herr1, A. Loerbroks2, J.A. Bosch3, M Schneider1 1 Heidelberg University, MANNHEIM, Germany 2 University of Düsseldorf, Düsseldorf, Germany 3 University of Amsterdam, Amsterdam, Nederland Introduction: Recent epidemiological research has identified supportive leaders’ behavior as a potential determinant of employees’ health. As there are various types of leadership style (LS) it is, however, unclear, which specific LS is particularly detrimental to employees’ health. This study aims to explore associations between several leadership styles (LSs) and self-rated health (SRH). Methods: Cross-sectional data from an occupational study of 1,632 individuals (response rate 40.8%; 51.6% male; mean age = 41.3 years with SD = 9.4) were used. Validated questionnaires measured SRH and several LSs: five dimensions of transformational LS (i.e., visionary, teamoriented, intellectual stimulating, role model and considering), transactional LS (i.e., conditional reward), non-listening LS, authentic LS (i.e., relational transparency) and fair LS (i.e., interactional justice). Associations were explored by logistic regressions modeling odds ratios (ORs) and 95% confidence intervals (95% CI) for good vs. poor SRH, mutually adjusting for all LSs and various confounders. Results: Controlling for the other LSs, only two LSs showed significant negative associations with SRH, these were, fair LS (interactional justice; OR = 0.58; 95% CI = 0.42-0.79) and conditional reward (transactional LS; OR = 0.79; 95% CI = 0.62-1.00). Conclusion: Some LSs might be more beneficial for employees’ health than others. In particular fairness and to a lesser extent transactional LS (praise and appreciation) seem to be relevant to health in the workplace. These LSs should be targeted in future longitudinal research. CORRESPONDING AUTHOR: Dr. B. Schmidt, Heidelberg University, Germany, [email protected] P641 PSYCHOSOCIAL MECHANISMS OF PSYCHOLOGICAL HEALTH DISPARITY IN JAPANESE WORKERS: A TWO-WAVE PANEL SURVEY A. Shimazu1, N. Kawakami1, A. Inoue2, S. Kurioka2, K. Miyaki3, M. Takahashi4, A. Tsutsumi5 1 The University of Tokyo, TOKYO, Japan 2 University of Occupational and Environmental Health, KITA-KYUSYU, Japan 3 National Center for Global Health and Medicine, TOKYO, Japan 4 National Institute of Occupational Safety and Health, KAWASAKI, Japan 5 Kitasato University, SAGAMIHARA, Japan

S190

Introduction: Recent epidemiological research has shown that people with higher socioeconomic status (SES) (e.g., educational attainment) have better psychological health than those with lower SES. However, the psychosocial mechanisms of underlying this relationship remain unclear. In addition, most previous studies focused on negative outcomes rather than positive ones. To fill this gap, the current study examines the mediating effects of job demands and job resources in the relationship of educational attainment with psychological distress and work engagement. Methods: The hypothesized model was tested using 2-wave panel data of 7,549 Japanese employees from 12 workplaces with a 1-year time lag per wave (J-HOPE_20130819 version). Results: Structural equation modeling revealed that (1) educational attainment had both a positive and a negative relationship to psychological distress (T2) through job demands and job resources (T1), respectively; (2) it had a positive relationship to work engagement (T2) through job resources and work engagement (T1); and (3) standardized total effects of educational attainment were -.006 (p >.05) for psychological distress and .093 (p <.01) for work engagement, respectively. Conclusion: The effects of educational attainment were weak and its underlying mechanisms were different by outcomes. The effect on psychological distress (T2) was mediated by job demands and job resources (T1), whereas that on work engagement (T2) was mediated by job resources and work engagement (T1). CORRESPONDING AUTHOR: Dr. A. Shimazu, The University of Tokyo, Japan, [email protected] P642 ASSOCIATIONS OF WORK DEMANDS AND RESOURCES WITH METABOLIC SYNDROME - FINDINGS FROM A CROSSSECTIONAL STUDY I. Getzendörfer, R.M. Herr, M.N. Jarczok, J.E. Fischer, B. Schmidt Heidelberg University, MANNHEIM, Germany Introduction: The prevalence of metabolic syndrome (MetS) increases steadily in industrial societies. Work stress, amongst others, is known as a modifiable risk factor. Work stress can be conceptualized in several ways, like in terms of demands (e.g., quantitative, physical) and resources (e.g., social support, control) at work. It is unclear however, which of these resources and demands are associated with higher or lower risk for MetS, which, therefore, was the aim of this study. Methods: Cross-sectional data of a healthy occupational sample (n=4154) was used. MetS was assessed according to latest America Heart Association criteria. Work demands (quantitative, physical, cognitive, emotional), resources (social support, operational control, skill stimulation) at work were measured by validated questionnaires. Logistic regressions estimated the associations between demands and resources at work and MetS. The analysis was adjusted for age, sex, smoking, alcohol consumption, physical exercise, job position and mutually for the other demands and resources. Results: In the fully adjusted model higher physical demands (OR = 1.21; 95%CI = 1.11-1.32) were positively, social support (OR = 0.89; 95%CI = 0.80-0.98) negatively associated with MetS while all other factors did not pass significance of below 0.05. Conclusion: Reducing physical demands and enhancing social support might provide an angle for early prevention of Cardiovascular Disease and Diabetes Mellitus Type II, thus protecting a large segment of the adult population. However more research regarding specific work stress factors and intervention studies are needed. CORRESPONDING AUTHOR: I. Getzendörfer, Heidelberg University, Germany, [email protected]

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

P643 STATUS OF THE IMPLEMENTATION OF HEAT STROKE MEASURES AT THE FUKUSHIMA DAIICHI NUCLEAR POWER PLANT T.S. Tateishi1, K. T. Kubo1, R.O. Okazaki2, S. K. Suzuki1, O. S. Okahara1, M. K. Mori1 1 University of Occupational and Environmental Health, Japan, KITAKYUSHU, Japan 2 Department of Radiation Biology and Health, KITAKYUSHU, Japan Introduction: At the Fukushima Daiichi Nuclear Power Plant, following the severe accident triggered by a tsunami in March 2011, Tokyo Electric Power Co. (TEPCO) and conductors have cooperatively implemented measures to prevent the heat stroke of workers. Plans to implement heat stroke measures presented by TEPCO are reviewed every year, and the number of people developing heat stroke has decreased (42 people in 2011, 10 people in 2012, and 12 people in 2013). The University of Occupational and Environmental Health in Japan provided an action checklist for the prevention of heat stroke for companies with no professionals. The checklist was distributed through TEPCO. The present study investigates the status of implementation of heat-stroke measures through the cooperation of each company with workers at the Fukushima Daiichi Nuclear Power Plant. Method: We collected information through a questionnaire. Questions related to whether items in the action checklist were implemented. Results: It was found that activities directly related to the protection of workers are being implemented; companies implement at least 70% of measures. However, companies are only implementing 25%-54.2% of measures continuously at the power plant. Conclusions: At the Fukushima Daiichi Nuclear Power Plant, about onethird of those in charge of health care have been replaced. Therefore, although the number of workers suffering heat stroke has decreased, there is a concern that the number will increase again. Continuous monitoring and education are required in the future. CORRESPONDING AUTHOR: T.S. Tateishi, University of Occupational and Environmental Health, Japan Japan, [email protected] P644 Physical & Psychosocial Stressors, Buffers & Resilience Affecting Health of Female Foreign Domestic Workers (FDW) Y.L. Wong1, S. Thambiah1, D. Spitzer2, S.S.H. Nudin3 1 University of Malaya, KUALA LUMPUR, Malaysia 2 Institute of Population Health, University of Ottawa, OTTAWA, Canada 3 Institute for Health Behavoral Research, Ministry of Health, KUALA LUMPUR, Malaysia Introduction: Malaysia is host to thousands of FDW from low income countries in Asia. Often confined to employers' private homes and under latter's control, FDW tend to have limited access to health determinants like nutrition, social support, and healthcare. Yet, they are vulnerable to multiple stressors arising from their work and migrant status. This paper explores the physical and psychosocial stressors; buffers and resilience affecting FDW's health. Method: Data on out-migration process, work conditions, health and illness experiences were collected among 54 FDW (30 Indonesian and 24 Filipino) in Malaysia. Results:Stressors included long working hours (76-90 hours/week), lack of separation between work and leisure, constrains/surveillance, social hierarchy, financial/emotional stress related to own children/families. They reported health problems like headaches, insomnia, gastric pains, menstrual disorders, and hypertension. Prayers and social support were the main buffers. Many coped through selfcare: trying to not think too much, practicing patience, refraining from anger or unpleasant thoughts, and use of traditional medicine.

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

Conclusion: The interplay of physical and psychosocial stressors, buffers and resilience affect FDW health. Coping and resilience borne out of the FDW's vulnerable conditions ought to be considered when addressing migrant health needs. CORRESPONDING AUTHOR: Assoc Prof Y.L. Wong, University of Malaya, Malaysia, [email protected] P645 DATA ANALYSIS OF THE HEALTH EFFECTS OF MAJOR ORGANIZATIONAL CHANGES IN A HOSPITAL S. Okahara University of Occupational & Environmental Health, Japan, KITAKYUSHU, Japan Introduction: Recent systematic reviews and observational research suggest that there is growing concern about the potential deleterious consequences of organizational changes on employee health. This analysis aimed to describe an association between organizational changes and employee mental health in hospitals. Methods: We analyzed personnel data from 2001 to 2012 concerning administrative leave and history of major organizational changes in a University Hospital. We confirmed the causes of mental health leave by consulting counseling records. Job type was categorized as medical doctor, nurse, engineer, and clerical staff. Results: We found 218 cases of mental health leave during 2001-2012. The average number of leave days was 206. Mental health leave was highest in nurses (n=149). There were two exponential increases in nurse leave in 2007 and 2012. Nurse mental health may have been affected by implementation of an electronic clinical record system in 2005, an increase in the number of new nurses in 2007 owing to a change of nursing system, and opening of a new hospital in 2011. However, the number decreased after 2008-2011. Organizational measures, such as group education, occupational self-improvement and return to work program were established for nurses in 2008, may have had positive effects on nurse mental health. Conclusions: The results suggested that major organizational changes had a negative effect on employee mental health in this hospital. If major organizational change is planned beforehand, it should be possible to estimate potential adverse effects on employee health and take appropriate organizational measures. CORRESPONDING AUTHOR: S. Okahara, University of Occupational & Environmental Health, Japan, [email protected] P646 EXPLORING CAUSAL ILLNESS ATTRIBUTIONS AMONG WORKERS WITH COMMON MENTAL DISORDERS I. B. O. Blø Olsen, S. Øverland, S. Reme, C. Løvvik Uni Research, BERGEN, Norway

S191

specific work-related attributions; work stress, leadership, unemployment, job dissatisfaction, work conflict, social work environment, job insecurity and change, workplace bullying, and physical strain. Conclusions: These findings point to several work-related risk factors for people suffering from CMDs and work disability. Work participation and mental health could for this population be improved through work coping and individual adjustments at the workplace. We suggest that workfocused individual therapy and workplace interventions that specifically target work-related causal illness attributions would be useful to include in this process. In CMDs, workplace interventions should aim at reducing work stress and enhancing coping at work by promoting social support, job satisfaction, employee-employer relations and organizational trust. CORRESPONDING AUTHOR: cand.psychol I. B. Olsen, Uni Research, Norway, [email protected] P647 QUALITY OF LIFE OF WORKING AND NON-WORKING INFORMAL CAREGIVERS: WHO IS BETTER OF AND WHY? M. Oldenkamp 1 , M. Hagedoorn 1 , R.P. Stolk 1 , R.P.M. Wittek 2 , U. Bultmann1, N. Smidt1 1 University Medical Center Groningen, GRONINGEN, Nederland 2 University of Groningen, GRONINGEN, Nederland Introduction: The combination of informal care and paid work is becoming increasingly common. Currently, in the Netherlands more than 70% of all caregivers aged 65 or younger combines informal care with work, and an increase is expected. The quality of life of informal caregivers plays a central role in their willingness and ability to provide and continue this important care. Our aim is to study whether fulltime working, parttime working, and non-working informal caregivers differ in their quality of life, and what the underlying factors of these differences are. Method: The study sample consists of informal caregivers in the LifeLines cohort study, who participate in an additional informal care questionnaire. Within 7 months of recruitment, more than 600 informal caregivers are included in the study. We aim to include 1000 informal caregivers. The questionnaire consists of questions about quality of life (EuroQol-5D), caregiver burden (Caregiver Reaction Assessment (CRA)), work, support, and the informal care situation (i.e. relation with the care receiver, hours and types of care). To compare the three groups of informal caregivers t-tests for more than 2 independent samples (continuous outcomes) and chi-square tests (categorical outcomes) are performed. Results: Results will be presented at the congress. Conclusions: Conclusions will show whether working and non-working informal caregivers should be approached differently when aiming to improve their quality of life. CORRESPONDING AUTHOR: M. Oldenkamp, University Medical Center Groningen, Nederland, [email protected] Innovations in emotion regulation and health

Introduction: Common Mental Disorders (CMDs) are major causes of sickness absence and disability. Disability prevention requires knowledge of causal mechanisms and risk factors. In this study we sought to examine causal illness attributions in a population struggling with work participation due to CMDs. Methods: Causal attributions of CMDs from 1122 participants were measured by an open-ended item of the Brief Illness Perception Questionnaire, asking participants to list the three most important causes of their illness. Responses were subjected to thematic content analysis, and organized in mutually exclusive causal attribution categories. Analysis was performed in two stages; a preliminary analysis to identify workrelated causal attributions, and a selective analysis to explore the workrelated attributions in more detail. Results: Work-related attributions were the most frequently reported causal illness attributions. The selective analysis identified the following

S648 INNOVATIONS IN EMOTION REGULATION AND HEALTH L. Cameron1, G. Ozakinci2, M.J. Zawadzki3, D. Wiebe1, L. Cameron1 1 University of California, Merced, MERCED, United States of America 2 University of St Andrews, STANDREWS, SCOTLAND, United Kingdom 3 The Pennsylvania State University, UNIVERSITY PARK, United States of America Emotion regulation and psychological health are intricately related to each other. Techniques that are typically used to regulate emotions include cognitive reappraisal, attentional deployment (e.g., avoidance), and emotional modulation (e.g., humour) have been related to different psychological outcomes. In this symposium, we will present findings using various theoretical frameworks that exemplify the relationship

S192

between emotion regulation and health. First, Zawadzki will present two studies comparing the effects of rumination with other emotion regulation strategies on health and a possible mechanism for rumination’s negative effects. Cameron will present findings from a longitudinal study on the predictive value of emotion regulation strategies for worry of cancer recurrence among cancer survivors using a common sense model to guide the development of illness-specific emotion regulation measures. Ozakinci will present findings from qualitative studies conducted with cancer survivors on the strategies they use to regulate fears about cancer coming back and also the impact of the communication with health care professionals on managing their emotions. Objectives: Understand the role of emotion regulation in health Overview of techniques used for emotion regulation Discussion of how interventions can assist with emotion regulation. Chair: Linda Cameron, University of California, Merced, USA Presenter 1: Matthew J Zawadzki, The Pennsylvania State University, USA Presenter 2: Gozde Ozakinci, University of St Andrews, Scotland UK Presenter 3: Linda Cameron Discussant: Deborah Wiebe, University of California, Merced, USA CORRESPONDING AUTHOR: Prof L. Cameron, University of California, Merced United States of America, [email protected] S649 EXAMINING THE IMPORTANCE OF RUMINATION AS A MALADAPTIVE EMOTION REGULATION STRATEGY M.J. Zawadzki, Joshua M. Smyth, William Gerin The Pennsylvania State University, UNIVERSITY PARK, United States of America Introduction: Research has consistently demonstrated that emotion regulation (ER) impacts health, with an extensive focus of this work on reappraisal and suppression. Yet a recent meta-analysis suggested that ruminative ER strategies may have the strongest relationship with health, however direct comparisons between different ER strategies and health are lacking. Furthermore, it is unclear why rumination may be particularly detrimental. The purpose of the following studies was first to compare the effects of rumination with other ER strategies on health. Second, it explored one possible explanation for rumination’s negative effects - that rumination prevents individuals from benefitting from recuperative, stress-relieving activities. Method: Study 1 enrolled a college sample (n = 218) and measured trait rumination, reappraisal, suppression, and proactive coping as ER strategies, and depression, anxiety, hostility, and sleep quality as health indicators. Study 2 enrolled a community sample (n = 38) and measured rumination, engagement in absorptive leisure activity, and 24-hour ambulatory blood pressure. Results: Study 1 found that rumination was strongly related to all health indicators, whereas inconsistent and small relationships were observed for the other ER strategies. Study 2 found that although engaging in absorbing leisure activity had a beneficial effect on daytime and nighttime blood pressure, trait ruminators did not experience this benefit. Discussion: The results from both studies suggest that among ER strategies, rumination may have particular importance for determining health. Moreover, they suggest rumination as an untapped and innovative way to improve intervention and treatment models, which often target reappraisal and suppression. CORRESPONDING AUTHOR: Dr. M.J. Zawadzki, The Pennsylvania State University, United States of America, [email protected] S650 REGULATING FEARS OF RECURRENCE G. Ozakinci1, E. Watson1, M. Sharpe2, G. Humphris1 1 University of St Andrews, ST ANDREWS, SCOTLAND, United Kingdom 2 University of Oxford, OXFORD, United Kingdom

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

Introduction: Following the end of treatment, cancer patients are often faced with the task of managing their emotions about uncertainty of future cancer risk. In this study, we examined f fears of recurrence (FoR) in people who have been treated for cancer, the methods they use for regulating their fears, and the role of health care professionals in emotion regulation guided by Leventhal’s Self-regulation model. Method: We conducted 6 focus group discussions (3 for breast and 3 for colorectal) with 18 people classified having either low, moderate, or high FoR (12 women; aged 49 and over). Participants were asked to talk about triggers of recurrence fears and their methods of emotion regulation. A constant comparative technique was utilised to analyse the data. Results: Themes: importance of symptoms in triggering emotional reactions (e.g., general aches/pains, annual appointments) and ways of regulating these emotions (e.g., distraction, trust in health care team, future orientation). Health care professionals’ own emotions regarding recurrence had an impact on participants’ management of emotions. Conclusion: The findings speak to the similarities in the experience of people who have been treated for cancer and can be transferred to other illnesses that involve uncertainty about future. Health care professionals’ communication of their fears impacts on patient’s fears. This work has illuminated the significance of assessing people’s FoR following completion of treatment and symptom-related triggers for FoR. CORRESPONDING AUTHOR: Dr G. Ozakinci, University of St Andrews, United Kingdom, [email protected] S651 CONCEPTUALIZING AND ASSESSING STRATEGIES FOR REGULATING ILLNESS-RELATED EMOTIONS: THE CASE OF WORRY ABOUT CANCER RECURRENCE L. Cameron1, Loshni Rogers2 1 University of California, Merced, MERCED, United States of America 2 The University of Auckland, AUCKLAND, New Zealand Introduction. Coping with illnesses such as cancer typically requires ongoing emotion regulation to control distress and worry. Based on an expanded Common-Sense Model (CSM) delineating specific emotion regulation strategies, we developed new measures assessing efforts to manage worry of cancer recurrence and used them to examine emotion regulation dynamics in women following breast cancer treatment. Methods. We recruited 143 women with breast cancer in remission from the New Zealand Cancer registry to complete a survey of worry about cancer recurrence and strategies to regulate worry, and to complete a similar survey one month later. Emotion regulation measures included problem-solving, attentional deployment (avoidance, distraction), cognitive change (optimistic reframing, spiritual focus), emotional modulation (expression, suppression, humor, substance use) and proactive behaviors (healthy diet, relaxation). Results. Regression analyses revealed that, controlling for clinical and demographic characteristics, specific emotion regulation strategies predicted worry of recurrence and changes in worry over time. Emotional expression (β=-.24, p<.05) and optimistic reframing (β=-.43, p<.01) predicted lower levels of concurrent worry, while optimistic reframing (β=-.23, p<.05) and spiritual focus (β=-.22, p<.05) predicted decreases in worry over time. Conversely, avoidance (β=.26, p<.05) and problem solving strategies (β=.21, p<.05) predicted greater concurrent worry; problem-solving strategies (β=.18, p<.05) and avoidance (β=.36, p<.01) also predicted increases in worry over time. Conclusion. The expanded CSM and emotion regulation measures provide new tools for conceptualizing and assessing emotion regulation dynamics in the context of worry about cancer recurrence. More generally, they provide a framework and approach for developing similar, illness-specific measures of emotion regulation. CORRESPONDING AUTHOR: Prof L. Cameron, University of California, Merced, United States of America, [email protected]

