Int.J. Behav. Med. (2011) 18:188–198