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

Going global with the Finnish goal program: a scaleable lifestyle intervention to improve diabetes prevention (22B) S652 GOING GLOBAL WITH THE FINNISH GOAL PROGRAM: A SCALEABLE LIFESTYLE INTERVENTION TO IMPROVE DIABETES PREVENTION R. Valve1, Brian Oldenburg2, Andre Kengne3, Pilvikki Absetz4 1 University of Helsinki, LAHTI, Finland 2 Monash University, MELBOURNE, Australia 3 South African Medical Research Council, JOHANNESBURG, South Africa 4 Collaborative Care Systems Finland, HELSINKI, Finland Two seminal efficacy trials published in the early 2000’s, the DPS in Finland and the DPP in the USA, established lifestyle change as the most effective means to prevent type 2 diabetes. The findings resulted in a global call for research translating the evidence into real world. The Finnish GOAL Program was one of the first implementation trials to demonstrate that similar outcomes could be achieved with a less resource intensive group-based program led by health professionals. Scalability and sustainability has now also been demonstrated more widely in Finland and other countries. In this symposium, the first presentation will describe the program background and initial evaluation results from Finland and a replication trial in Australia, the GGT DPP. The second presenter will describe how the GOAL has been adapted for implementation in low and middleincome countries with an example from India, where the model was translated into the Kerala Diabetes Prevention Program, a peer-led support program that seeks to empower not just participants but also their families and communities. Finally, the last presenter will describe the adaptation of the GOAL program for South Africa and other countries in Sub-Saharan Africa. There will then be a discussion focusing on key factors behind successful translation and implementation of lifestyle interventions such as the GOAL Program. CORRESPONDING AUTHOR: Ph.D. R. Valve, University of Helsinki, [email protected] S653 DIABETES PREVENTION IN THE REAL WOLD: ESTABLISHING EFFECTIVENESS AND FEASIBILITY OF THE GOAL MODEL R. Valve1, Pilvikki Absetz2, Brian Oldenburg3, Tiina Laatikainen1, James A Dunbar4 1 University of Helsinki, LAHTI, Finland 2 Collaborative Care Systems Finland, HELSINKI, Finland 3 Monash University, MELBOURNE, Australia 4 Flinders University and Deakin University, WARRNAMBOOL, Australia Introduction. Efficacy trials such as the Finnish Diabetes Prevention Study, DPS, showed that type 2 diabetes could be prevented with lifestyle changes. The GOAL Lifestyle Implementation Trial (GOAL LIT) adapted the evidence into routine health care, and the Australian GGTDPP established cultural transferability of the GOAL model. Methods.The GOAL model included six group sessions over 8 months, delivered by trained nurses and dieticians. Participants with elevated diabetes risk were recruited from primary health care for the GOAL LIT (N = 352, mean age = 58) and from GP practices for the GGT-DPP (N=???, mean age). Nutrition, physical activity and clinical risk factors were documented at baseline and 12 months in both studies. In GOAL LIT, maintenance was measured at 36 months. Lifestyle outcomes were compared with the relevant efficacy trials, and within-subject changes were tested for risk reduction. Results.The GOAL model was successful in replicating the lifestyle outcomes in both studies. Risk factor reduction was slightly lower than

S193

in the reference trial. In the GOAL LIT, the weight loss outcome was maintained at three years, and blood glucose had reverted to normal among 43% of those with elevated blood glucose at baseline. Both studies were able to establish high feasibility of implementation. Conclusion.The GOAL model seems feasible for routine care in highincome countries. With established effectiveness, and no need for additional resources, it provides an alternative to individual-based routine risk factor control for patients at risk. In Finland and Australia, the program has been successfully scaled up. CORRESPONDING AUTHOR: Ph.D. R. Valve, University of Helsinki, [email protected] S654 CLUSTER RANDOMISED CONTROLLED TRIAL OF COMMUNITY-BASED DIABETES PREVENTION: THE KERALA DIABETES PREVENTION PROGRAM B. Oldenburg University of Melbourne, MELBOURNE, Australia Background. After China, India has the largest number of people with Type 2 Diabetes Mellitus. We describe findings from a study which is evaluating a “real world”community-based diabetes prevention program in rural Kerala, India which is adapted from the Finnish GOAL program. Methods. A total of 60 rural neighbourhoods (“clusters”) are randomised to intervention or control conditions. Participants at “high risk”of diabetes are recruited in 2 steps: (1) If people aged 30-60 years have an Indian Diabetes Risk Score (IDRS) value of ≥60, they are invited to attend a mobile clinic (HOME SCREENING); and (2) At the clinic they complete questionnaires, undergo physical measurements, and provide blood samples (MOBILE CLINIC). Participants in the intervention arm participate in a group-based, lay peer support intervention program over 12 months to motivate, guide and support planning, initiation and maintenance of lifestyle changes aimed at reducing diabetes risk. Participants in the control arm are provided with a health education booklet. Follow-up assessments are conducted at 12 and 24 months. Results. 3202 individuals underwent home screening, 44% (n=1410) of these had an IDRS ≥60. 86% (n=1211) of these individuals attended the mobile clinic. Two hundred were excluded as having diabetes, leaving 1011 participants in the trial, with an average cluster size of 17. Intervention fidelity is very good in 29/30 intervention clusters and intervention feasibility and acceptability is high. Discussion. Results from this trial will contribute to improved policy and practice regarding community-based diabetes prevention programs in India and other resource-constrained settings. CORRESPONDING AUTHOR: Professor B. Oldenburg, University of Melbourne, Australia, [email protected] S655 CONTEXT AND SUITABILITY FOR AN ADAPTED GOAL PROGRAM IN SOUTH AFRICA AND SUB-SAHARAN AFRICA A. Kengne South African Medical Research Council, CAPE TOWN, South Africa Background. The largest current and projected relative increase in the population with diabetes around the world is expected to occur in subSaharan Africa (SSA). This is driven by the large and increasing number of people at risk, who will progress to the full stage of the disease with time. Therefore, in addition to providing care to those with diabetes in SSA, efforts are needed to prevent the development of the disease in those who are at high risk. Methods. We have reviewed the successful diabetes prevention implementation programs and models of translation, with a focus on the Finnish GOAL program and the sister projects. We have reviewed the health system, the culture of SSA countries using South Africa as a

S194

model, to identify the context specific barriers and opportunities for a successful diabetes prevention implementation program. Results. Most components of the GOAL program can be adapted to the SSA settings. Community-based intervention delivered via groups is a potential resource-sensitive approach. Empowering non-professional health workers will help overcoming the shortage of trained health professionals. The high penetration rate of mobile phone positions mobile technology as a very useful tool for sustaining the intervention. Collaboration with health decision makers and community leaders can facilitate the scalability of the program. Conclusion. This evidence is been used to inform to design of the South African Diabetes Prevention Implementation Project and use it as a model for other sub-Saharan African countries. CORRESPONDING AUTHOR: Prof A. Kengne, South African Medical Research Council, South Africa, [email protected] INVITED SYMPOSIUM: Behaviour change techniques to facilitate patients' self-management of their rheumatic disease S656 BEHAVIOUR CHANGE TECHNIQUES TO FACILITATE PATIENTS' SELF-MANAGEMENT OF THEIR RHEUMATIC DISEASE E. Dures1, R. Arrends2, H. Zangi3, J. Dekker4 1 University of the West of England, BRISTOL, United Kingdom 2 University of Twente, ENSCHEDE, The Netherlands 3 Diakonhjemmet Hospital, OSLO, Norway 4 VU Medical Center, AMSTERDAM, The Netherlands The implementation of behavior change techniques to reduce the impact of rheumatic and musculoskeletal disease is not widespread in rheumatology settings. However, evidence is accumulating that supporting patients’ self management is effective. This symposium will present research on the use of a range of behavior change techniques designed to facilitate patients’ self-management of their rheumatic disease. The first speaker will present findings from qualitative research with rheumatology clinicians who have undertaken brief cognitivebehavioural skills training, and who are applying the techniques in routine consultations. The second speaker will present results from a randomised controlled trial of a mindfulness-based intervention, the Vitality Training Program, to support patients with inflammatory arthritis. The third speaker will present an in-depth process-evaluation of a goal management intervention given by rheumatology nurses. The processevaluation focused on both specific and unspecific factors of the delivery of the intervention. Optimal ways of training health professionals for a psychosocial intervention will be discussed. The last speaker will review approaches towards promoting adherence to exercise in rheumatic and musculoskeletal diseases, and evidence supporting the application of these approaches. Together the four presentations will demonstrate the effectiveness of using behaviour change techniques to facilitate patients’ self management in a rheumatology setting.

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

make lifestyle changes. Taking an active role in their own care is rarely a choice that patients can enact alone, and is highly influenced by the dynamic created in the way care is provided. Cognitive-behavioural techniques can facilitate self-management, but understanding is needed of the challenges and benefits of learning and implementing such techniques. Methods: Semi-structured interviews with 16 rheumatology clinicians who have undertaken brief skills training, to identify barriers and facilitators of using cognitive-behavioural techniques in routine clinical care. Results: Barriers included: a lack of time in clinic; the difficulties of changing the existing pattern and focus of interaction; and concerns about exploring social and emotional aspects of living with arthritis. Facilitators included: training that balanced theory with time to practice skills and receive feedback; access to clinical supervision to gain confidence and develop more advanced techniques; and the perception that patients gained a greater sense of control and were better able to take responsibility for their treatment, as a consequence of a more collaborative consultation. Conclusions: Rheumatology clinicians are well placed to utilise a range of basic cognitive-behavioural skills and techniques to facilitate a collaborative interaction and support self-management. These include helping patients to identify their priorities and formulation to unpick the relationship between thoughts, feelings, symptoms and behaviours. CORRESPONDING AUTHOR: Dr E. Dures, University of the West of England, United Kingdo, [email protected] S658 USING MINDFULNESS-BASED APPROACHES TO SUPPORT PATIENTS WITH RHEUMATIC DISEASES H. Zangi1, Kåre B. Hagen1, Arnstein Finset2 1 Diakonhjemmet Hospital, OSLO, Norway 2 Dept. of Behavioural Science in Medicine, University of Oslo, OSLO, Norway

S657 R H E U M ATO L O G Y C L I N I C I A N S U S I N G C O G N I T I V E BEHAVIOURAL TECHNIQUES IN ROUTINE CONSULTATIONS: BARRIERS AND FACILITATORS E. Dures, S. Hewlett University of the West of England, BRISTOL, United Kingdom

Introduction: Living with chronic inflammatory arthritis (IA) may be stressful to patients. In recent years mindfulness-based therapies have been applied in the treatment of the mental distress that may accompany various chronic diseases. It is believed that experiencing the present moment with an attitude of openness, curiosity and acceptance can effectively counter the effect of stressors, whereas excessive orientation towards the past or future can be related to feelings of depression and anxiety. The Norwegian Vitality Training Program (VTP) is a 10-session group intervention that combined mindfulness training with various creative exercises, aiming at enhancing patients' health promoting resources. Methods: The VTP was evaluated in a randomised controlled trial including 73 patients with IA, 37 were allocated to the VTP group, and 36 to the control group. Effects in psychological distress, self-efficacy, pain, fatigue and emotion-focussed coping were assessed by patient-reported outcome measures post-treatment and at 12-months follow-up and analysed by mixed-models repeated measures. Results: Participants were 79% female, mean age 54 years and mean disease duration 16 years. Sixty-eight persons (96%) were available for follow-up post-treatment, and 67 (94%) at 12 months follow-up. Significant treatment effects in favour of the VTP were found post-treatment in psychological distress, self-efficacy, fatigue and emotional processing. Effects were maintained or increased at 12 months follow-up (effect sizes 0.92 to 0.50). Conclusion: A mindfulness-based intervention, such as the VTP, can support patients who experience their condition as distressing and be a valuable complement to their disease management. The long-term effects are notably promising.

Introduction: Self-management requires informed, activated patients to manage the physical and psychosocial consequences of arthritis, and to

CORRESPONDING AUTHOR: H. Zangi, Diakonhjemmet Hospital, Norway, [email protected]

CORRESPONDING AUTHOR: Dr E. Dures, University of the West of England, United Kingdom, [email protected]

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

S195

S659 AN IN-DEPTH PROCESS-EVALUATION OFA GOAL MANAGEMENT INTERVENTION GIVEN BY RHEUMATOLOGY NURSES R.Y. Arends, C. Bode, E. Taal, M.A.F.J. Van de Laar University of Twente, ENSCHEDE, Nederland

O661 MEDICAL STUDENTS BECOMING YOUNG DOCTORS: IMPACTING MENTAL HEALTH? J.K. Sluiter, M.H.W. Frings-Dresen Academic Medical Center, AMSTERDAM, Nederland

Introduction. “Right on target!”is a psychosocial goal management group intervention, given by trained rheumatology nurses, that focuses on the threatened goals of patients. The new intervention goes beyond traditional self-management courses as it is aimed at the needs and capacities of the individual participant. The detailed process-evaluation was focused on both specific and unspecific aspects of the delivery of the intervention. Method. Both qualitative and quantitative analyses were applied to (i) written weekly evaluations from 63 participants and 5 trainers, (ii) 14 indepth structured interviews with a random selection of participants post intervention, and (iii) random audio recordings of 16 sessions. Results. Specific factors: trainers accurately followed the manual, and participants evaluated the intervention as good (mean is 4.36 on a scale of 1= very poor till 5=very good). The exercises were evaluated as mixed, with as prominent example the mental simulation that was evaluated as either an eye-opener, or challenging but helpful, or personally ineffective. Unspecific factors: trainers needed additional guidance for applying the appropriate coaching attitude and psychological communication skills. Participants evaluated the atmosphere as open and familiar. Conclusions. Participants evaluated the exercises as mixed; however, final evaluation of the intervention was good. Supervision during the intervention period was applied to reduce differences between trainers. Inclusion of unspecific factors is recommended for training of future trainers.

Introduction: Medical study is considered stressful. Two study phases are distinguishable: the non-clinical phase (yr. 1-4) and clinical phase (yr. 5-6). The question of whether study phase and it’s stability over a one-year period is associated with the prevalence of common mental health complaints was studied. Methods: Two Dutch medical faculties participated in a cohort study with online questionnaires. Stress, depression and anxiety were assessed with screening instruments. Prevalence of common mental health complaints at follow-up was compared in three group of students with respect to their study phase stability: 1) stable (non-clinical), 2) stable (clinical), 3) unstable (from non-clinical to clinical). Results: Data from 958 medical students was used. The participation rate at follow-up was 57%. In group 1, high mental complaints, were: 21% (stress), 32% (anxiety) and 39% (depression). In group 2, high mental complaints, were: 9% (stress), 24% (anxiety) and 28% (depression). In group 3, high mental complaints, were: 10% (stress), 24% (anxiety) and 35% (depression). Significantly more stressed students were found in the young non-clinical phase of the study compared to students further in the process. Significant differences between groups were also found for depressive complaints. More students have depression complaints in the group that changed from non-clinical to the clinical phase compared to their colleagues already active in the clinical phase. Conclusions: Different mental health complaints play a significant role in the early phase of the medical study compared to while transitioning to, and within, the clinical phase. Guidance for medical students is called for.

CORRESPONDING AUTHOR: MSc R.Y. Arends, University of Twente, Nederland, [email protected] S660 PROMOTING EXERCISE ADHERENCE IN RHEUMATIC AND MUSCULOSKELETAL DISEASES-STATE OF THE SCIENCE AND NEW DEVELOPMENTS J. Dekker VU University Medical Center, AMSTERDAM, Nederland Background. Maximizing patient adherence is a key element to success of medical interventions, exercise included. A range of approaches towards promoting exercise adherence have been developed and evaluated. Purpose. To review approaches towards promoting adherence to exercise in rheumatic and musculoskeletal diseases, and evidence supporting the application of these approaches. Method. This review draws upon a systematic review and a meta-analysis on improving adherence to exercise, supplemented with recent developments. Promoting exercise in osteoarthritis of the knee or hip is used as an illustrative example . Results. Evidence from the systematic review and meta-analysis supports the application of (i) behavioral approaches towards promoting exercise adherence (e.g. self-management program, behavioral graded activity), (ii) dedicated delivery of exercise (supervised exercise, providing audio- or videotapes, and refresher sessions), and (iii) mobile devices (personal digital assistants, mobile phones). Although empirical evidence is lacking, expert opinion suggests that (iv) tailoring exercise to the patient’s primary medical condition and to comorbidity improves adherence. Conclusion. Although traditionally seen as a difficult task, the current state of the science and recent developments suggest substantial progress in promoting exercise adherence in rheumatic and musculoskeletal diseases. CORRESPONDING AUTHOR: Professor J. Dekker, VU University Medical Center, Nederland, [email protected]

CORRESPONDING AUTHOR: Professor J.K. Sluiter, Academic Medical Center, Nederland, [email protected] O662 WORKPLACE BULLYING AND DEPRESSIVE SYMPTOMS: A PROSPECTIVE STUDY AMONG JUNIOR PHYSICIANS A. Loerbroks1, M. Weigl2, J. Li1, J Glaser3, C. Degen1, P. Angerer1 1 University of Düsseldorf, DÜSSELDORF, Germany 2 Ludwig-Maximilians-University, MUNICH, Germany 3 University of Innsbruck, INNSBRUCK, Austria Introduction: Prospective evidence on the potential association between workplace bullying and depressive symptoms remains sparse. We set out to explore this relationship in a cohort of junior physicians, who represent an occupational group frequently exposed to bullying in their workplace. Methods: In 2004, a total of 621 junior doctors working in hospitals in the city of Munich, Germany, and surrounding communities participated in a baseline survey. Baseline participants were followed-up in 2005 (Fup1) and 2007 (Fup2). The current investigation was restricted to those with complete data at all assessments (n=507 or 82%). To measure workplace bullying, an initial description of this phenomenon was provided followed by an item inquiring whether the respondent felt she/he had been exposed. Depressive symptoms were assessed by the State-Trait Depression Scale. Associations of bullying at baseline with depressive symptoms (both trait and state scores) at Fup1 or Fup2 were estimated by linear regression models. Results: Adjusting for demographics, work conditions, health-related lifestyles, health outcomes and the respective baseline depression score, exposure to workplace bullying at baseline predicted increased depressive symptoms (state) at both Fup1 (beta=1.44, p=0.01) and Fup2 (beta=1.53, p=0.01). Similar associations were observed for bullying as a predictor of trait depressive symptoms (beta=1.32, p=0.01, and beta=1.54, p=0.00, respectively). Conclusion: Exposure to bullying at the workplace predicts depressive symptoms among junior physicians. These findings highlight the need for

S196

interventions aiming at the prevention or reduction of workplace bullying in order to protect or improve the mental health of junior physicians. CORRESPONDING AUTHOR: Dr. A. Loerbroks, University of Düsseldorf, Germany, [email protected] O663 PHYSICIANS' JOB DEMANDS, SUSTAINED ACTIVATION FROM WORK AND INTENTION TO LEAVE MEDICAL PRACTICE C. Degen1, M. Weigl2, A. Müller1, J. Glaser3, P. Angerer1 1 University of Düsseldorf, DÜSSELDORF, Germany 2 Ludwig-Maximilians-University, MUNICH, Germany 3 University of Innsbruck, INNSBRUCK, Austria Introduction: The evolving physician shortage in many countries requires examination of factors that determine intention to leave medical practice. Employee’s intention to leave as a reaction on sustained activation from work during free time has not yet been given attention in research. We aim to research the impact of sustained activation from work on the relationship between job demands and physicians` intention to leave clinical practice. Methods: The study is based on a three-wave prospective cohort study of 613 German junior doctors. Two aspects of sustained activation from work are researched. Cognitive activation was assessed by the Overcommitment Scale and behavioural activation was measured by a quotient derived from the share of work activity at leisure time and share of relaxation at leisure time. For the relationship between job demands measured by work overload and intention to leave a mediation analysis for sustained activation from work was conducted. Results: Mediation analyses were adjusted for intention to leave at baseline, age, gender, partnership, children, free weekends and working hours. High job demands significantly predicted intention to leave and cognitive as well as behavioural activation from work. Both aspects of activation predicted intention to leave. The effect of job demands on intention to leave was partially mediated by both aspects of sustained activation. Conclusion: Sustained activity from work during free time can explain parts of the effects of high job demands in physicians’ intention to leave medical practice. Intervention trainings for physicians should therefore promote methods that facilitate detachment from work. CORRESPONDING AUTHOR: Dr. C. Degen, University of Düsseldorf, Germany, [email protected] O664 JOB CHARACTERISTICS AND GOAL ORIENTATION PREDICT WORK ENGAGEMENT AND BURNOUT IN EMERGENCY NURSES L. Sibilia SIMPS, ROMA, Italy Stan Maes, Jef Adriaenssens & Véronique De Gucht, Health Psychology, Leiden University, The Netherlands Introduction. This study, carried out in 13 emergency departments (ED) in Belgium, investigated to what extent goal orientation adds to the explanation of burnout and work engagement in ED-nurses after controlling for job characteristics (demands, control and social support). Method. 170 out of 274 ED-nurses (response rate 62%) completed the Leiden Quality of Work Questionnaire for Nurses, the Goal Orientation Questionnaire, The Maslach Burnout Inventory (MBI-HSS) and the Utrecht Work Engagement Scale (UWES). Hierarchical multiple regression analyses were conducted. Results. Work characteristics and goal orientation explained 38% and 37% of variance in burnout and work engagement respectively. Job control and social support were strong predictors of both outcomes (p<.001). Job demands was predictive of burnout only (p<.001). MasteryApproach goal orientation was related to an increase in work engagement

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

(p<.001) and a decrease in burnout (p<.05). Performance-avoidance goal orientation was related to a decrease in work engagement (p<.01) and an increase in burnout (p<.001). Performance-approach and masteryavoidance goal orientation were not predictive for the two outcome variables. Conclusion. A mastery-approach goal orientation is beneficial for occupational well-being while a performance-avoidance goal orientation has a negative effect. CORRESPONDING AUTHOR: Prof. L. Sibilia, SIMPS, Italy, [email protected] O665 PREDICTORS FOR RE-EMPLOYMENT SUCCESS AMONG UNEMPLOYED PERSONS: A PROSPECTIVE COHORT STUDY S. Brouwer1, J.M.H. Schellekens2 1 University Medical Center Groningen, GRONINGEN, Nederland 2 Department of Experimental & Work Psychology, University of Groningen, GRONINGEN, Nederland Background: In an effort to prevent long-term unemployment, promoting re-entry into the workforce as early as possible is an important goal for the unemployed as well as society. From literature in the field of return-towork and in long-term unemployment is it known that personal, healthrelated and behavioral (psychosocial) factors influence re-employment success. Little evidence has been found about the predictive value of these factors on re-employment success in persons in their first-year of unemployment. Aim: The aim of the present study was to investigate which factors predict re-employment success among persons who become unemployed and are claiming unemployment benefit in the Netherlands. Method: In a prospective cohort study with a one-year follow-up, we investigated a sample of 3.618 subjects who became unemployed. A survey was sent to all participants, including health-related, personal and behavioral factors, based on Wanberg’s model and the Theory of Planned Behavior model. Data about re-employment status were obtained from the SSI registry. Multivariate logistic regression analyses were performed. Results: After one-year follow-up N=2082 (57.5%) were re-employed. In the multivariate analysis, we found ten factors predicting re-employment success. Age, number of working years in previous job, understanding Dutch language, expectancies about re-employment, active job search behavior, job search intention and external variable attribution, perceived health and general work ability were found as predictors for reemployment success. Conclusion: This study showedthat personal-, health-related and behavioral factors play a role in predicting the re-employment success in the first year after becoming unemployed and to prevent long-term unemployment. CORRESPONDING AUTHOR: dr S. Brouwer, University Medical Center Groningen, Nederland, [email protected] O666 DO EXPECTANCIES OF RETURN TO WORK AND OVERALL JOB SATISFACTION PREDICT ACTUAL RETURN TO WORK? J. Opsahl1, T.H. Tveito2, H.R. Eriksen2 1 Uni Research, BERGEN, Norway 2 Uni Helse, Uni Research, BERGEN, Norway Introduction: Musculoskeletal disorders including low back pain have major individual and socioeconomic consequences as it often leads to disability and sick leave. Predictors of disability and return to work often differs, hence it is important to identify key predictors for return to work. The aim of the study was to assess if expectancies of return to work and overall job satisfaction predicts actual return to work after 12 months, among employees with long lasting low back pain. In addition, potential gender differences were explored.

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

Methods: Data from the multicenter cognitive interventions and nutritional supplements trial was used. Predictors for return to work was examined in 574 employees sick listed 2-10 months for low back pain. Data were analysed with multiple logistic regression models stratified by gender, and adjusted for potential confounders. Results: Regardless of gender, high expectancies were significant in predicting return to work at 12 months, while high levels of job satisfaction were not significant. Men with high expectancies had higher odds of returning to work compared to women. Men and women reported similar levels of expectancies and overall job satisfaction Conclusions: Among individuals with long lasting low back pain, high expectancies of returning to work were strongly associated with successful return to work. Screening expectancies and giving individuals with low expectancies interventions targeted at changing expectancies of return to work, such as CBT or self-management interventions, may contribute to increase actual return to work. CORRESPONDING AUTHOR: J. Opsahl, Uni Research, Norway, [email protected] O667 NEGATIVE AFFECTIVITY PREDICTS INCREASED PAIN SENSITIVITY DURING LOW-GRADE INFLAMMATION IN HEALTHY WOMEN T.E. Lacourt1, J.H. Houtveen2, J.J.C.S. Veldhuizen van Zanten3, J.A. Bosch4, M.T. Drayson5, L.J.P. Van Doornen4 1 The University of Texas MD Anderson Cancer Center, HOUSTON, TEXAS, United States of America 2 Utrecht University, Clinical and Health Psychology, UTRECHT, Nederland 3 University of Birmingham, School of Sport, Exercise and Rehabilitation Sciences, BIRMINGHAM, United Kingdom 4 University of Amsterdam, Department of Clinical Psychology, AMSTERDAM, Nederland 5 University of Birmingham, Division of Immunity and Infection, Medical School, BIRMINGHAM, United Kingdom Introduction: Experimental animal studies provided evidence for a synergistic effect of immunological and psychological stressors on subsequent changes in behavior and mood, among which increased pain sensitivity. This possibly provides a mechanism leading to the expression of functional (or medically unexplained) somatic symptoms. However, up to now little corroborating evidence for such synergy exists for humans. The aim of the present study was to determine an interaction between psychological stress and stress-vulnerability and an immunological activator on experimental pain sensitivity in healthy humans. Methods:In healthy female participants (n = 25, mean age 22.3 years), negative affectivity and experienced stress (perceived stress and life events) were assessed by questionnaire before receiving a Salmonella typhi vaccine or saline control in a randomized blinded cross-over design. Pain sensitivity was assessed as pressure pain thresholds on the lower back and calves and pain tolerance on the thumbnail, before and six hours after each injection. Results:Vaccination induced low-grade inflammation as shown by leukocytosis (+100%) and increased serum interleukin-6 (+670%). Negative affectivity predicted decreased pain tolerance after vaccination (β = -.57, p = .007), but not after placebo (β = .25, p = .26). Conclusions:Negative affectivity moderated the effects of inflammation on pain sensitivity. Negative affectivity has repeatedly been related to somatic symptom reporting in prior reports. Thus, our finding is consistent with a synergistic model whereby inflammation may lower the threshold for pain reporting in individuals with increased vulnerability for somatic symptom reporting. CORRESPONDING AUTHOR: T.E. Lacourt, The University of Texas MD Anderson Cancer Center, United States of America, [email protected]

S197

O668 PARENTAL ACCURACY REGARDING ADOLESCENT DAILY EXPERIENCES: ASSOCIATIONS WITH ADOLESCENT PSYCHOLOGICAL ADJUSTMENT AND INFLAMMATORY REGULATION L.J. Human1, M. Chan2, A. DeLongis3, L. Roy3, G.E. Miller2, E. Chen2 1 University of California, San Francisco, SAN FRANCISCO, United States of America 2 Northwestern University, EVANSTON, United States of America 3 University of British Columbia, VANCOUVER, Canada Introduction. There is evidence that parents play an important role in their adolescent’s health and well-being, but the specific psychobiological processes by which parents influence adolescent outcomes remain to be determined. In this study, we examined the role of parental accuracy - that is, whether parents who are more accurate about their adolescents’ daily experiences have adolescents with better psychological functioning and inflammatory regulation. Methods. In a two-week daily diary study of 116 parent-adolescent dyads, we examined whether parental accuracy about their adolescent’s daily demands and the positivity of their day together were associated with markers of psychological functioning and with regulation of the inflammatory response in terms of glucocorticoid (GC) sensitivity (the extent to which cortisol is able to dampen the production of inflammatory proteins) in adolescents. Results. Adolescents whose daily experiences were perceived more accurately by their parents reported better psychological adjustment (lower stress and depression) and a greater sensitivity of their immune cells to antiinflammatory signals from cortisol (i.e., diminished production of inflammatory proteins when cells were stimulated with the combination of a bacterial product (lipopolysaccharide, LPS) and cortisol). Conclusions. Greater parental accuracy regarding adolescents’ daily experiences is associated with better adolescent psychological adjustment and a more sensitive anti-inflammatory response to cortisol. These results provide preliminary evidence that parental accuracy regarding their adolescent’s daily experiences may be one specific daily parent factor that plays a role adolescent health and well-being. CORRESPONDING AUTHOR: L.J. Human, University of California, San Francisco, United States of America, [email protected] O669 SICKNESS BEHAVIOR DURING EXPERIMENTALLY-INDUCED SYSTEMIC INFLAMMATION: HIGHER SENSITIVITY SPECIFICALLY FOR PAIN OR ALL INTEROCEPTIVE SIGNALS? T. Roderigo 1 , L. Rebernik 2 , A. Wegner 3 , E. Engelbrecht 2 , M. Schedlowski2, S. Elsenbruch2, S. Benson2 1 Institute of Medical Psychology and Behavioral Immunobiology, ESSEN, Germany 2 Inst. of Med. Psychology and Behavioral Immunobiology, University Hospital Essen, ESSEN, Germany 3 Clinic for Trauma Surgery, University Hospital Essen, ESSEN, Germany Roderigo, T., Rebernik, L., Wegner, A.*, Engelbrecht, E., Schedlowski, M., Elsenbruch, S., Benson, S. Institute of Medical Psychology and Behavioral Immunobiology, *Clinic for Trauma Surgery, University Hospital Essen, Germany Introduction: Systemic inflammation reportedly leads to sickness symptoms including visceral hyperalgesia. This study aimed to determine if experimentally-induced inflammation specifically increases pain sensitivity or leads to increased interoceptive awareness in general. Methods: Based on a randomized double-blind crossover study-design, we assessed visceral pain sensitivity using an established rectal barostat distension protocol as well as interoceptive awareness using an established heartbeat perception (HBP) task in N=40 healthy males and females. On two study days, subjects received either an i.v. injection of

S198

endotoxin (lipopolysaccharide; 0.4ng/kg) to induce a transient systemic inflammation or saline. HBP and visceral pain sensitivity were assessed 2h post injection, along with repeated assessments of circulating cytokines. Results: Endotoxin administration expectedly led to increased levels of pro-inflammatory cytokines including TNF-alpha and interleukin-6. As expected, visceral pain thresholds were significantly decreased following endotoxin-induced immune activation. At the same time, interoceptive awareness (i.e., HBP) was significantly reduced during endotoxemia. Endotoxin-induced changes in nociception and interoception were not related (r=-.184, p=.269). Conclusion: Our findings suggest that systemic inflammation specifically increases visceral pain sensitivity, whereas subjects’ general ability to detect bodily signals (i.e., HBP) is lowered. Hence, inflammation-induced visceral hyperalgesia does not appear to be mediated by an increased general interoceptive awareness. These findings may help to further elucidate the relevance of inflammatory processes in functional (pain) disorders. CORRESPONDING AUTHOR: Dipl.-Psych. T. Roderigo, Institute of Medical Psychology and Behavioral Immunobiology, Germany, [email protected] O670 BLUNTED CARDIOVASCULAR AND CORTISOL STRESS REACTIVITY AND THE METABOLIC SYNDROME S.R. de Rooij AMC, AMSTERDAM, Nederland Introduction: The metabolic syndrome is a constellation of interrelated metabolic risk factors including glucose intolerance, insulin resistance, central obesity, dyslipidemia and hypertension that predisposes to type 2 diabetes and cardiovascular disease. Recently, it was shown that adiposity is associated with blunted cardiovascular and cortisol responses to psychological stress. The aim of the present study was to investigate the prevalence of the metabolic syndrome in those with low and high cardiovascular and cortisol stress reactivity. Methods: We performed a psychological stress protocol in a populationbased cohort of 725 late middle aged men and women. The stress protocol included three stress tests (Stroop, mirror drawing and speech) during which cardiovascular activity was continuously measured and saliva cortisol samples were taken. We assessed prevalence of the metabolic syndrome in the 10% lowest and 10% highest stress reactors (peak stress response minus baseline activity). Metabolic syndrome was defined according to the International Diabetes Federation. Results: Prevalence of the metabolic syndrome was higher in those with low heart rate than in those with high heart rate reactivity (52% compared to 29%, p < .01). This association survived adjustment for sex, age and basal activity (p = .02), but disappeared after adjustment for waist circumference (p = .82). There was no significant difference between those with low and high cortisol reactivity (53% compared to 45%, p = 0.43). Conclusion: Findings of the present study suggest that blunted cardiac stress reactivity is associated with an increased risk for type 2 diabetes and cardiovascular disease via increased adiposity. CORRESPONDING AUTHOR: dr S.R. de Rooij, AMC, Nederland, [email protected] O671 INFLAMMATION IS AN INDEPENDENT PREDICTOR OF DEPRESSIVE SYMPTOMS AND FATIGUE IN STABLE KIDNEY TRANSPLANT RECIPIENTS J.A. Bosch1, W. Chan2, A. Phillips2, D. Jones3, O. Kaur2, P. McTernan4, N. Inston2, S. Moore2, A. McClean2, L. Harper2, R. Borrows2 1 University of Amsterdam, AMSTERDAM, Nederland 2 University of Birmingham, BIRMINGHAM, United Kingdom 3 Manchester Metropolitan University, MANCHESTER, United Kingdom 4 University of Warwick, COVENTRY, United Kingdom

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

BACKGROUND: Experimental animal studies have demonstrated potent neurological effects of inflammation, which induces depressive-like and lethargic behaviours. While human data is comparatively sparse, these findings lead to the hypothesis that inflammatory activity may partly account for the high incidence of depression and fatigue in various chronic patient groups. The present study investigated if inflammatory activity is a predictor of depression and fatigue in stable kidney transplant recipients. METHODS: This cross-sectional study enrolled 128 stable kidney transplant patients beyond 12 months post transplantation [mean age = 50 ± 15 years; 56% male]. Depression and fatigue were assessed with the Hospital Anxiety and Depression Scale (HADS) and the Multidimensional Fatigue Inventory-20 (MFI-20), respectively. Fasting blood was taken for highsensitive analysis of the inflammatory marker C-reactive protein (hsCRP). RESULTS: hsCRP predicted individual differences in depressive symptoms (R=.32, p<.01) as well as the fatigue dimensions ‘reduced activity’ and ‘physical fatigue’ (R>.30, p<.01, analyses adjusted for age, ethnicity, BMI and gender). In patients reporting high depressive symptom levels (HADS>8), the mean level of hsCRP was 7.85 mg/L versus 3.63 mg/L in non-depressed patients (p<.01). Further statistical adjustments for potential confounding by co-morbidity (Index of Coexistent Disease) or kidney function (estimated glomerular filtration rate, haemoglobin levels) did not attenuate any of the observed associations. CONCLUSION: In stable kidney transplant patients, inflammatory activity is an independent predictor of depressive symptoms and lethargy. These findings suggest that control of inflammation may be a relevant target for interventions that aim to improve well-being and activity levels in these patients. CORRESPONDING AUTHOR: Dr J.A. Bosch, University of Amsterdam, Nederland, [email protected] O672 LIFESTYLE MEDIATED EFFECTS OF PSYCHOLOGICAL DISTRESS ON HS-CRP AND FIBRINOGEN IN NONOBSTRUCTIVE CORONARY ARTERY DISEASE P.M.C. Mommersteeg 1 , S.H. Meeuwis 1 , J. Denollet 1 , J.W. Widdershoven1, W. Aarnoudse2, B.L.W.J.J Westerhuis3, W.J. Kop1 1 Tilburg University, TILBURG, Nederland 2 TweeSteden Hospital, TILBURG, Nederland 3 Elisabeth Hospital, TILBURG, Nederland Objective: Psychological factors associated with distress are predictive of acute coronary syndromes and poor prognosis in coronary artery disease (CAD). These factors have been associated with markers of inflammation that play a role in atherosclerosis. However, biobehavioral processes in non-obstructive CAD are not well understood. The aim of the present study is to establish the association between psychological distress with high sensitive C-reactive protein (hs-CRP) and fibrinogen, and to identify mediating factors. Methods: Patients with non-obstructive CAD (“wall irregularities”, stenosis <60%, N=414, mean age 62.1±9.25, 52% women) participating in the TweeSteden Mild Stenosis Study (TWIST) completed the hospital anxiety and depression scale (HADS) to measure psychological distress. Blood samples were analysed for hs-CRP and fibrinogen. Multivariate regression analysis was used to examine the association of distress, anxiety and depressive symptoms with hs-CRP and fibrinogen, adjusting for demographic and biomedical covariates. Mediating effects of lifestyle factors were assessed using bootstrap sampling procedures. Results: Depressive symptoms were positively associated with hs-CRP level (β=.135, p=.009), but not fibrinogen (β=.075, p=.153), adjusted for age and sex. No associations were observed for distress or anxiety (p>0.10). Mediation analysis showed that the association of depressive symptoms with hs-CRP was mediated via body mass index (BMI) and smoking.

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

Conclusion: Among patients with non-obstructive CAD, depressive symptoms were associated with increased hs-CRP levels, mediated by BMI and smoking. Because acute coronary syndromes can occur in the setting of minimally obstructive CAD, these findings provide an important biobehavioral pathway for the association between depression and acute coronary syndromes. CORRESPONDING AUTHOR: Dr. P.M.C. Mommersteeg, Tilburg University, Nederland, [email protected] O673 THE HEALTH BEHAVIOR TAXONOMY G. Nudelman, S. Shiloh Tel Aviv University, TEL AVIV, Israel Introduction: Health behaviors (HBs) are important causes of morbidity and mortality. They encompass a large and heterogeneous group of behaviors (e.g. smoking, vaccination) which obscures the psychological processes involved and necessitates developing a taxonomy representing lay-people's perceptions of HBs. Method: Four studies were conducted. Study 1 created a comprehensive list of HBs based on reports from 70 laypersons, 30 health promoters and a literature review. Study 2 examined the perceived importance of those HBs (n=268). Study 3 applied a card sorting technique to assess perceived similarities between HBs (n=450), and Study 4 replicated the sorting task through an internet sample (n=627). Results: An initial list of 66 HBs was reduced to 45 behaviors perceived as most important. Hierarchical clustering analysis conducted on proximity matrices representing the sorted data yielded a structure (rsquared=0.93) comprised of four main clusters: Risk Avoidance, Nutrition Behaviors, Health Routines and General Well-Being (psycho-social factors), that were further divided into sub-clusters. This taxonomy was replicated across samples and sub-groups divided by gender, age, income and education (r-spearman=0.97). Nutrition was perceived as the most important cluster, and General well-being as the least important one. A structure based on reported practice of the same HBs was only weakly associated with the taxonomy (r-spearman=0.39). Discussion: The taxonomy of HBs is a novel and stable classification representing lay perceptions and their cognitive schema of HBs. It is different from classifications based on behavioral reports, and enhances our understanding of laypersons’ perspective, which may help design more comprehensive and cost-effective intervention programs. CORRESPONDING AUTHOR: G. Nudelman, Tel Aviv University, Israel, [email protected] O674 HEALTH IDENTITY THEORY: A NEW APPROACH TO UNDERSTANDING HEALTH BEHAVIOR R.R.H. Hyatt, K. McKay, N. Sullivan Tufts University School of Medicine, BOSTON, United States of America Introduction: Our ability to design effective and sustainable interventions within complex public health and medical contexts requires theoretical frameworks that allow for the definition and testing of hypotheses to better understand health related behaviors. Many of the extant theoretical frameworks fall short of this goal in their ability to address the social-behavioral mechanisms underlying health behavior. As a result, our ability to generate new hypotheses to support research on issues such as health disparities, adherence to behavioral interventions, and intergenerational transfer of health behavior within families is limited. Methods: This paper presents results from 28 one-hour telephone interviews with a voluntary sample of adults in Boston, MA. Quantitative and qualitative results are presented which support the preliminary

S199

construction of a new health behavior theory based on social identity theory (Burke and Stets, 2009). Results: This approach suggests that individual health behavior rests upon one’s unique health identity and argues that health behavior is a product of socialization. The Health Identity framework is useful for generating hypotheses about the underlying mechanisms of health behavior that facilitate examination of relationship between health behaviors and associated outcomes. Examples are presented which demonstrate the efficacy of Health Identity Theory in the study of health disparities, obesity, and adherence to medication protocols. Conclusion: Health Identity Theory presents opportunities to design behavioral public health interventions that are sustainable from multiple perspectives including individual lifestyle, public and community health, and cost. CORRESPONDING AUTHOR: Dr. R.R.H. Hyatt Tufts University School of Medicine, United States of America, [email protected] O675 WHAT HELPS YOUNG PEOPLE TO STAYAWAY FROM DRINKING AND SMOKING? N. Cable University College London, LONDON, United Kingdom Introduction: Current research evidence for harm associated with risky health behaviours suggests importance of educating young people as a preventive measure. Yet little is known what determines young people not to drink alcohol or smoke cigarettes. Methods: Available cases (N=1714) from the Understanding Society youth study (age 10-15) waves 2 and 3 were used to test whether young people’s attitudes toward drinking alcohol or smoking, happiness, selfesteem, supportive friend network size at wave 2 would determine four patterns of consecutive drinking and smoking behaviours (1=persistent non user, 2=quitters, 3=starters, 4=persistent users, reference category) from the waves 2 to 3. Multinomial logistic regression was used to test this question, while young people’s sex, age, religiosity, household social position, residential characteristics, and previous health status were included in the study model as confounders. Findings were corrected for clustering and response patterns. Results: Preliminary findings showed greater awareness of drinking alcohol or smoking was associated with persistent none use as well as new use. Smaller supportive friend networks were also associated with both types of users; however, greater happiness differentiated persistent none users from starters. Quitters were likely to be religious, boys, and higher social position. Conclusions: Promoting children’s happiness can be used as a preventive measure for young people to stay away from harm due to drinking or smoking.as well as education for young people to stay away from harm due to drinking or smoking. CORRESPONDING AUTHOR: Dr. N. Cable, University College London, United Kingdom, [email protected] O676 NEED FOR ACTION: PREVENTIVE CARE FOR CHRONIC DISEASE RISK FACTORS WITHIN COMMUNITY MENTAL HEALTH SERVICES K. Bartlem 1, J Bowman1 , M. Freund1 , P. Wye 1, J. Knight 2, K. McElwaine1, K. Gillham2, J. Wiggers2 1 University of Newcastle, CALLAGHAN, Australia 2 Population Health, Hunter New England Local Health District, WALLSEND, Australia INTRODUCTION: People with a mental illness experience a significant chronic disease burden, contributed to by an increased prevalence of health risk behaviours such as smoking, inadequate nutrition, alcohol

S200

over-consumption, and physical inactivity (SNAP).This study reports the findings of a survey of Australian community mental health (CMH) clients to determine the prevalence of SNAP risks, the receipt of preventive care for these risks, and the acceptability towards receiving such care. METHODS: A cross-sectional telephone interview study was conducted within one local health district in Australia. Clients (n=558) of CMH services were asked about their engagement in SNAP health risk behaviours, their receipt of preventive care (assessment, advice, and referral) for SNAP risks, and their acceptability towards receiving such care. Associations between receipt of care and client and service characteristics were explored. RESULTS: Alcohol consumption was most commonly assessed (76%), followed by smoking (73%), physical activity (57%), and nutrition (26%). Smokers were most commonly offered referral for treatment (42%), compared to alcohol overconsumption (32%), inadequate physical activity (35%), and inadequate nutrition (19%). Preventive care was more likely provided to clients who reported it was acceptable to receive such care, and those who had a greater number of appointments in the previous 12 months. CONCLUSIONS: The findings demonstrate the sub-optimal provision of preventive care within CMH, and highlight the need for such services to improve preventive care delivery. Intervention research is required to determine effective strategies to increase such care in CMH settings. CORRESPONDING AUTHOR: Ms K. Bartlem, University of Newcastle, Australia, [email protected] O677 CAN PROFESSIONAL FOOTBALL CLUBS HELP MEN ACHIEVE SUSTAINED IMPROVEMENTS IN WEIGHT, PHYSICAL ACTIVITY AND DIET? C. Gray1, K. Hunt2, C. Bunn3, S. Wyke3 1 Glasgow University, GLASGOW, United Kingdom 2 MRC/CSO Social and Public Health Sciences Unit, University of Glasgow, GLASGOW, United Kingdom 3 Institute of Health and Wellbeing, University of Glasgow, GLASGOW, United Kingdom Introduction: The prevalence of male obesity is rising, but men often avoid traditional weight management programmes; many perceive them as ‘feminised spaces’. Football Fans in Training (FFIT) draws on psychological and sociological understandings of masculinity to offer a weight management programme designed to appeal to men. FFIT comprises 12, weekly, group-based sessions delivered by community coaches at Scotland’s top professional football clubs, with ongoing, light-touch maintenance support. We evaluated FFIT’s effectiveness in helping men (aged 35-65, BMI≥28kg/m2) make sustained positive weight, physical activity (PA) and dietary changes. Methods: A two-arm, pragmatic randomised controlled trial at 13 football clubs. Men were randomly assigned to intervention (N=374) and waitlist (N=374) comparison groups, and followed at 12 weeks and 12 months. The primary outcome was 12 month weight loss. Focus group discussions (N=13) used a self-determination theory framework to explore men’s experiences of weight loss maintenance at 12 months. Results: Men in the intervention group had lost significantly more weight at 12 months than the comparison group: 5·56 kg (95% CI 4·70-6·43) vs. 0·58 kg (0·04-1·12). Improvements in self-reported PA and diet were also significantly greater in the intervention group. Men described a process of internalisation of motivation and the development of new positive PA and healthy eating habits. Work, injury and life events were barriers to maintenance, but being part of a group helped some men deal with relapse. Conclusion: FFIT was successful in engaging men in weight management and helped them achieve sustained weight loss, PA and dietary change. CORRESPONDING AUTHOR: Dr C. Gray, Glasgow University, United Kingdom, [email protected]

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

O678 TACKLING ALCOHOL-RELATED HARM IN SPORT: A CLUSTER RANDOMIZED CONTROLLED TRIAL WITH FOOTBALL CLUBS M. Kingsland1, L. Wolfenden1, J. Tindall2, B. Rowland3, K. Gillham2, P. Dodds2, C. Lecathelinais2, P. McElduff1, M. Sidey4, I. Crundall4, J. Rogerson4, J. Wiggers1 1 The University of Newcastle, WALLSEND, Australia 2 Hunter New England Population Health, WALLSEND, Australia 3 Deakin University, BURWOOD, Australia 4 Australian Drug Foundation, MELBOURNE, Australia Aim: To examine the effectiveness of an alcohol management intervention in reducing risky alcohol consumption by football club members and associated alcohol-related harms. Methods: The study used a cluster randomized controlled design, with football clubs randomized to control or intervention groups. The 2.5 year intervention included multiple strategies designed to decrease supply of alcohol to intoxicated members, increase purchase of non/ low-alcoholic drinks and cease drinking games and discounted/ free alcohol. The main outcome measures were risky alcohol consumption (5+ drinks) at the club and overall risk of alcohol-related harm (AUDIT tool). Results: 88 football clubs agreed to participate in the trial (43 control; 45 intervention) and 1411 and 1143 club members completed pre- and postintervention surveys. At follow-up, significantly more intervention clubs had implemented at least 13 of 16 alcohol management practices (88% v 65% OR:3.7; p=0.04) and significantly lower proportions of intervention club members reported: risky alcohol consumption at the club (19% v 24%; OR:0.63; p=0.05); overall risk of alcohol-related harm (38% v 45%; OR:0.58; p<0.01); alcohol consumption risk (47% v 55%; OR:0.60; p<0.01) and possible alcohol dependence (1% v 4%; OR:0.20; p<0.01). The intervention did not impact club income (p=0.38). Conclusions: Sports clubs can reduce risky alcohol consumption at the club as well as overall risk of alcohol-related harm to members. With large numbers of people worldwide playing, watching and officiating sport, club-based alcohol management interventions could make a substantial contribution to reducing alcohol-related harm in communities without the need for formal enforcement and without negatively impacting club income. CORRESPONDING AUTHOR: M. Kingsland, The University of Newcastle, Australia, [email protected] O679 BELIEFS ABOUT MEDICINES AND MEDICATION ADHERENCE IN PATIENTS WITH RHEUMATOID ARTHRITIS IN GERMANY S. Brandstetter1, N. Bartsch1, G. Riedelbeck1, M. Steinmann1, J Loss1, B. Ehrenstein2, C. Apfelbacher1 1 University of Regensburg, REGENSBURG, Germany 2 Asklepios Clinic, Department of Rheumatology and Clinical Immunology, BAD ABBACH, Germany Introduction: Adherence to medication is crucial to achieve treatment control in chronic disease. Little is known about psychological factors predictive of adherence in patients with rheumatoid arthritis (RA). We investigated the associations between beliefs about medicines and medication adherence in patients with RA. Methods: Patients with physician-diagnosed RA completed the “Beliefs about Medicines-Questionnaire”(BMQ) and the “Medication Adherence Report-Scale”(MARS). 362 patients recruited in inpatient and outpatient care (30.7% male, mean age 60.2 years (SD: 13.3)) provided informed consent and were included in the study. In order to predict non-adherence multivariate logistic regression analyses were computed for each BMQsubscale. All analyses were controlled for sex, sociodemographics and disease characteristics. BMQ-subscales were used as main explanatory variables. A binary variable derived from MARS was used as dependent

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

variable (full adherence vs. suboptimal adherence). All analyses were computed using SPSS.19 for Windows. Results: 73.9% of male and 62.9% of female patients were non-adherent. Non-adherence was predicted by general beliefs about medication overprescription by doctors (OR = 1.51; CI: 1.07-2.13), general beliefs about harm of medication (OR=1.52; CI: 1.05-2.19) and specific concerns regarding individual medication (OR=1.71; CI: 1.20-2.43). Beliefs about general utility and specific necessity were not significantly associated with adherence. Conclusion: The finding that negative, but not positive beliefs were weakly associated with adherence in RA may inform the development of targeted educational measures. CORRESPONDING AUTHOR: Dr. S Brandstetter University of Regensburg Germany [email protected] O680 STRUCTURAL EQUATION MODELLING TO PREDICT MEDICATION ADHERENCE IN RHEUMATOID ARTHRITIS OVER SIX MONTHS L.D. Hughes Institute of Psychiatry, King's College London, LONDON, United Kingdom Introduction: Medication for Rheumatoid Arthritis (RA) prevents irreversible joint damage but non-adherence levels remain at 40-75%. The Self Regulatory Model (SRM), Theory of Planned Behaviour (TPB) and Health Belief Model (HBM) have some utility individually to predict medication non-adherence in RA. This study used structural equation modelling to test all three models in parallel to establish the most predictive elements in this patient group for future intervention. Methods: A longitudinal questionnaire study was carried out with 171 adult RA patients. Validated questionnaires completed at baseline were used to predict self-reported adherence (5-item Compliance Questionnaire for Rheumatology) at six months using structural equation modelling. Results: The model showed a reasonably good fit and explained 24.7% of the variance in adherence; Χ2(766)=1166.75, p<0.001; CFI=0.86; RMSEA=0.06; SRMR=0.09. Higher baseline chronic timeline and consequences led to higher perceptions of medication necessity, improving adherence. Stronger emotional impact of RA led to higher medication concerns, reducing adherence. TPB important others and TPB perceived behavioural control (self-efficacy) explained an additional 4.4%. The Health Belief Model did not predict adherence. Conclusion: Illness and treatment beliefs predict medication adherence but are best utilised in conjunction with TPB factors including important others and self-efficacy which has not been previously demonstrated. This study is the first to use all three models to predict adherence and suggests that self-efficacy training, involvement of important others and targeting illness and treatment perceptions may improve medication adherence in Rheumatoid Arthritis, although consideration of the effect on mood and wellbeing is essential. CORRESPONDING AUTHOR: Dr L.D. Hughes, Institute of Psychiatry, King's College London, United Kingdom, [email protected] O681 ADHERENCE TO METHOTREXATE IN ARTHRITIS: IMPACT OF WRITTEN PATIENT INFORMATION ON MEDICATION BELIEFS AND CONCERNS C.F. Hayden1, C Tarrant1, W Hassan2 1 University of Leicester, LEICESTER, United Kingdom 2 University Hospitals of Leicester NHS Trust, LEICESTER, United Kingdom Introduction: Although methotrexate (MTX) efficacy and tolerability in the treatment of rheumatoid arthritis are well-proven, there were 165

S201

adverse incidents reported in 2006. This prompted the provision of safety-focussed information to those commencing on methotrexate. Written information is recognised as important for promoting adherence, but safety-focused information may disrupt medication beliefs and raise patient concern. This study aims to determine the impact of safetyfocussed written information on medication beliefs and subsequent influence on medication-taking behaviour. Methods: Consecutively-attending adult patients prescribed MTX were approached in out-patient clinic. 15 semi-structured interviews lasting around 1 hour were transcribed verbatim and analysed thematically. Results: Findings suggest patients struggle with adjustment to chronic disease and are often initially resistant to taking medication. Decisions to start MTX are influenced by healthcare professionals’ recommendations, but also reflect the patient adjusting to their condition and recognising necessity. Subsequently, patients experience dissonance between their concerns about MTX as a powerful drug and strong feelings of necessity. Whilst adherence is good, patients continuing on MTX need reassurance. Patients found leaflets empowering, enabling and legitimising, valuing them as a trusted source of information, but had unmet needs for information and support to manage expectations and concerns. Conclusions: This study improves understanding of how patients utilise and value written drug information and the impact on beliefs and adherence. This will guide clinicians and service providers in tailoring information to ensure patients are informed about safety risks without causing excessive concern and so reducing risk of non-adherence. CORRESPONDING AUTHOR: Miss C.F. Hayden, University of Leicester, United Kingdom, [email protected] O682 DO ILLNESS PERCEPTIONS AND MEDICATION BELIEFS PREDICT MEDICATION ADHERENCE IN ASTHMA? CROSSSECTIONAL AND PROSPECTIVE FINDINGS M.O.M. van de Ven, D. Tiggelman Behavioural Science Institute, Radboud University Nijmegen, NIJMEGEN, The Netherlands Introduction. Adherence to asthma preventer medication is low, especially in adolescents. Previous cross-sectional studies in adults with asthma have shown that illness perceptions and medication beliefs are associated with asthma medication adherence. However, prospective designs are needed to test whether perceptions and beliefs contribute to changes in adherence over time. This is the first known longitudinal study testing whether illness perceptions and medication beliefs not only relate to asthma medication adherence, but also predict changes in adherence over time. Method. Adolescents with asthma (n=177) using daily preventer medication completed the Beliefs about Medicines Questionnaire, the Brief Illness Perception Questionnaire, and the Medication Adherence Report Scale for Asthma in the spring/summer of 2011, 2012 and 2013. Path analyses (controlling for age, gender, asthma duration, asthma control) using Mplus6 were conducted. Results. Medication adherence significantly decreased over time. At baseline, strong associations between beliefs about asthma medication (i.e., necessity and concerns) and medication adherence, and weaker but significant associations between illness perceptions (i.e., timeline and treatment control) and medication adherence were found. However, illness perceptions and medication beliefs did not predict changes in medication adherence over time. Conclusions. In line with earlier cross-sectional studies in adults, results showed that illness perceptions and medication beliefs are related to medication adherence in adolescents with asthma. Although we might conclude from cross-sectional research that medication beliefs and illness perceptions are important to target in medication adherence interventions, longitudinal results showed no predictive effect of illness perceptions and beliefs on changes in adherence over time.

S202

CORRESPONDING AUTHOR: Dr. M.O.M. van de Ven, Behavioural Science Institute, Radboud University Nijmegen, The Netherlands, [email protected] O683 MEDICATION BELIEFS, TREATMENT COMPLEXITY, AND NONADHERENCE TO DIFFERENT DRUG CLASSES IN PATIENTS WITH DIABETES S.T. de Vries1, J.C. Keers2, R. Visser3, D. De Zeeuw1, F.M. HaaijerRuskamp1, J. Voorham1, P. Denig1 1 University of Groningen, University Medical Center Groningen, GRONINGEN, The Netherlands 2 Van Swieten Research Institute, Martini Hospital, GRONINGEN, The Netherlands 3 University of Groningen, UMCG, The LifeLines Cohort Study, GRONINGEN, The Netherlands Objective: To assess the relationship of patients' medication beliefs and treatment complexity with unintentional and intentional non-adherence for three therapeutic groups commonly used by patients with type 2 diabetes. Methods: Survey data about adherence (Medication Adherence Report Scale) and beliefs about medicines (Beliefs about Medicines Questionnaire) were combined with prescription data from the Groningen Initiative to ANalyse Type 2 diabetes Treatment (GIANTT) database. Patients were classified as being adherent, mainly unintentional non-adherent, or partly intentional non-adherent per therapeutic group (glucose-, blood pressure-, and lipid-lowering drugs). Treatment complexity was measured using the Medication Regimen Complexity Index, which includes the dosage form, dosing frequency and additional directions of taking the drug. Analyses were performed using Kruskal-Wallis and Mann-Whitney U-tests. Results: Of 257 contacted patients, 133 (52%) returned the questionnaire. The patients had a mean age of 66 years and 50% were females. Necessity beliefs were not significantly different between the adherers, mainly unintentional non-adherers, and partly intentional non-adherers (differences smaller than 5 points on a scale from 5 to 25). For blood pressurelowering drugs, patients reporting intentional non-adherence had higher concern beliefs than adherers (8 point difference, P = 0.01). Treatment complexity scores were lower for adherers but similar for mainly unintentional and partly intentional non-adherers to glucose- and blood pressure-lowering drugs. Conclusion: Treatment complexity was related to non-adherence in general. Beliefs about necessity were not strongly associated with nonadherence, while patients' concern beliefs may be associated with intentional non-adherence. However, the role of these determinants differs per therapeutic group. CORRESPONDING AUTHOR: S.T. de Vries, University of Groningen, University Medical Center Groningen, The Netherlands, s.t.de.vriesumcg.nl O684 DEVELOPMENT OF A MEASURE TO ASSESS ADHERENCE TO EXERCISE: THE EXERCISE ADHERENCE RATING SCALE (EARS) E.L. Godfrey, J. Weinman, N. Beinart King's College London, LONDON, United Kingdom Introduction:Exercise is an important way of treating chronic illness but adherence to exercise is often poor and there is currently no validated measure of it. The Exercise Adherence Rating Scale (EARS) is a new measure assessing adherence to prescribed exercise. Validation of the EARS was conducted using a study of patients with chronic low back pain in the UK. Methods:The EARS has 13-items generated from a systematic review and interviews with patients, physiotherapists and psychologists. Items are scored using a 5-point Likert scale (completely agree to completely

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

disagree), where a higher score indicates greater adherence. Construct validity was investigated using exploratory factor analysis (EFA) and convergent validity by correlating relationships between EARS factors and baseline pain and disability. Intraclass correlation coefficients were calculated to assess the test-retest reliability. Results:150 participants, mean age 50 and 60% female, were recruited from physiotherapist-led back rehabilitation classes at three hospitals in London and completed baseline demographic, pain and disability data and six week adherence data. EFA revealed a two factor solution explaining 47% of the variance in adherence to exercise. Sub-scale 1 (7 items) assesses negative adherence factors and sub-scale 2 (6 items) assesses positive adherence factors. Sub-scale 1 was found to be significantly associated with pain. Internal consistency was acceptable and test re-test reliability was high for both subscales. Conclusion:The EARS is a validated self-report measure that can discriminate between levels of adherence to exercise that has been prescribed by a health care professional. CORRESPONDING AUTHOR: Dr E.L. Godfrey, King's College London, United Kingdom, [email protected] O685 PROVIDING CARE TO A CHILD WITH CANCER: A LONGITUDINAL STUDY ON PARENTS' EXPERIENCES WITH CAREGIVING E. Sulkers, W.E. Tissing, A. Brinksma, P.F. Roodbol, W.A. Kamps, R.E. Stewart, J. Fleer, R. Sanderman University Medical Center Groningen, GRONINGEN, Nederland Introduction: Caring for a child with cancer is potentially stressful. Parents of children with cancer are confronted with several extra demands (e.g. responsibility for disease related tasks, dealing with role conflicts, and communication with the healthcare personnel). This may contribute to their already existing stress. The aim of the present study was to describe the course, predictors and impact of caregiving stress on the functioning of mothers of children with cancer during the 1st year post diagnosis. Method: Mothers (N=95, 86% response rate) of consecutive newly diagnosed pediatric cancer patients (0-18 years) completed measures of caregiving stress, depressive symptoms anxiety and self-reported health at diagnosis, and 3, 6, and 12 months thereafter. Results: Results indicated a significantdecreasein caregiving stress (especially during the first 3 months after diagnosis), and improvement of mothers’ functioning across the 12 months follow-up. Caregiving stress was predicted by marital status and the ill child being the mother’s only child. Multilevel analyses, controlled for sociodemographic and medical covariates, showed that, over time, the decline in caregiving stress was accompanied by a reduction in depressive symptoms and anxiety. The amount of variance explained by caregiving stress was 55% for depressive symptoms, 36% for anxiety, and 5% for self-reported health. Conclusions: The findings of the present study indicate that caregiving stress is an important factor related to maternal adjustment in pediatric oncology. This offers possibilities for developing interventions aimed at preventing caregiving stress, and strengthening parents’ confidence in their ability to provide good care. CORRESPONDING AUTHOR: E. Sulkers, University Medical Center Groningen, Nederland, [email protected] O686 CLINICAL EFFECT OF NEUROFEEDBACK TRAINING FOR CHILDREN WITH TIC DISORDER Y. Yang, S. Wang, L. Ma The Third Affiliated Hospital of Soochow, CHANGZHOU, China

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

Objective To investigate the effect of neurofeedback training for tics symptoms and electroencephalogram (EEG) spectrum change in children with tic disorder. Methods 19 patients with transient tic disorder (TTD), 20 with chronic motor or vocal tic disorder (CTD) and 15 with Tourette’s disorder (TS) received 40 sessions of neurofeedback training. The clinical effect was evaluated with Yale Global Tic Severity Scale, and brain activity improvement with EEG theta /beta, theta / sensorimotor rhythm (SMR) power ratio. Results The patients with three types of tic disorder all showed a reduction of tics symptoms (TTD 10.32±0.82 vs. 5.91±5.74; CTD 12.35±1.50 vs. 4.00±2.22; TS 25.33±2.80 vs. 13.73±4.22; P<0.01). Additionally, the patients showed theta decreasing and SMR enhancing after the sessions. The ratio of theta /beta (TTD 2.52±0.38 vs. 2.33±0.32; CTD 2.43±0.39 vs. 2.21±0.30; TS 2.64±0.35 vs. 2.39±0.26) and theta/SMR (TTD 3.28±0.44 vs. 3.09±0.38; CTD 3.27±0.51 vs. 2.98±0.34; TS 3.36±0.40 vs. 3.09±0.34)was significantly lower that before treatment (P <0.01). Conclusions Neurofeedback training is an effective and valuable treatment in tic disorder especially in the patients with transient tic disorder and mild chronic motor or vocal tic disorder. CORRESPONDING AUTHOR: Y. Yang, The Third Affiliated Hospital of Soochow, China, [email protected] O687 IMPROVING PHYSICAL PLAY IN CHILDREN WITH DISABILITIES: IDENTIFYING MODIFIABLE, CLINICALLY MEANINGFUL TARGETS FOR INTERVENTIONS N. Kolehmainen1, C. Ramsay2, L. McKee2, C. Owen3, J.J. Francis4 1 Newcastle University, NEWCASTLE, United Kingdom 2 University of Aberdeen, ABERDEEN, United Kingdom 3 NHS Lothian, LOTHIANS, United Kingdom 4 City University London, LONDON, United Kingdom Introduction: Participation in physical play (‘PPP’) is important for children with disabilities, but evidence-base for PPP interventions for them is scarce. The first step forward is to identify modifiable, clinically meaningful factors that can be targeted by interventions. Methods: A mixed methods study using the WHO framework International Classification of Functioning, Health and Disability (ICF). Participants: children (6-8yrs) with motor impairments, mobilising independently with/without equipment, seen by therapy services; and their parents. Self-reported PPP was assessed by Children’s Assessment of Participation and Enjoyment(CAPE). Data about modifiable impairment, activity, environmental, and personal factors were collected by therapists’ observations and parent questionnaires. Data were analysed using descriptive statistics and multiple linear regression. Results: Children’s (n=195) PPP [mean 18 times/week (IQR=11-25)] was mainly ‘recreational’ (e.g. pretend play, playing with pets) rather than ‘active physical’ (e.g. riding a bike/scooter). Parents (n=152) reported positive beliefs about children’s PPP but variable levels of family PPP behaviour. Therapists’ reported 23 unique impairments (e.g. muscle tone), 16 activity limitations (e.g. walking), and 3 personal factors (e.g. child’s PPP confidence). The final model, adjusting for family PPP behaviour, impairments in child’s psychomotor functions, and impairments in movement-related body structures, explained 19% of variation in PPP. Family PPP behaviour explained most of this variation. Conclusions: The results converge with other emerging evidence that interventions to increase PPP in children with disabilities should consider the behaviours of those around the child. The results also indicate that the clinical data currently collected has limited usefulness in explaining these children’s PPP. CORRESPONDING AUTHOR: N. Kolehmainen, Newcastle University, United Kingdom, [email protected]

S203

O688 LONG-TERM FOLLOW-UP OF A TELEPHONE-BASED INTERVENTION TO IMPROVE PRESCHOOLERS' HEALTHY EATING: A RANDOMISED CONTROLLED TRIAL R.J. Wyse1, L. Wolfenden1, E. Campbell2, L. Brennan3, K.J. Campbell4, A. Fletcher2, J. Wiggers2, J. Bowman5, T. Heard2 1 School of Medicine and Public Health, NEWCASTLE, Australia 2 Hunter New England Population Health, NEWCASTLE, Australia 3 School of Psychology, Australian Catholic University, MELBOURNE, Australia 4 Centre for Physical Activity & Nutrition Research, Deakin University, MELBOURNE, Australia 5 School of Psychology, University of Newcastle, NEWCASTLE, Australia Introduction: Telephone-based interventions have been found to have a positive long-term impact in improving the diet of adults. The aim of this study was to assess the long-term impact of a telephone-based intervention on the intake of fruit and vegetables and non-core foods in preschool aged children. Method: Parents were recruited from Australian preschools and randomly allocated to receive print materials and a brief, four contact telephone intervention, or to a print information only control group. A validated food frequency questionnaire (the Children's Dietary Questionnaire) was used to assess child diet at baseline and 12 and 18 months post intervention. Results: Fruit and vegetable subscale scores were significantly higher among children of the intervention group, indicating greater child fruit and vegetable intake, at the 12 month (16.77 vs. 14.89, p<0.01) but not the 18 month follow-up (15.98 vs. 16.82 p=0.14). There was no significant difference between groups at either follow-up period in non-core food subscale score. Conclusions: Further research to identify effective maintenance strategies are required to maximise the benefits of telephone-based interventions on child diet. CORRESPONDING AUTHOR: Dr R.J. Wyse, School of Medicine and Public Health, Australia, [email protected] O689 MENTAL HEALTH AND RESILIENCE IN ADOLESCENCE: A RESILIENCE-BASED INTERVENTION J. Dray1, M. Freund2, J. Bowman3, E. Campbell2, J. Wiggers2, L. Wolfenden2, R. Hodder2, K. Gillham4 1 Hunter New England Population Health/University of Newcastle, WALLSEND, Australia 2 Hunter New England Population Health;The University of Newcastle;HMRI, NEWCASTLE, Australia 3 Hunter New England Population Health; University of Newcastle, WALLSEND, Australia 4 Hunter New England Population Health; HMRI, NEWCASTLE, Australia Background: Research indicates that young people who are exposed to significant disadvantage or adversity are at greater risk of developing a range of negative developmental outcomes, including mental health problems. One suggested approach for supporting positive life outcomes in adolescents involves enhancing their resilience; the ability when faced with adversity to actively employ individual factors, and environmental factors, to return to a positive state of mental health and functioning. Research has suggested that high levels of resilience may promote positive mental health. Method: A web-based survey was undertaken by 7935 adolescents, aged 12-16 years, in Grades 7-10 across 32 disadvantaged secondary schools in Australia. The survey included the Strengths and Difficulties Questionnaire (SDQ) as a measure of risk of mental health problems, and the

S204

California Healthy Kids Survey (CHKS) as a measure of student resilience factors. The study is part of a randomised controlled trial of a multistrategy resilience-based intervention fostering internal and external resilience characteristics of participating students. Results: The presentation will report results from preliminary analysis of the relationship between resilience and risk of mental health problems. Key findings include, a significant association between external resilience and total SDQ score (X2(2, N=7935) =721.58, p<0.001), internal resilience and total SDQ score (X2(2, N=7935)=698.95, p<.0001), and overall resilience and total SDQ score (X2(2, N=7935)= 834.23, p<.0001). Conclusions: Understanding the relationship between adolescent resilience and mental health is fundamental to developing effective interventions for building adolescents awareness of social supports, and internal and external strengths that allow positive adaptation to adversity. CORRESPONDING AUTHOR: J. Dray, Hunter New England Population Health/University of Newcastle, Australia, [email protected] O690 EVALUATION OF A MULTIDISCIPLINARY AFTER-SCHOOL GROUP-BASED INTERVENTION TO IMPROVE WEIGHT STAT US IN OVERWEIGHT CHILDREN G.C. Kloek, Marie Pau Luger, Tinus Jongert, Sanne de Vries, Rob Oudkerk The Hague University of Applied Sciences, THE HAGUE, Nederland IntroductionThe intervention, called WIJS (Dutch for “What is Your Style”), is an after-school group-based programme aimed at overweight children aged 8-12 years, living in deprived neighbourhoods. During the school year, participants follow group sessions of 90-minutes including activity-based exercise, dietary education, and lifestyle counselling. This study evaluates the outcome of the programme on weight status in participants.MethodBody height and weight were measured with standardized protocols at baseline (T0), at the end of the programme (T1) and at follow-up (T2), 6 months later. BMI standard deviation scores (BMISDS) were calculated using age- and sex-specific criteria. Two groups were enrolled in the programme: group I at the start (n=72) and group II (n=98) halfway through the school year. Measurements were also conducted in 61 matched controls (with group II).ResultsBMI-SDS in group I at T0, T1 and T2 were 1.91, 1.82, and 1.92, respectively. BMI-SDS in group II at T0, T1 and T2 were 1.72, 1.64, and 1.66. BMI-SDS in the control group at T0, T1 and T2 were 1.44, 1.50, and 1.54. Participants in both groups achieved a BMI-SDS reduction at T1 (p<0.05). In group II compared to the control group the programme led to improvements in weight status directly after the intervention (p<0.05) and at follow-up (p<0.05).ConclusionsThe WIJS programme is an effective intervention to improve weight status in overweight children. However this improvement does not seem to be sustainable because participants gradually fall back in their weight status they had at the start of the programme. CORRESPONDING AUTHOR: G.C. Kloek, The Hague University of Applied Sciences, Nederland, [email protected] Fear of cancer recurrence S691 FEAR OF CANCER RECURRENCE M.F.M. Gielissen Radboud University Medical Center, NIJMEGEN, Nederland The proposed symposium addresses a neglected area in both research and oncology care, namely fear of cancer recurrence (FCR). FCR appears to be a universal problem and is one of the most prevalent unmet psychosocial need for help amongst survivors. Some survivors report more fear than others, levels of fear that are so high and overwhelming that they

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

have difficulties in performing their daily and social activities. This high FCR represents a form of subjective distress related to the medical situation like cancer itself, to recurrence or metastasis, follow-up care or periodic examinations. On the other hand, this fear is related to psychosocial concerns like relying on others to perform daily activities, worries about the future life, disability or death. In the literature, the percentages of cancer survivors with high levels of FCR are up to 50%. In this symposium background literature will be presented by discussing a systematic review (Ozakinci), more insight will be gained on how to measure FCR (Simard), and data of a retrospective cohort study (Custers) and longitudinal study (Mehnert) will be presented. Chair: Dr. MFM Gielissen Presenter 1: Dr. G. Ozakinci: Fear of cancer recurrence in adult cancer survivors: a systematic review of quantitative studies. Presenter 2: Dr. S. Simard: The challenge of the measurement of Fear of Cancer Recurrence Presenter 3: J. Custers (MSc): Fear of Cancer Recurrence in colorectal cancer survivors Presenter 4: Dr. A. Mehnert: Fear of recurrence in hematological cancers patients in the course of allogeneic hematopoietic stem cell transplantation CORRESPONDING AUTHOR: Dr. M.F.M. Gielissen, Radboud University Medical Center, Nederland, [email protected] S692 FEAR OF CANCER RECURRENCE IN ADULT CANCER SURVIVORS: A SYSTEMATIC REVIEW OF QUANTITATIVE STUDIES G. Ozakinci1, B. Thewes2, G. Humphris1, S. Simard3 1 University of, ST ANDREWS, SCOTLAND, United Kingdom 2 University of Sydney, SYDNEY, Australia 3 Institut universitaire de cardiologie et de pneumologie Québec,Laval University, QUÉBEC, Canada Gozde Ozakinci (1), Belinda Thewes (2), Gerry Humphris (1), Sebastien Simard (3) (1)University of St Andrews, UK (2) University of Sydney, Australia (3) Laval University Introduction:Fears of cancer recurrence (FCR) are among the most commonly reported problems experienced by cancer survivors and also among the most frequently reported unmet needs for survivors and their carers. We conducted a comprehensive systematic review of quantitative studies that summarises the current scientific knowledge on FCR and provide evidence-based recommendations for future research. Methods:We conducted a systematic review to identify quantitative studies using Medline, CINAHL, PsychINFO, and AMED databases (1996 to December 2011). We identified 130 eligible papers, summarised and synthesised its findings. Results:Studies show that survivors reported low to moderate levels of FCR but considered these fears as one of the greatest concerns and the most frequently endorsed unmet need. Younger age, experiencing physical symptoms, distress, and lower quality of life were associated with higher FCR. Having FCR had implications for health behaviours, psychological responses, and functional impairments. Moreover, FCR remains stable over time. Carers reported higher FCR than the patients. Limited data on interventions for FCR currently exist. Conclusions:This systematic review shows that FCR remains a worry for many cancer survivors, even in the long-term. There is a growing research interest in the area of FCR. However, there is little consensus in defining FCR. To assess the long-term impact of these fears, longitudinal studies are required. Developing and evaluating theoretically based FCR interventions for cancer survivors and their carers are needed CORRESPONDING AUTHOR: Dr. G. Ozakinci, University of United Kingdom, [email protected]

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

S205

S693 THE CHALLENGE OF THE MEASUREMENT OF FEAR OF CANCER RECURRENCE S. Simard1, B. Thewes2 1 Laval University, QUÉBEC (QC), Canada 2 University of Sydney, SYDNEY, Australia Sébastien Simard(1) & Belinda Thewes(2) (1) Institut universitaire de cardiologie et de pneumologie de Québec, (2) University of Sydney

diagnosis, stoma) variables. For psychological variables, high FCR was associated with more distress, fatigue and lower emotional well-being. Conclusions A long time after the diagnosis of colorectal cancer, FCR is a problem for a substantial part of survivors. Survivors with high levels of FCR experience significantly more psychological distress, functioning impairments and triggers compared to survivors with low levels of FCR.

Introduction: Fear of Cancer Recurrence (FCR) has become a major focus in cancer survivorship research. However, relatively few efforts have been made to summarize knowledge on this issue. This may be partly explained by the diversity of the assessment methods used. Our presentation summarises the current self-reports measures available and provides recommendations for future development. Method: Two systematic reviews were undertaken using Medline, CINAHL, PsychINFO, and AMED databases. The first one selected papers describing a FCR assessment instrument and including data on their psychometric properties. The second selected quantitative studies associated with FCR (1996-2011) and reported self-report measures used. The Medical Outcomes Trust criteria were used to examine the psychometric properties. Results: A total of 32 relevant scales were identified. Of those, FCR was assessed using purpose-designed longer scale (>10 items), brief FCR specific scales (2-10 items), FCR-specific subscales within quality-oflife or needs questionnaires. Several single items were also identified. The majority of self-report questionnaires demonstrated reliability and preliminary validity. Three longer specific questionnaires and one brief scale are particularly promising. Few efforts have been made to establish clinical criteria of FCR. Conclusions: Scientific community needs to adopt a consensual definition and consolidate the development of FCR assessment methods currently available. More research on what constitutes clinical level of FCR is needed. Rapid screening procedures with a clinical cut-off score are necessary to standardize research and clinical activity.

S695 FEAR OF RECURRENCE IN CANCERS PATIENTS IN THE COURSE OF ALLOGENEIC HEMATOPOIETIC STEM CELL TRANSPLANTATION A. Mehnert1, S. Sarkar2, A. Scherwath2, F. Schulz-Kindermann2, U. Koch2 1 University Medical Center Leipzig, LEIPZIG, Germany 2 University Medical Center Hamburg-Eppendorf, HAMBURG, Germany

CORRESPONDING AUTHOR: Dr. S. Simard, Laval University, Canada, [email protected] S694 FEAR OF CANCER RECURRENCE IN COLORECTAL CANCER SURVIVORS J.A.E. Custers, M.F.M. Gielissen, S.H.V. Janssen, J.H.W. de Wilt, J.B. Prins Radboud umc, NIJMEGEN, Nederland Introduction In general, colorectal cancer survivors report a good quality of life (QoL). However, they may encounter specific problems. One of the most commonly reported problems and important unmet psychosocial need is fear of cancer recurrence (FCR). For some survivors this fear becomes so overwhelming that they experience difficulties in performing daily activities. This study focuses on severity and characteristics of FCR in colorectal cancer survivors. Method Two hundred thirty six patients diagnosed with colorectal cancer in the period 2003 - 2010 from the Radboud University Medical Center in the Netherlands were asked to participate. An information letter was mailed to all colorectal cancer survivors together with questionnaires on demographic variables and medical data, FCR, QoL, distress, fatigue, coping and social support. Results One hundred six patients (45%) returned the questionnaires. Of these, 81 patients (median age 67.7; range 41-88 years) were disease free and did not have a recurrence or metastasis (median time since diagnosis 5.07; range 1.3-9.1 years). About a third of these patients (34.6%) experience high levels of FCR. No relation was found between high levels of fear and demographic (age, gender) or medical (time since

CORRESPONDING AUTHOR: J.A.E. Custers, Radboud umc, Nederland, [email protected]

Introduction: We examined the course of FCR and the prevalence of high FCR levels in patients undergoing allogeneic peripheral blood stem cell transplantation before conditioning (T0,N = 239), 100” days after (T1,n = 150), and 12”months after allogeneic HSCT (T2,n = 102). Predictors for high FCR and the association between FCR and quality of life were investigated. Methods: The Fear of Progression Questionnaire-Short Form (FoP-Q-SF), the EORTC QLQ-C30, and the Hospital Anxiety and Depression Scale (HADS) were used. Results: At T0, 36% of patients, at T1 15% of patients, and at T2 10% of patients fulfilled the criteria for high FCR (FoP-Q-SF cut-off = 34). FCR declined significantly over time (p < .001). Married patients had a significantly lower risk (OR = 0.37, p = .018) and patients with higher anxiety (OR = 11.36, p < .001) and depression (OR = 3.11, p = .021) had a higher risk for high FCR at baseline. At T1 higher baseline anxiety predicted high FCR (OR = 5.27, p < .01); at T2 female sex predicted high FCR (OR = 3.24, p = .031). FCR made a substantial contribution to the prediction of HRQoL (at T1 ΔR2 = 0.47 and at T2 ΔR2 = 0.38 for emotional functioning; at T1 ΔR2 = 0.10 and at T2 ΔR2 = 0.28 for physical functioning). Conclusion: Our data provide evidence that FCR is a prevalent problem in patients with hematological malignancies and has a significant adverse impact on HRQoL. CORRESPONDING AUTHOR: Prof. Dr. A. Mehnert, University Medical Center Leipzig, Germany, [email protected] Digital health promotion: innovations, evidence and directions S696 DIGITAL HEALTH PROMOTION: INNOVATIONS, EVIDENCES AND DIRECTIONS S.L. Lippke Jacobs Center, BREMEN, Germany Digital Health Promotion is an area of research, teaching, and practice concerned with the intersection of computing and health promotion. Key words like eHealth, iHealth, mHealth, ICT in medicine and health, online counseling, internet-based interventions, virtual interactive interventions are well known and widely applied to promote health. While increasing numbers of applications and studies are outperformed, it is important to review innovations, to test and compare their evidences and to discuss directions. With this goal this symposium brings together experts from different continents presenting research from various settings and populations, debating about theoretical backdrops, mechanisms and moderators. Brian Oldenburg (AUS) evaluates an automated interactive telephone system, the Australian Diabetes Telephone-Linked Care (TLC), regarding

S206

the improvement of diabetes self-management. He shares the results on behavior change and HbA1c decrease. Wayne Velicer (USA) uses Relational Agents, computer-based virtual characters in an intervention. He reports on whether these Relational Agents can establish a continuing personal relationship and increase participant engagement. Hein deVries (NL) reviews results from different RCTs testing computer tailored technology including videos and games. He presents findings on changes in smoking, alcohol, nutrition, and physical activity when addressed separately as well as in combination. Lilian Lechner (NL) reports on different modes of computer-based intervention (web-based versus print-based version). All interventions are tailored to the specific personal characteristics, needs and preferences of the individual user. The symposium chair Sonia Lippke (GER) invites the audience to discuss the value of these and other innovations, the need for specific evidences and future directions. CORRESPONDING AUTHOR: Prof. S.L. Lippke, Jacobs Center, Germany, [email protected] S697 HELPING PEOPLE WITH DIABETES TO IMPROVE THEIR SELF-MANAGEMENT OF DIABETES BY CONVERSING WITH A COMPUTER B. Oldenburg University of Melbourne, MELBOURNE, Australia We have developed an automated interactive telephone system, TLC Diabetes, to improve diabetes self-management. Participants upload their blood glucose results via cellphone and 'converse' on topics including blood glucose monitoring, nutrition, physical activity and medication taking. A randomized study of adults with sub-optimally controlled type 2 diabetes has demonstrated improved glycaemic control (mean HbA1c decrease from 8.7% to 7.9% for TLC condition vs 8.9% to 8.7% in control arm, p = 0.002). The program also improved mental health functioning. Program use was high with 80% of expected calls completed. Medication costs were also lower for the intervention group. Program evaluation shows how this program influences people to change multiple diabetes self-care behaviors. It provides people with very valuable information (“I thought I knew a lot but I still had a lot to learn”), it enhances motivation (“Good to know someone is motivating me each week”), it teaches new strategies (“Gave me the know how”) and it helps establish routines (“I became more disciplined”). It also leads to behavioral improvements in blood glucose testing, eating (“more aware of the food I eat”) and other actions to improve health (“I joined a gym”; “I now visit my GP more often”). It also had some important psychosocial effects (“I am socializing more”; “I was depressed before the program, but I am now doing OK”). The program enhances self-efficacy (“Speaking with a computer is good, it reminds you that you are managing diabetes yourself”). The presentation will discuss wider applications and potential scalability.

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

behaviors in a general population. Effectiveness has been demonstrated in multiple studies. However, these interventions have produced only small effect sizes, partially due to a lack of participant engagement. Relational Agents, computer-based virtual characters who can establish a continuing personal relationship, have the potential to increase engagement. This paper reports the Month 12 outcomes for the first randomized clinical trial study involving Relational Agents. Method. The design was a 3 (groups; Control, Expert System intervention, and Relational Agent (plus Expert System) intervention) x 3 (Occasions; Months 0, 12, and 24) with all intervention occurring during the first 12 months. A representative USA sample of 1639 individuals at risk for both exercise and UV protection was recruited. Both interventions employed the Transtheoretical Model and empirically based decision rules. Results. Month 12 data demonstrated that the Relational Agent Intervention increased participant engagement with the average number of sessions viewed per week of 0.048 in the Expert System group compared to 0.142 in the Relational Agents group. The addition of the Relational Agent also increased the satisfaction and was viewed as more helpful. Preliminary outcome data found significant differences between the three groups in the predicted direction on outcome for exercise (proportion at criteria for exercise was 13.6% in Control, 15.7% in Expert System and 16.6% in Relational Agent. Conclusions. The Relational Agent Intervention has the potential to outperform the previous best practice for computer-based interventions. CORRESPONDING AUTHOR: Professor W. Velicer, University of Rhode Island, United States of America, [email protected] S699 INNOVATIONS IN COMPUTER TAILORED E-HEALTH AND T H E I R A P P R E C I AT I O N , E F F E C T I V E N E S S , C O S TE F F E C T I V E N E S S A N D P O T E N T I A L E D U C AT I O N A L DIFFERENCES H. de Vries Maastricht University, MAASTRICHT, Nederland

S698 USING RELATIONAL AGENTS TO INCREASE ENGAGEMENT IN COMPUTER-BASED INTERVENTIONS: PRELIMINARY OUTCOMES W. Velicer1, C.A. Redding1, B. Blissmer1, S.F. Babbin1, A.L. Paiva1, T. Bickmore2, J.J. Johnson3 1 University of Rhode Island, KINGSTON, United States of America 2 Northeastern University, BOSTON, United States of America 3 ProChange Behavior Systems, WAKEFIELD, United States of America

Introduction: eHealth computer tailored technology (CTT) provides specific feedback to an individual and has shown to be effective. We have conducted several studies to test innovations, such as how to offer multiple behaviors (sequentially or simultaneously), the comparison of video and text tailoring, game-tailoring, and measured potential differences between groups differing in educational level. Methods: We will present the data of several RCT’s (N.10.000) that used eHealth CTT in order to analyze appreciation, behavioral outcomes, and cost-effectiveness. Results: Computer tailored technology resulted in significant behavior changes (p<.01) on smoking, alcohol, nutrition, and physical activity, when addressed separately as well as when addressed in combination. Video tailored approaches were significantly more appreciated, effective and cost-effective than text-based tailoring. The utilization of games provided interesting new potential. Lower educated respondents appreciated computer tailored technology, and mass media approaches were a more effective and cheaper way to attract them than via GPs. Finally, results of four studies assessed and demonstrated that computer tailoring was a cost-effective approach. Conclusions: Tailored eHealth programs are appreciated positively by lower and higher educated groups, sometimes even slightly better by lower educated adults. Programs were effective for both educational groups, indicating that the use of these programs are unlikely to contribute to increasing health disparities between the two groups. New strategies, such as video tailoring and game-tailoring can have interesting added value.

Introduction. Multimedia computer-tailored (Expert System) interventions represent a low cost, easily disseminated method of changing

CORRESPONDING AUTHOR: Prof. Dr. H. de Vries, Maastricht University, Nederland, [email protected]

CORRESPONDING AUTHOR: Professor B. Oldenburg, University of Melbourne, Australia, [email protected]

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

S700 COMPUTER TAILORED INTERVENTIONS TO ENHANCE PHYSICAL ACTIVITY: EFFECTS OF DIFFERENT APPROACHES L. Lechner1, D. Peels1, S. Friederichs1, A. Oenema2, H. De Vries2, C. Bolman1 1 Open University of the Netherlands, HEERLEN, Nederland 2 Maastricht University, MAASTRICHT, Nederland INTRODUCTION. Computer tailoring (CT) modifies an intervention to the specific personal characteristics, needs and preferences of the individual user. However, the way CT is applied and theoretically embedded varies substantially, possibly resulting in different effects. METHODS. Two CT interventions targeted at increasing physical activity (PA) will be presented. The ActivePlus intervention, targeting older adults, is based on theories such as Social Cognitive Theory, Theory of Planned Behavior and Self-Regulation Theory. In an RCT we compared the effects on PA of four versions of the intervention: an online web-based versus a print-based version and for both versions we tested the added value of supplementary environmental PA information. Further, we tested the I Move intervention, a web-based CT intervention aimed at increasing PA in adults, which is based on Self Determination Theory, SelfRegulation Theory, and integrates Motivational Interviewing techniques in online CT. In an RCT we compared I Move to a more traditional CT intervention, to test the additional PA effects of the more client centered I Move approach. RESULTS. All versions of the ActivePlus intervention were highly effective six months after baseline compared to a control group, while the print-based intervention remained effective up to one year after baseline. The I Move intervention proved effective in increasing short term PA compared to a control group. The six months effects, in comparison to a control group and a more traditional CT intervention, will be presented. CONCLUSIONS. CT is potentially effective in increasing PA. However, different approaches can lead to substantially different effects. CORRESPONDING AUTHOR: Prof. dr. L. Lechner, Open University of the Netherlands, Nederland, [email protected] Psychology of infertility and its treatment S701 PSYCHOLOGY OF INFERTILITY AND ITS TREATMENT C.M. Verhaak1, J. Boivin2, S. Gameiro2, Z. Donarelli3 1 Radboud University medical center, NIJMEGEN, Nederland 2 Cardiff University, CARDIFF, United Kingdom 3 Centro andros, PALERMO, Italy This symposium addresses crucial behavior medicine related issues regarding infertility and its treatment.. Focus of attention is the interplay between psychological factors (compliance, fertility specific distress, couple relationships, psychosocial vulnerability) and outcomes of fertility treatment in terms of pregnancy yes or no, as well as in terms of emotional adjustment to these treatment outcomes. The first presentations will focus on psychological factors and its relationship to treatment behavior as well as outcome. The second presentation will focus more on psychological processes regarding compliance to fertility treatment. To what extend do people comply to recommendations regarding fertility treatment and what are reasons and predictors for non compliance. In the third presentation, psychological risk factors for adjust problems to infertility in both women and men will be addressed with special attention to couple dynamics. The last presentation will focus on the feasibility of personalized psychosocial care in reproductive medicine from screening on psychosocial risk profile to tailored online psychological interventions. The symposium will end with a discussion about state of the art of behavioral medicine in reproduction and what should be next steps in research.

S207

CORRESPONDING AUTHOR: Dr C.M. Verhaak, Radboud University medical center, Nederland, [email protected] S702 WHY DO PATIENTS DISCONTINUE FERTILITY TREATMENT PREMATURELY AND HOW DO WE PREDICT DISCONTINUATION BEHAVIOUR? S. Gameiro1, C.M. Verhaak2, L. Peronace3, J. Boivin1 1 Cardiff University, UK, CARDIFF, United Kingdom 2 Radboud University Nijmegen, NIJMEGEN, The Netherlands 3 Merck Serono, S.A., GENEVA, Switzerland Chances of achieving parenthood are high for couples who comply with recommended fertility treatment. However, many discontinue before conceiving. A systematic review was conducted to investigate patients' stated reasons for and predictors of treatment discontinuation at five fertility treatment stages. Six databases were searched. Search-terms referred to fertility treatment and discontinuation. Studies reporting on patients' stated reasons for or predictors of discontinuation were included. A list of all reasons for discontinuation presented in each study was made, different categories of reasons were defined and the percentage of selections of each category was calculated. For each predictor, it was noted how many studies investigated it and how many found an association with discontinuation. 22 studies that sampled 21 453 patients were included. The most selected reasons for discontinuation were: postponement of treatment (39.18%), physical and psychological burden (19.07), relational and personal problems (16.67), treatment rejection (13.23%) and organizational (11.68%) and clinic (7.71%) problems. Some reasons were common across stages (e.g. psychological burden). Others were stage-specific (e.g. treatment rejection during workup). None of the predictors reported were consistently associated with discontinuation. There is a lack of longitudinal and theory led research to explain discontinuation from fertility treatment. Results from an ongoing prospective study using the Theory of Planned Behaviour to predict discontinuation will also be presented. Meanwhile, treatment burden should be addressed by better care organization and support for patients. Patients should be well informed, have the opportunity to discuss values and worries about treatment and receive decisional-aid about continuing treatment. CORRESPONDING AUTHOR: Dr. S. Gameiro, Cardiff University, UK United Kingdom, [email protected] S703 FERTILITY STRESS, BEHAVIOURAL RISK FACTORS, NONCOMPLIANCE AND FERTILITY TREATMENT OUTCOME IN A FIVE-YEAR PROSPECTIVE EVALUATION J. Boivin1, Lone Schmidt2, Carla Gameiro1 1 Cardiff University, CARDIFF, United Kingdom 2 University of Copenhagen, COPENHAGEN, Denmark Introduction: Emotional distress could have direct effects on the chance of pregnancy (not supported by past research) or indirect effects by increasing behaviours that reduce fertility (e.g., smoking, treatment discontinuation). The aim of the present prospective study was to evaluate the direct and indirect effects of fertility-related stress on the outcome of fertility treatment. Method: Two weeks before treatment 264 couples completed the Fertility Problem Stress which measures stress and disruption caused by the fertility problem. Baseline behavioural risk factors were extracted from the medical file and comprised smoking more than 10 cigarettes per day, drinking more than 14 units of alcohol per week (21 units for men), and body mass index more than 28 (women only). Non-compliance was measured as self-reported discontinuation of treatment. Treatment outcome was the occurrence of an ongoing pregnancy or live birth during the 5-year study period. Structural Equation Modelling (SEM) tested a base and extended model comprising all study variables.

S208

Results: Average age for women was 31.8 (SD=3.8) and men 34.4, (SD=5.0), and 31.4% of couples had had an on-going pregnancy or live birth with treatment). A SEM showed fit of the proposed model but stress and behavioural risk factors were not consistently associated with outcome (chisquare (15, N = 264) = 34.66, p = .003, RMSEA = .07, CFI = .94, NFI = .91). Conclusion: Stress and behavioural risk factors may be implicated in the outcome of treatment but associations are complex (direct, indirect) and have small effect sizes. CORRESPONDING AUTHOR: J. Boivin, Cardiff University, United Kingdom, [email protected] S704 ATTACHMENT AND ANXIETY IN COUPLES EXPERIENCING INFERTILITY: AN ACTOR PARTNER INTERDEPENDENCE ANALYSIS Z. Donarelli1, G. Lo Coco2, S. Gullo2, D.M. Kivlighan3, L. Salerno1, A. Volpes1, A. Allegra1 1 ANDROS Day Surgery Clinic, PALERMO, Italy 2 Department of Psychology, University of Palermo, PALERMO, Italy 3 Department of Counseling, Higher Education and Special Education, University of, MARYLAND, United States of America Introduction: There is a lack of dyadic research examining how infertile individuals’ attachment dimensions influence their infertility-related stress and anxiety. In the current study the actor partner interdependence model (APIM) was used to examine whether: (a) both males and females in a couple experiencing infertility who are high on attachment avoidance (AAv) and/or attachment anxiety (AAn) will experience high levels of stress (actor effects), (b) women’s attachment anxiety will be related to men’s infertility stress and men’s attachment avoidance will be related to women’s infertility stress (partner effects). Method: 385 couples undergoing IVF treatments at the Reproductive Medicine Unit of ANDROS Day Surgery Clinic (Palermo, Italy) were studied. Participants filled out the Experience in Close Relationships, the Fertility Problem Inventory, and the State-Trait Anxiety Inventory prior to start the treatment. Results: As hypothesized, significant actor effects for attachment dimensions and infertility-related stress, both for male (AAn: .67; AAv: 1.01, p<.05) and female (AAn: .67; AAv: .85, p<.05) members of the couple were found. Regarding the partner effects, we found a significant partner effect for both female attachment dimensions and male infertility stress (AAn: .16; AAv: .23, p<.05). Contrary to the hypothesis, the partner effect for male attachment avoidance and female infertility stress was not significant. Conclusio: The current study found that actor and partner effects are present in infertile couples, supporting the idea that individuals' attachment anxiety and avoidance can impact his or her infertility stress, and their partner’s reports of infertility stress. CORRESPONDING AUTHOR: PSY.D Z. Donarelli, ANDROS Day Surgery Clinic, Italy, [email protected] S705 TIMELY AND TAILORED IN COGNITIVE BEHAVIORAL THERAPY C.M. Verhaak, W. Nelen, J. Kremer, A. Dongen van Radboud University medical center, NIJMEGEN, Nederland Introduction: One out of thirty children is born after assisted reproduction. Still, about one in three women starting fertility treatment will not get pregnant. They have to adjust to the unfulfilled child wish. Research showed that most people are able to adjust well but about 25% shows severe adjustment problems continuing to exist six months after treatment. Pre-treatment wellbeing, cognitions regarding fertility problems and social support predict emotional adjustment. This provide the opportunity to identify patients at risk already in pre-treatment. In addition, it

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

allows offering psychological treatment tailored to the individual risk profile. A screening instrument is implemented as well as an online cognitive behavioral therapy treatment to offer timely tailored interventions in the infertile patients. Methods: A 32 item screening instrument was developed, validated and implemented in daily care. Patients scoring above the cut off for higher risk of emotional adjustment problems were invited to participate in the RCT on the effect of tailored online cognitive behavioral therapy. Patients were offered modules that corresponded to their personal risks. Results: 646 couples were screened, uptake rate was 78-80% and 93% of the patients recognized themselves in the risk profile. However, only 21% of the patients at risk planned to seek professional support for themselves. Online treatment was offered to 120 patients, 21% agreed to participate but dropped out prematurely. Conclusion: Screening of patients on risks for emotional adjustment problems is feasible, allowing for timely and tailored interventions. CORRESPONDING AUTHOR: Dr C.M. Verhaak, Radboud University medical center, Nederland, [email protected] S706 CLOSING THE GAP IN FERTILITY AWARENESS: PUBLIC AND PROFESSIONAL EVALUATIONS OF AN EMPIRICAL DECISIONMAKING TOOL B.F. Fulford1, L. Bunting2, J. Boivin1 1 Cardiff University, CARDIFF, United Kingdom 2 National Collaborating Centre for Cancer, CARDIFF, United Kingdom Introduction: The majority of adults want to have children. However an ever-increasing number of people are at risk of not achieving their parenthood goals due to factors that affect their fertility, including lifestyle and lack of fertility help-seeking. This study evaluated an empiricallyvalidated tool that allows individuals to assess their lifestyle, reproductive and biological risks for fertility problems and provides personalised guidance (the Fertility Status Awareness Tool; FertiSTAT). Method: A think-aloud protocol and qualitative interviews were conducted amongst women of reproductive age (n=14) and medical and healthcare professionals (n=10). Thematic analysis was performed. Results: In general women were aware that lifestyle during pregnancy can affect the baby but did not know that lifestyle could reduce their chances of pregnancy. Women and medical/healthcare professionals agreed that fertility and lifestyle choices were difficult to discuss during consultations without patients feeling stigmatised. Professionals believed the FertiSTAT would facilitate a much needed dialogue with patients about fertility health and reduce the stigma felt by patients. Women felt the personalised information provided by the FertiSTAT was non-judgemental and empowered them to discuss fertility concerns with their doctor. Conclusions: Given considerable knowledge gaps in the population about risk factors for fertility problems, the FertiSTAT may be a timely tool to promote awareness and informed decision-making about fertility. The FertiSTAT could be used in medical practice to facilitate discussions about fertility health in a way that is seen by patients as scientificallyjustified and non-prejudiced. Research should investigate the impact of the FertiSTAT on fertility choices and outcomes. CORRESPONDING AUTHOR: Miss B.F. Fulford, Cardiff University, United Kingdom, [email protected] Behavioral health economics and the development of behavioral interventions S707 BEHAVIORAL HEALTH ECONOMICS AND THE DEVELOPMENT OF BEHAVIORAL INTERVENTIONS P. Brown Univ California Merced, MERCED, United States of America

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

Behavioral economic research has been touted as having the potential to lead to new approaches to interventions and policies aimed at influencing health-related behaviors such as exercise, weight loss, and medication adherence. The symposium presents an overview of developments in behavioral health economics as it applies to interventions aimed at promoting physical activity in older adults and attempts to identify new insights that behavioral economics is providing to behavioral medicine researchers. The basis of the two presentations are the results from systematic reviews of two literatures: behavioral interventions aimed at promoting physical activity interventions in older adults and behavioral interventions that draw upon behavioral economics. The first paper by Associate Professor Denise Taylor provides the results from a systematic review of the first literature, highlighting the most commonly used methods for encouraging adherence in this area. The second paper by Professor Paul Brown describes the results of a systematic review of interventions based on behavioral economics, with the emphasis being on the use of financial incentives and behavioral 'nudges'. The session concludes with a roundtable discussion led by Professor Linda Cameron and Professor Christina Lee on the implications of behavioral economics for interventions aimed at promoting exercise in older adults and insights from a health psychology perspective for behavioral medicine researchers. CORRESPONDING AUTHOR: Professor P. Brown, Univ California Merced, United States of America, [email protected] S708 INTERVENTIONS FOR IMPROVING THE HEALTH OF OLDER ADULTS: PHYSICAL ACTIVITY INTERVENTIONS PROMOTING LONG-TERM BEHAVIOR CHANGE D. Taylor Auckland University of Technology, AUCKLAND, New Zealand Background: Physical inactivity is a major contributor to mortality and morbidity. Research emphasizes the direct link between physical inactivity, low cardiovascular fitness and the presence of chronic health conditions. With increasing age, the risk of developing non-communicable chronic health conditions increases. Physical Activity Guidelines have been published globally, yet these have not resulted in significant changes in activity. This is especially true for adults aged 75 years and older, where only 9% of men and 6% of women achieve the recommended guidelines. The most common reasons given by older adults for not participating in physical activity are ill- health, pain and injury, and a lack of knowledge about appropriate exercise. Method: A literature review was undertaken to determine behavioral interventions aimed at promoting physical activity interventions in older adults. Different approaches are categorized according to methods to promote adherence to the program during and after the formal intervention period. Cinahl, Scopus, Medline and PeDRO databases were searched using terms, older adults, physical activity intervention, randomized controlled trial and adherence. Results: A total of 7 publications fitted the selection criteria. Of these ,all just reported percentage adherence as a reflection of participation in the exercise program, with limited follow-up to determine long-term behavioral change. The most commonly used methods of encouraging adherence were calendar/diary use, exercise logs and reminder telephone calls. Discussion: Consideration of methods of structuring physical activity interventions for older adults to encourage ongoing physical activity is limited, with little attention paid to changing long-term behavior. CORRESPONDING AUTHOR: Dr D. Taylor, Auckland University of Technology, New Zealand, [email protected]

S209

S709 INTERVENTIONS PROMOTING HEALTH BEHAVIOR CHANGE USING BEHAVIORAL ECONOMICS P. Brown Univ California Merced, MERCED, United States of America Background: Behavioral economists have categorized behavioral interventions as utilizing one or more of four types of techniques: Restrictions/ constraints on choice, marketing/persuasion, incentives/disincentives, and choice architecture involving 'nudges'. Proponents of choice architecture suggest that individuals can be 'nudged' toward making the decisions that are in their 'best interest' without limiting free choice ('libertarian paternalism'). Nudges are seen as being useful for promoting healthy behaviors such as exercise, weight loss, and medication adherence. This paper reviews evidence on the use of financial incentives and nudges to promote health behavior change. Method: A systematic literature review identified behavioral interventions that either use financial incentives or choice architecture/nudges to motivate health behavior change. Cinahl, Scopus, Medline, Google Scholar, Cochran and PeDRO databases were searched using terms behavior change, financial incentives, nudges, behavioral economics, and choice architecture. Results: A total of 17 studies fit the selection criteria. Of these, the use of financial incentives was most common. Those studies using choice architecture included reference to a number of common heuristics, with the most commonly used being anchoring and availability heuristics. This included the use of defaults/automatic enrolments, loss aversion, and strategic use of decision points to influence behavior. Discussion: The use of incentives or nudges to influence behavior, including those aimed at boosting self control or desired behavior, providing external motivation, increasing deliberative decision making or reducing cognitive overload, or discourages competing activities, are grounded in expected utility theory and incorporate impulsive and deliberative decision making. CORRESPONDING AUTHOR: Professor P. Brown, Univ California Merced, United States of America, [email protected] O710 COGNITIVE BEHAVIORAL THERAPY FOR THE TREATMENT OF ANTENATAL SYMPTOMS OF DEPRESSION AND ANXIETY H. Burger1, J. Ormel1, T. Verbeek1, M.G. Van Pampus1, J.L. Meijer1, C.G.H.M. Beijers1, C.L.H. Bockting2 1 University Medical Center Groningen, GRONINGEN, The Netherlands 2 University of Groningen, GRONINGEN, The Netherlands Introduction. Around 10-20% of pregnant women suffer from antenatal symptoms of depression or anxiety. They are distressing to the women themselves and they form potentially modifiable risk factors for health problems in the offspring . Evidence of effectiveness of cognitive behavioural therapy (CBT) in reducing these symptoms is sparse and conflicting. Method. A pragmatic randomized controlled trial was conducted among 226 unreferred pregnant women with at least moderate levels of anxiety or depression (State Trait Anxiety Inventory (STAI) >42 or Edinburgh Postnatal Depression Scale (EPDS) >11) at the end of the first trimester (baseline). Women were 1:1 allocated to either 6-10 antenatal sessions of CBT or care as usual (CAU). The outcome of the present analysis was the between group differences in mean scores on the EPDS and the STAI at 36 weeks of gestation, adjusted for baseline. Results. No marked inequalities in baseline characteristics were observed between the CBT (N=115) and CAU (N=111) group. The mean (SD) baseline score on the EPDS was 10 (4), on the STAI 48 (9). In both groups, 27% dropped out during the trial. At 36 weeks of gestation, differences between the CBT and the CAU group in EPDS and STAI scores were nihil, (0.0 (95%CI: -1.2;1.2; p=0.99) and -0.6 (95%CI: -

S210

3.9;2.6) p=0.69), respectively. Sensitivity analyses using multiple imputation for dropouts showed similar results. Conclusions. CBT is not effective in reducing moderate antenatal symptoms of anxiety and depression. Future research may address the background of these findings as well as effects on child outcomes. CORRESPONDING AUTHOR: H. Burger, University Medical Center Groningen, The Netherlands, [email protected] O711 THE ROLE OF EXPECTATIONS IN CONCENTRATION/MEDITATION TRAINING EFFECTS ON THE RESPONSE TO HUMAN ENDOTOXEMIA H. van Middendorp1, M. Kox2, P. Pickkers3, A.W.M. Evers1 1 Leiden University / Radboud university medical center, LEIDEN, Nederland 2 Radboud university medical center, Department of Anesthesiology, NIJMEGEN, Nederland 3 Radboud university medical center, Department of Intensive Care, NIJMEGEN, Nederland Introduction:Based on preliminary findings of a controlled inflammatory response to human endotoxemia (lipopolysaccharide (LPS-)administration) by means of a concentration/meditation technique (Kox et al., 2012), a randomized-controlled pilot study in 24 healthy young men has very recently shown that this technique can be trained and can induce a reduced inflammatory and symptom response to LPSadministration (Kox et al., under revision). Evidence is growing that psychological processes such as expectations can impact physiological stress systems, including the immune system. We explored the potential role of expectations in this pilot study. Method:Twelve healthy young men received a concentration/meditation training (experimental group) and 12 received no training (control group). Before, during, and after the training, expectations of the impact of the training on health and LPS-administration were assessed. Results:The experimental group expected fewer physical symptoms than the control group to the LPS-administration (t(22)=8.12) and showed increased expectations of the training on their response to the LPSadministration from before to after the training (F(2,22)=9.08). Expecting more physical symptoms to the LPS-administration was associated with a higher symptom report in the experimental group (r>.25), as well as a higher TNF-α level and lower IL-10 level in both groups together (r>.40). Participants showing higher expectations of the training on their health reported fewer symptoms during the LPS-administration (r>.50). Conclusions:Preliminary indications are provided that expectations are related to the physical complaints and immune response to an inflammatory stimulus. When verified in larger samples, interventions might be strengthened by explicitly focusing on optimized adequate expectation trainings. CORRESPONDING AUTHOR: H. van Middendorp Leiden University / Radboud university medical center, Nederland, [email protected] O712 PROGRESSIVE MUSCLE AND BIOFEEDBACK RELAXATIONS' POTENTIAL FOR RAISING STUDENTS' DISTRESS TOLERANCE AND LOWERING PERCEIVED STRESS G. Jarasiunaite1, I.P. Peciuliene2, A.P. Perminas3 1 Vytautas Magnus University, KAUNAS, Lithuania 2 Ieva, KAUNAS, Lithuania 3 Aidas, KAUNAS, Lithuania Introduction: Students are a social group that, despite general sources of stress, experience some specific challenges such as adaptation to a new place of residence, new social network and a load of academic requirements. Individuals who have lower distress tolerance also have more difficulties adapting to new changes and challenges that they face. In

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

stressful situations such people tend to react with impulsive behavior which leads to abuse of psychotropic substances, overeating, physical violence, avoidance of situations such as class attendance and taking exams. This study aims to assess the potential of progressive muscle relaxation and skin-conductance biofeedback relaxation in order to raise distress tolerance and lower perceived stress in students’ population. Methods: 65 undegraduate university students aged between 18 and 23 participated in the study. The participants of this study were randomly assigned to 3 different groups. Two groups received 4 progressive muscle relaxation (23, 35.4% and skin-conductance relaxation (23, 35.4%) training sessions once a week between two measurements of their distress tolerance and level of perceived stress and the third group was a control group (19, 29.2%). Distress Tolerance Scale (Simons, Gaher, 2005) was used to evaluate distress tolerance. Subjects completed Perceived Stress Scale (Cohen, 1983) to measure their perceived stress level. Results and conclusion: The results of the study showed that distress tolerance increased in students who had attented skin-conductance biofeedback training, while perceived stress lowered in students who had had progressive relaxation training. CORRESPONDING AUTHOR: G. Jarasiunaite, Vytautas Magnus University, Lithuania, [email protected] O713 REDUCING WORRY AND SOMATIC HEALTH COMPLAINTS: EXAMINING THE EFFECTIVENESS OF AN INTERNETDELIVERED WORRY POSTPONEMENT INTERVENTION A. Versluis, B. Verkuil, J.F. Brosschot Leiden University, LEIDEN, Nederland Introduction: Several studies have shown that perseverative, worrisome thoughts are prospectively related to somatic complaints and that a short worry intervention can be effective in reducing these complaints in children, adolescents and patients suffering from severe work stress. As worry and somatic complaints are prevalent and costly, we attempted to replicate these findings in an adult sample. We tested whether postponing worry can be used to reduce worry and somatic complaints in the general population. Methods: Participants were recruited in the Dutch population (N=1037). They were instructed to register the frequency and duration of their worry episodes online for 6 consecutive days. Half of the participants were additionally instructed to postpone their worrying to a special 30 minute-period in the early evening. The Subjective Health Complaints (SHC) questionnaire was administered before and after the worry postponement intervention. Data was analyzed using multilevel modeling. Results: Three hundred and forty-nine participants completed the study. Contrary to our expectation, the registration group (n = 186) did not differ from the postponement group (n = 163) in the amount and duration of worry episodes, or in the amount of health complaints at the end of the study. Conclusions: In contrast to previous studies, the current study showed that a worry postponement intervention was not effective in reducing worry and somatic complaints. Given the high burden associated with worry and health complaints in the general population, there is still need for short and evidence-based interventions. CORRESPONDING AUTHOR: A. Versluis, Leiden University, Nederland, [email protected] O714 LISTENING TO MUSIC IS GOOD FOR YOUR HEALTH - FINDINGS FROM TWO MOMENTARY ASSESSMENT STUDIES A. Linnemann1, B. Ditzen2, J. Strahler1, J.M. Doerr1, U.M. Nater1 1 University of Marburg, MARBURG, Germany 2 University of Zurich, ZURICH, Switzerland Introduction Music has been used as an adjuvant treatment option in various health care settings due to its proposed ability to reduce stress.

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

This stress-reducing effect appears in controlled laboratory settings, but it is unclear whether it can be observed in everyday life. We tested the potentially stress-reducing effect of music in two ecologically valid momentary assessment studies. Methods Study 1 (N=56) tested the effects of music on subjective stress levels in daily life on seven consecutive days. Study 2 (N=55) longitudinally investigated the effect of music on stress parameters during two periods of varying stressfulness (five days each). In both studies, healthy university students rated their perceived stress and current music listening behavior six times per day. Additionally, a sub-sample (n=25) from study 2 provided for two days saliva samples at each time point for the later analysis of the stress hormone cortisol. Results Hierarchical linear modeling revealed that in both studies music listening led to a reduction of stress (p<0.05, p<0.001), but only if participants intended to use music to relax. Study 2 found that music listening was associated with lower stress, but only during the less stressful period (p<0.05). Music listening led to decreases in cortisol, but, again, only if participants intended to use music to relax (p<0.05). Conclusions Music listening appears as effective means of stress reduction in daily life. Both researchers and clinicians in behavioral medicine should be aware of the possibility to reduce stress by asking patients to listen to music in daily life. CORRESPONDING AUTHOR: A. Linnemann, University of Marburg, Germany, [email protected] O715 STRESS WITHOUT DISTRESS: META-ANALYSIS OF THE PHYSIOLOGICAL STRESS RESPONSE TO VIDEO GAMING T. Nagy, Z. Kekecs, G. Bárdos, M. Fülöp Eotvos Lorand University, BUDAPEST, Hungary Background: In recent decades video games were occasionally used to study the stress response, although it is not clear whether video gaming can be considered a source of distress. The present study aimed to examine whether exposure to video games yield comparable physiological response to standard stressors, such as mental arithmetic task, public speaking, or the cold pressor test. Method: Literature search was conducted on major online databases to identify studies on the effects of video gaming on physiological parameters. Measures of sympathetic and parasympathetic activity, as well as heart rate and cortisol level were used as outcome variables. Random effects meta-analysis was performed to estimate the pooled effect sizes of video gaming on physiological variables. Results: Video games increased sympathetic activity and heart rate, and decreased parasympathetic activity. However cortisol levels did not show to increase in response to video gaming. Conclusion: A standard characteristic of laboratory stressors is that they elicit cortisol increase. As video gaming does not seem to provoke a cortisol response reliably, we conclude that video gaming should not be used as a model for distress. On the other hand, video gaming might prove to be a useful model for the research of eustress. CORRESPONDING AUTHOR: T. Nagy, Eotvos Lorand University, Hungary, [email protected] O716 ACUTE STRESS DISORDER SYMPTOMS AND ALL-CAUSE MORTALITY AMONG MYOCARDIAL INFARCTION PATIENTS: 14 YEAR FOLLOW-UP K. Ginzburg Tel Aviv University, TEL-AVIV, Israel Introduction: There is evidence that about 15% of myocardial infarction (MI) patients endorse acute stress disorder (ASD) in the immediate aftermath of MI, manifested by dissociation, intrusion, avoidance, and

S211

hyperarousal symptoms. These symptoms were shown to be risk factors for long term distress and impaired quality of life. Yet, although there are indications that long term distress affects prognosis in MI patients, no study to date has evaluated the prediction of ASD symptoms, expressed during hospitalization, for long term mortality. Method: 193 MI patients filled out a battery of self-report questionnaires during the first week following MI. Clinical data evaluating MI severity was obtained from patients’ hospital records. Endpoint was all-cause mortality at 14-year follow-up. Results: 42 (21.8%) of the MI patients died during follow-up period. Mortality was associated with higher levels of acute dissociation (p<.001) and hyperarousal (p<.05). Adjusted Cox proportional hazards regression analysis indicated that the relative mortality risk (multivariate adjusted for sex, age, level of education, clinical characteristics, and depression) hazard ratio was 2.07 (95% confidence interval 1.29-3.35, p<.01) for acute dissociation. Conclusions: The findings suggest that dissociative symptoms in the immediate aftermath of MI constitute risk factor for mortality, thus pointing to the prognostic benefit in ASD screening during hospitalization following MI. CORRESPONDING AUTHOR: Prof. K. Ginzburg, Tel Aviv University, Israel, [email protected] O717 FACTORS ASSOCIATED WITH PRE-HOSPITAL DELAY IN PATIENTS WITH ACUTE MYOCARDIAL INFARCTION A.H. Madani1, Hossein Farshidi2, Shafei Rahimi2, Abdi Ahmadnoor2, Sarah Salehi2, Abdoulhos Madani2 1 Research Center for Social DeterminantsHormozgan University of Medical Sciences, BANDAR ABBAS, Iran 2 Research Center for Cardiovascular, Di ,Hormozgan University of Medical Sciences, BANDAR ABBAS, Iran Introduction: Treatment of patients with acute myocardial infarction (AMI) is time related, so delay in treatment could affect prognosis. Recognizing pre-hospital or in-hospital delays in initiating treatment and reducing these factors is very efficacious in treatment of these patients.The aim of this study is evaluate the causes of pre-hospital delay just as other studies on effect of different variables such as socioeconomic and personal factors on pre-hospital delay in with patients with AMI. Methods:A cross sectional study was carried out on 227 patients with acute myocardial infarction and demographic data, educational level, marital status, type of transfer to hospital and delay in arrival to hospital were recorded. Results: 35.7% patients arrived during one hour of symptom onset, and 7.9% arrived after 24 hours. Patients having high level education (P = 0.0492) and with a family history of coronary artery disease (P = 0.01) had significantly less delay in arriving to hospital. Age, marital status, gender, and route of transfer to hospital were not related with pre-hospital delay (P > 0.05). Patients thought most common cause of delay in arrival was unawareness of coronary artery disease (38.8%) and self-medication (34.3%). Conclusions:Increasing awareness of patients about cardiovascular symptoms and their risk factors could be helpful in patient's decision in seeking medical help. So general education via media and primary and middle schools could be helpful. CORRESPONDING AUTHOR: Professor A.H. Madani, Research Center for Social DeterminantsHormozgan, University of Medical Sciences, Iran, [email protected] O718 AFFECTIVE OUTCOMES IN THE SUPRIM STRESS MANAGEMENT PROGRAM FOR POST-MYOCARDIAL INFARCTION PATIENTS G. Burell, F. Norlund, E. Olsson, P. Pavulans, K. Svardsudd Uppsala University, UPPSALA, Sweden

S212

Background. The Interheart Study showed that psychosocial stress increased the risk of myocardial infarction (OR 2.67). The Secondary Prevention in Uppsala Primary healthcare project (SUPRIM) tested whether stress management after myocardial infarction could decrease risk of recurrent cardiovascular events. SUPRIM was a randomized controlled trial in which CHD patients attended a CBT group stress management program. During 94 months of follow-up there was a 45% reduction of recurrent myocardial infarctions (MI) in the intervention group. There were no differences between intervention and control patientes in standard risk factors. Thus, we assumed that psychological mechanisms could explain the effects on hard endpoints. Method. 362 MI patients were assessed by a package of self-rating psychometric instrument, and this paper examined whether hostility, time urgency, depression, somatic anxiety, and vital exhaustion decreased more in the intervention group than in the control group. Data collected from 5 points of measurement over 24 months were analyzed with multi-linear regression. Results. The intervention had a significant effect on hostility, time urgency, and somatic anxiety. Vital exhaustion and depression scores showed no differences. There was a significant relationship between attendance rate and decrease of hostility, but no association between attendance rate and levels of vital exhaustion or depression. Conclusions. Treatment effects were demonstrated in measures that assessed emotional reactivity. These changes are possibly linked to physiological and hormonal mechanisms. However, it is to a considerable degree still unknown why and how the stress management intervention impacts disease endpoints. CORRESPONDING AUTHOR: Ph.D. K. Burell, Uppsala University, Sweden, [email protected] O719 REDUCED MOTIVATION AND MENTAL FATIGUE EFFECTS COGNITIVE FUNCTIONING AFTER ACUTE CARDIAC EVENTS J.B. Burkauskas, Julija Brozaitiene, Robertas Bunevicius Lithuanian University of Health Sciences Behavioral Medicine Institute, PALANGA, Lithuania Introduction. Contribution of fatigue to the cognitive functions comprising mental capacity of coronary artery disease (CAD) patients remains to be identified. Method. A 5 year study included 722 CAD patients two weeks after acute myocardial infarction or unstable angina; 529 (73%) men and 193 (27%) women; mean age of 58 years (SD=9). Patients were evaluated for sociodemographic characteristics, left ventricular ejection fraction (LVEF), New York Heart Association (NYHA) functional class. The Multidimensional Fatigue Inventory (MFI-20) was used to assess mental fatigue and reduced motivation. Digit Span Test, Digit Symbol Test, Trail Making Test A and B were used to assess cognitive functioning. Participants were considered cognitively impaired or markedly fatigued if they fell below the 25th percentile of the study population in the specific cognitive function and fatigue tests. Multiple logistic regression models were used to evaluate effects of reduced motivation and mental fatigue on impairment in each of the cognitive tests. Results. Controlling for potential confounders such as age, gender, education, NYHA class and LVEF, reduced motivation was associated with a 1.56-fold increase (95% CI, 1.05~2.32) in risk for impairment in test completion time. CAD patients with marked mental fatigue respectively had 1.92 (95% CI, 1.30 to 2.84); 1.56 (95% CI, 1.06 to 2.31) higher odds for impairment in incidental learning and executive control compared to patients with less severe mental fatigue. Conclusions. In CAD patients two weeks after acute cardiac events reduced motivation and mental fatigue effect selective impairment in test completion time, incidental learning and executive functioning. CORRESPONDING AUTHOR: MSc J.B. Burkauskas, Lithuanian University of Health Sciences Behavioral Medicine Institute, Lithuania, [email protected]

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

O720 LONGITUDINAL CHANGE IN CARDIOVASCULAR RISK FACTORS OVER SEVEN YEARS IN RURAL KERALA, INDIA T. Sathish1, S. Kannan2, P.S. Sarma2, O. Razum3, K.R. Thankappan2 1 Monash University, MELBOURNE, Australia 2 Sree Chitra Tirunal Institute for Medical Sciences and Technology, TRIVANDRUM, India 3 Bielefeld University, BIELEFELD, Germany Introduction: Cognitive impairment and physical limitation in older people are often correlated. We investigate the association of cognitive impairment with functional dependence, and the role of cardiovascular diseases (CVDs) and atherosclerosis in their association among Chinese elderly people. Methods: This cross-sectional study included 1526 participants (age ≥60 years, 59.1% women) of the Confucius Hometown Aging Project in Shandong, China. From June 2010 to July 2011, data were collected through interviews (e.g., demographics, lifestyles, and medical history), clinical examinations (e.g., CVDs and carotid artery plaques), and laboratory tests (e.g., blood glucose and serum lipids). Global cognitive function was assessed with the Mini-Mental State Examination, and the education-specific cutoffs were employed to define cognitive impairment. Physical function was assessed with basic activity of daily living (ADL) and instrumental ADL (IADL). Data were analyzed with multiple logistic models. Results: The overall prevalence was 5.9% for cognitive impairment, 15.5% for only IADL-dependence, and 2.2% for any ADL-dependence. Controlling for demographic features and major cardiovascular risk factors, the likelihoods of cognitive impairment were 4.40 (2.69-7.17) for only IADL-dependence and 6.24 (2.46-15.82) for any ADL-dependence; the corresponding figures were 4.36 (2.67-7.12) and 6.06 (2.39-15.40) when further controlling for carotid plaques, and 4.06 (2.48-6.64) and 4.25 (1.60-11.31) when CVDs (stroke, coronary heart disease, and heart failure) were added to the model. Conclusions: Cognitive impairment is associated with functional dependence among Chinese elderly people, and the association with ADLdependence is partly attributable to atherosclerotic diseases, especially CVDs. CORRESPONDING AUTHOR: Dr T. Sathish, Monash University, Australia, [email protected] O721 HOW ARE EFFECTS OF LOW SES MEDIATED? K. Orth-Gomer, C. Weber, C. Herrman-Lingen, C. Albus, H.C. Deter Karolinska Institutet, STOCKHOLM, Sweden Background Psychosocial factors tend to cluster together. Thus, interventions which reduce mortality risk may use multiple pathways of action. Depression effects are enhanced in low as compared to high socioeconomic status (SES) and cardiac mortality is high in low SES and in depression. Low heart rate variability (HRV), due to autonomic cardiac dysfunction, has been proposed as one of the mediating pathways that might explain the increased mortality associated with depression in patients with coronary heart disease (CHD). We evaluated the relationship of SES with HRV MethodsThe SPIRR-CAD trial, is a large multicenter trial which screened and randomized German patients with coronary disease to either a stepwise psychosocial intervention to reduce depression or to care as usual. All 570 patients (450 men; 120 women) were mildly (or severely) depressed, i.e. scored 8 or higher on the HADS. In the SPIRR-CAD trial, SES (education), HRV and depressive symptoms (HADS, PHQ, HDRS) were evaluated at baseline (t0) and at 18-month assessment (t3). Here baseline results are reported. Time domain measures (SDNN), standard deviation of the beat to beat interval and frequency domain measures (RMSSD) root mean square of successive differences, were recorded with the “polar watch”before, during and after

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

an aritmethic stress task. SES was classified into low: 8-9 years of schooling (46%), middle: 10-12 years (30%) and high: 13 years or more in school (24%). Results At baseline, depression scores were highest in the lowest SES group and vice versa (HADS 11.5+/-3.4 vs 10.1 +/- 2.4 p=0.04) and so was vital exhaustion (91% of low SES vs 61% of high SES patients were exhausted p=0.004). Low SES was also associated with financial difficulties, with social isolation, with smoking. In low SES, 57% of men, but only 31% of women had had an AMI p<0.01, whereas 40.4% of women but only 21% of low SES menp<0.001 were living alone. In a random subsample (N=217) complete HRV data were obtained(37 female, 180 male, mean age 58.2 (SD = 9.1) y.). In response to the intervention, HADS depression scores had decreased by about two points, between t0 and t3, but there was no relationship with HRV parameters or gender. At baseline, SDNN and RMSSD were significantly lower during the arithmetic stress task compared to the resting conditions (p <.001). Conclusion Social and psychological risk factors (depression, exhaustion, isolation and stress) were more common in low SES. In contrast, there was no SES effect on HRV measures, nor on other biological mechanisms,including S-lipids, S-coagulation factors, renal or thyroid function. CORRESPONDING AUTHOR: K. Orth-Gomer, Karolinska Institutet, Sweden, [email protected] O722 THE EFFECT OF ATTRIBUTIONS OF RESPONSIBILITY FOR ACCIDENTS ON PHYSICAL, PSYCHOLOGICAL AND FUNCTIONAL RECOVERY J.H. Thompson1, M. Berk2, M. O'Donnell3, L. Stafford4, T. Nordfjaern5 1 Monash University, CLAYTON, Australia 2 Deakin University, GEELONG, Australia 3 University of Melbourne, MELBOURNE, Australia 4 Royal Womens Hospital, MELBOURNE, Australia 5 Norwegian Institute for Alcohol and Drug Research, OSLO, Norway Whilst the potential role of attributions of responsibility on post-injury recovery has been recognised for some time, the extent and direction of its effect has remained uncertain. Rarely has research specifically set out to assess the comparative influence of attributions of responsibility on outcomes among people injured in motor vehicle accidents (MVAs). This presentation represents a significant body of work specifically designed to understand the effect of attributions of responsibility for MVAs on four areas of recovery; mental health outcomes, physical health outcomes, return to work, and patient satisfaction with the personal injury compensation scheme. Three studies using interviews from over 1100 participants over 2 time periods were conducted. All participants had been injured in an MVA and were current or previous clients of the Victorian Transport Accident Commission - a no-fault personal injury compensation scheme. Results showed that attributions of responsibility for motor vehicle accidents were significantly associated with poorer post-injury mental health outcomes, greater symptoms of depression, and lower levels of satisfaction with the compensation scheme. Attributions of responsibility were also associated with poorer physical health outcomes and return to work outcomes through the mediating role of mental health variables. Compensation schemes should include routine assessment of attributions of responsibility for accidents to identify clients at risk of both poor recovery and low levels of satisfaction. Future applied and academic research should control for attributions of responsibility when comparing mental and physical health outcomes, return to work, or patient satisfaction within or between injured populations. CORRESPONDING AUTHOR: Mr J.H. Thompson, Monash University, Australia, [email protected]

S213

O723 PATIENTS' DRAWINGS OF BLOOD CELLS REVEAL PATIENTS' PERCEPTION OF THEIR BLOOD DISORDER S.R. Ramondt1, J. Tiemensma1, L.D. Cameron1, E. Broadbent2, A.A. Kaptein3 1 University of California, Merced, MERCED, United States of America 2 University of Auckland, AUCKLAND, New Zealand 3 Leiden University Medical Center, LEIDEN, Nederland Introduction: Sickle cell disease (SCD) and thalassemia are rare but chronic blood disorders. Recent literature showed impaired quality of life (QOL) in patients with these blood disorders. Assessing one of the determinants of QOL (i.e. illness perceptions) therefore, is an important next research area. We aimed to explore illness perceptions of patients with a blood disorder with the Drawing Test in addition to the Brief Illness Perception Questionnaire (Brief IPQ). The Drawing Test is a novel test to assess illness perceptions and the free-range answers drawings offer can add additional insight into how patients perceive their illness. Method: We conducted a cross-sectional pilot study including 17 patients with a blood disorder. Patients’ illness perceptions were assessed by the Brief IPQ and the Drawing Test. Brief IPQ scores were compared with reference groups from the literature (i.e. patients with asthma or lupus erythematosus). Results: Patients with SCD or thalassemia perceived their blood disorder as being more chronic and reported more severe symptoms than patients with either asthma or lupus erythematosus. In the drawings of these patients with a blood disorder, a greater number of blood cells drawn was negatively correlated with perceived personal control (P<0.05) indicating that a greater quantity in the drawing is associated with more negative or distressing illness beliefs. Conclusion: Patients with a blood disorder perceive their disease as fairly threatening compared with patients with other chronic illnesses. Drawings can add additional insight into how patients perceive their illness by offering free-range answers. CORRESPONDING AUTHOR: S.R. Ramondt, University of California, Merced, United States of America, [email protected] O724 PERSISTENT NEGATIVE DISEASE PERCEPTIONS DESPITE LONG-TERM REMISSION OF ACROMEGALY: NOVEL APPLICATION OF THE DRAWING TEST J. Tiemensma1, E. Broadbent2, J.A. Romijn3, A.M. Pereira3, N.R. Biermasz3, A.A. Kaptein3 1 University of California, Merced, MERCED, United States of America 2 University of Auckland, AUCKLAND, New Zealand 3 Leiden University Medical Center, LEIDEN, Nederland Introduction: Patients with acromegaly have persistent complaints despite long-term biochemical control of the disease. Drawings can be used to assess disease perceptions of patients. We aimed to explore the utility of the Drawing Test and its relation to illness perceptions and quality of life (QoL). Method: We included 50 patients after long-term remission of acromegaly. Patients completed the Drawing Test (with drawings on their body perception before acromegaly, during the active phase of acromegaly, and after long term biochemical remission of acromegaly) , Illness Perception Questionnaire-Revised, Physical Symptom Checklist, EuroQoL-5D, and the AcroQoL. Results: Patients perceived a dramatic change in body size during the active state of the disease compared with the healthy state prior to awareness of acromegaly. Patients reported that their body did not completely return to the original proportions after long term biochemical control of acromegaly. There were strong correlations between the size of the drawings and the perceived negative consequences of acromegaly (P<0.05, larger drawings indicated more negative consequences). Furthermore, larger drawings indicated a higher score on emotional

S214

representations (P<0.05) and larger drawings also indicated more perceived symptoms that were attributed to acromegaly (P<0.01). The association between drawings and QoL was less apparent. Conclusion: The Drawing Test appears to be a novel and relatively easy tool to assess the perception of patients after long-term remission of acromegaly. The assessment of drawings may enable doctors to appreciate the perceptions of patients with long-term remission of acromegaly, and will lead the way in dispelling idiosyncratic beliefs. CORRESPONDING AUTHOR: J. Tiemensma, University of California, Merced, United States of America, [email protected] O725 CARDIAC MISCONCEPTIONS IN FIRST-MI PATIENTS: PRELIMINARY RESULTS FROM AN INTERVENTION STUDY M.J. Figueiras1, J. Maroco2, R. Monteiro1, R. Caeiro1, M. Trigo1 1 Instituto Piaget, ALMADA, Portugal 2 UIPES / ISPA, LISBOA, Portugal Introduction: Cardiac patients hold specific beliefs about their illness that can be misconceived and negatively affect their adjustment. This study tested the efficacy of an intervention aimed at dispelling common cardiac misconceptions in myocardial infarction (MI) patients. Method: Eighty-six first-MI patients were randomized to control group (standard care) or intervention group. The intervention consisted of 1 inhospital session followed by 4 phone calls after discharge. Patients were assessed at baseline and 4 months after hospital discharge. The main outcomes were cardiac misconceptions, illness perceptions, perceived health-related quality of life, anxiety, depression and health behaviours. Results: At the 4-month , patients in the control group had significantly higher misconceptions, more negative illness perceptions of identity, concern and emotional response and higher levels of depression than patients in the intervention group. Also, a significantly higher proportion of participants reported doing exercise in the intervention group than in the control group. No significant differences between groups were found in the remaining outcomes. Conclusion: A cardiac misconception intervention can be effective in altering MI patient’s erroneous cardiac beliefs and negative illness perceptions, with beneficial effects over depression levels and exercise behaviours. CORRESPONDING AUTHOR: Prof. M.J. Figueiras, Instituto Piaget, Portugal, [email protected] O726 THE RELATIONSHIP BETWEEN THE BELIEF IN A GENETIC C A U S E F O R B R E A S T C A N C E R A N D B I L AT E R A L MASTECTOMY Keith Petrie1, S.M. Myrtveit2, Ann H. Patridge3, Melika H. Stephens1, A.L. Stanton3 1 Department of Psychological Medicine, University of Auckland, AUCKLAND, New Zealand 2 Department of Clinical Science, BERGEN, Norway 3 Dana-Farber Cancer Institute, Brigham and Women’s Hospital, Harvard Medical Scho, LOS ANGELES, United States of America Introduction: Most women develop causal beliefs following diagnosis with breast cancer and these beliefs can guide decisions around care and management. Bilateral mastectomy rates are increasing, although the benefits of this surgery are only established in a small percentage of women. In this study we investigated the relationship between causal beliefs and the decision to undergo bilateral mastectomy. Method: 2,269 women from the Army of Women’s breast cancer research registry completed an on-line survey. Women were asked what they believed caused their cancer and responses were coded into eight causal categories. Participants were also asked about the type of surgery they underwent

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

following their breast cancer diagnosis. The odds ratio for having had bilateral mastectomy were calculated for each causal category using random/bad luck as referent category. Results: Hormonal factors (22%) and genetics (19%) were the most common causal belief, followed by don’t know (19%), environmental toxins (11%), negative emotions (9%), poor health behavior (8%), other (6%) and random/bad luck (6%). Compared to the referent category, the odds ratio of having a bilateral mastectomy was significantly higher in both the genetics and hormonal causal belief groups (OR=2.58 [95%CI: 1.56-4.26] and OR=1.87 [95%CI: 1.13-3.08] respectively). Conclusions: Beliefs in a genetic cause for breast cancer are common and are associated with high rates of bilateral mastectomy. This is despite evidence that the actual genetic contribution to breast cancer is much lower than perceived and that bilateral mastectomy is, in most cases, unlikely to reduce the future risk of cancer or improve survival. CORRESPONDING AUTHOR: Keith Petrie, Department of Psychological Medicine, University of Auckland, New Zealand O727 WOMEN'S EXPERIENCES OF HORMONAL THERAPY FOR BREAST CANCER; EXPLORING POTENTIAL INFLUENCES ON MEDICATION TAKING BEHAVIOUR C.C. Cahir1, A. Timmons2, C. Kelly3, M.J. Kennedy1, L. Sharp2, K. Bennett1 1 Trinity College Dublin, DUBLIN 8, Ireland 2 National Cancer Registry, CORK, Ireland 3 Mater Misericordiae University Hospital, DUBLIN, Ireland Introduction: Five to ten years of hormonal therapy is recommended for the prevention of breast cancer recurrence and mortality; however, rates of non-adherence range from 16% to 32% at 5 years. Existing evidence, albeit limited, suggests that side-effects, co-morbidities, socioeconomic status may be associated with medication taking behaviour (MTB) in breast cancer. This study is investigating potential influences on hormonal therapy MTB using the Theoretical Domains Framework (TDF). Method: Thirty-one semi-structured face-to-face interviews were conducted with women with stage I-III breast cancer prescribed hormonal therapy. Participants were classified by their MTB (adherent/persistent; non-adherent/persistent; non-persistent). Data analysis was ongoing and iterative using thematic analysis based on the Framework approach, with the TDF informing the analysis framework. The TDF is an integrative framework consisting of fourteen domains of behavioural change to inform intervention development and implementation. Results: Preliminary analysis has identified eleven theoretical domains as potentially relevant to hormonal therapy MTB. Key enablers for adherent/ persistent women were identified within the domains of knowledge (medication), social influences (relationship with healthcare providers, support), beliefs about capabilities (side-effects), optimism, intentions, goals (high-priority), beliefs about consequences (breast cancer recurrence) and emotion (fear). Barriers were identified within the domains of skills, memory, attention and decision processes (forgetting) and environmental context and resources (stressors) for non-adherent/persistent women and knowledge, social influences, beliefs about capabilities and beliefs about consequences (lack of belief in efficacy) for non-persistent women. Conclusion: The results will inform the development of an intervention to improve MTB and associated outcomes in women with breast cancer. CORRESPONDING AUTHOR: Dr C.C. Cahir, Trinity College Dublin, Ireland, [email protected] O728 WHEN IMMEDIATE EFFECTS DO NOT PREDICT SUSTAINED EFFECTS: THE CASE OF HEALTH WARNINGS R. Borland Cancer Council Victoria, MELBOURNE, Australia

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

Experimental studies of graphic health communications often report negative or reactance effects. However, research on the effectiveness of emotionally evocative health warnings and advertisements to discourage smoking show clear positive effects. Why does this inconsistency occur? This paper reviews the evidence. Population-based studies of actual warning regimes show that the frequency of reporting avoidance of health warnings is positively correlated with the frequency of reporting quittingrelated thoughts, both of which are prospectively predictive of making quit attempts, although the relationship with avoidance disappears when controlling for quit related thoughts. It may be that in the period immediately after avoiding health warnings a person is less likely to engage in quit related activity. However, as warnings are potentially presented many times a day, those sufficiently concerned to avoid the warnings are also likely to sometimes find themselves confronted by them. Further, as their level of self-efficacy efficacy for doing something about their smoking varies, on some such occasions their confidence will be great enough for them to take action, even though on others they may avoid the issue. Thus there are plenty of opportunities for the warnings to trigger a move towards quitting, some of which result in quit attempts. What happens in isolated occasions cannot be simply assumed to mirror the impact of repeated events. This has important implications for our understanding of how experimental studies can inform population-level interventions. CORRESPONDING AUTHOR: Dr R. Borland, Cancer Council Victoria, Australia, [email protected] O729 DOES AA WORK DIFFERENTLY FOR YOUNGER PEOPLE? A MODERATED MULTIPLE MEDIATION ANALYSIS B. Hoeppner, S. S. Hoeppner, J. F. Kelly Harvard Medical School, BOSTON, MA, United States of America Aims: Research has shown that participation in Alcoholics Anonymous (AA) confers significant recovery benefit to adults suffering from alcohol use disorder (AUD). Concerns persist, however, that AA may not work as well for younger adults, who tend to have shorter addiction histories, different social circumstances, and less spiritual/religious interest than adults aged 30+. Design: Secondary data analysis of Project MATCH, using a prospective, moderated multiple mediation analysis to test and compare six previously identified mechanisms of change in younger adults (n=266) vs. adults aged 30+ (n=1,460). Setting: Nine clinical sites within the United States. Participants: Treatment-seeking adults (n=1,726) suffering from AUD who participated in 12 weeks of outpatient treatment and completed follow-ups at 3-, 9and 15-months. Measurements: AA attendance during treatment; mediators at 9 months; and outcomes [percentage of days abstinent (PDA) and drinks per drinking day (DDD)] at 15 months. Findings: AA attendance was associated with improved drinking outcomes in both younger adults and adults aged 30+ . Similarly, multiple mediation was significant for both outcomes in both younger adults (PDA: t=3.0, p<0.01; DDD: t=-2.4, p<0.05) and adults aged 30+ (PDA: t=8.4, p<0.01; DDD: t=-6.8, p<0.01), yet only two of the six hypothesized pathways (i.e., decreases in pro-drinking social networks, self-efficacy in social situations), were significant in younger adults. Conclusion: Young adults appear to benefit from AA attendance similarly to adults aged 30+, yet the underlying mechanisms of behavior change are less well understood. Once identified, these heretofore unidentified mechanisms could yield novel intervention targets for this population. CORRESPONDING AUTHOR: Dr. B. Hoeppner, Harvard Medical School, United States of America, [email protected] O730 SHOULD BRIEF INTERVENTIONS IN PRIMARY CARE MORE STRONGLY ADDRESS ALCOHOL PROBLEMS? J. McCambridge1, S. Rollnick2 1 LSHTM, LONDON, United Kingdom 2 Cardiff University, CARDIFF, United Kingdom

S215

Background: Brief interventions have well established small effects on alcohol consumption among hazardous and harmful drinkers in primary care and national large scale programmes are being implemented in many countries for public health reasons. Methods: This paper examines data from reviews and draws upon older brief intervention studies and recent developments in the literature on motivational interviewing to consider the capacity of brief interventions to benefit those with problems, including those with severe problems. Results: Effects on alcohol problems have been much less consistently shown, and evidence cannot be claimed to be strong for any outcomes other than reduced consumption. Combinations of advice and motivational interviewing are a promising target for evaluation in trials and more detailed studies of the conduct of brief interventions are needed. Conclusions: We propose that brief interventions in primary care may be more effective if they offer appropriate content in a person-centred manner, more directly addressing patient concerns. CORRESPONDING AUTHOR: Dr J. McCambridge, LSHTM, United Kingdom, [email protected] O731 SMOKING CESSATION IN CARDIAC PATIENTS: THE INFLUENCE OF ACTION AND COPING PLANS AND SELF-EFFICACY N. de Hoog Open University of the Netherlands, HEERLEN, Nederland Introduction. Smoking cessation is the most effective treatment for smokers with coronary heart disease, yet more than half of cardiac patients continue to smoke. This study examined some factors that could explain (un)successful smoking cessation in cardiac patients in the postmotivational phase, as postulated by the I-change model (De Vries et al., 2003). In particular, we examined the influence of making action and coping plans and self-efficacy on intention to quit and actual smoking cessation. Method. Cardiac patients from eight hospitals in the Netherlands filled in a baseline questionnaire (N = 245) during their hospital stay, consisting of demographic and smoking characteristics, intention to quit smoking, smoking behavior, self-efficacy, and making of action plans and coping plans. The follow up six months later (N = 184) included smoking behavior (continued abstinence). Results. Both individual and number of action plans had a direct effect on continued abstinence, whereas coping plans were unrelated to continued abstinence. Self-efficacy explained 36% of the variance of intention to quit smoking and was also an indirect predictor of continued abstinence, through intention. Intention to quit smoking and number of action plans both directly influenced continued abstinence, explaining 31% of the variance in continued abstinence. Conclusions. Future interventions to facilitate smoking cessation in cardiac patients should put more emphasis on improving self-efficacy. Moreover, making specific action plans could increase the effectiveness of smoking cessation. CORRESPONDING AUTHOR: dr N. de Hoog, Open University of the Netherlands, [email protected] O732 PREVENTING SMOKING RELAPSE AMONG SMOKERS VIA A WEB-BASED MULTIPLE-SESSION ATTENTIONAL BIAS MODIFICATION TRAINING I. Elfeddali1, C. Bolman2, T. Pronk3, H. De Vries1, R.W. Wiers3 1 School for Public Health and Primary Care (Caphri), Maastricht University, MAASTRICHT, The Netherlands 2 Open University of the Netherlands, HEERLEN, The Netherlands 3 University of Amsterdam, AMSTERDAM, The Netherlands

S216

Introduction: Implicit cognitive processes such as attentional and approach biases have been associated with smoking behavior. This study aimed to assess the efficacy of a multiple-session attentional bias modification (ABM) training in modifying these implicit cognitive biases and fostering six months continued smoking abstinence. Differential effects of the training were assessed for light-moderate (6-15 cigarettes/day) and heavy (>15 cigarettes/ day) smokers. Methods: Respondents were daily smokers aged 18-65 years and motivated to quit smoking within one month. The study was a randomized controlled trial with two conditions: ABM training and a continued assessment control group. Both conditions consisted of 1. a pre- and post-training assessments with a selfreport questionnaire on smoking-related cognitions (e.g., selfefficacy), a Visual Probe (VP) task for measuring attentional bias and a Stimulus Response Compatibility (SRC) task for measuring approach bias; 2. six intervention (training or assessment) sessions; and 3. a follow-up measurement after six months (assessing continued abstinence). Results: The ABM-training had a significant positive effect on continued abstinence, in heavy smokers only (abstinence rates were 50% in the training group versus 23% in the control group when using complete cases and 15% versus 6% when considering dropouts as relapsers). In heavy smokers, the training resulted in small effects with regard to attentional bias and no effects on approach bias. The training had no effect for light to moderate smokers. Conclusions: Providing multiple sessions of attentional bias modification training via the Internet has the potential to foster continued smoking abstinence in heavy smokers who make a quit-attempt. CORRESPONDING AUTHOR: Dr. I. Elfeddali, School for Public Health and Primary Care (Caphri), Maastricht University The Netherlands, [email protected]

Int.J. Behav. Med. (2014) 21 (Suppl 1):S1–S216

O733 A BEHAVIOR CHANGE INTERVENTION TO REDUCE ARECA NUT CHEWING AND IMPROVE PERCEPTIONS ABOUT ITS HAZARDS K. Shafique1, Z.I. Arain2, M.I. Arain2, S.S. Mirza3 1 Dow University of Health Sciences, KARACHI, Pakistan 2 Isra Medical University, HYDERABAD, Pakistan 3 University of Rotterdam, ROTTERDAM, Nederland Introduction.Areca nut chewing has increased in recent years among teenagers in many countries; however, little is known about potential interventions to reduce its burden. This study aimed to assess the impact of pictorial anti-areca chewing warning messages on prevalence of areca nut chewing and perceptions about its hazards among teenage school children. Methods. A multistage random sampling was done to select 13-16 years old school children (n=620) from secondary schools of Hyderabad, Pakistan. Pictorial anti-areca chewing warning messages (having pictures of lip, oral and esophageal cancers) were shown to students twice during the baseline screening. The point-prevalence of areca nut chewing and perceptions about hazards of areca nut chewing were assessed at baseline and at week 4 using a validated questionnaire. The McNemar’s test was used to assess the difference in prevalence of areca nut chewing while the Mann Whitney U test to assess the difference of perception score, before and after intervention. Results. The point-prevalence of areca nut chewing reduced (absolute difference 4.7%, p-value <0.001) from 33.9% (n=211) to 29.2% (n=182) and the perception score increased significantly (p-value <0.001) after the intervention with higher rank sum. Watching a picture of oral cancer had the highest rank score in convincing the harmful effect of areca nut chewing compared with pictures of lip and esophageal cancers. Conclusion. This behavior change intervention using pictorial warning messages reduced areca nut chewing and also perceived effective by the teenage school children in convincing about harmful effects of areca nut chewing. CORRESPONDING AUTHOR: Dr. K. Shafique, Dow University of Health Sciences, Pakistan, [email protected]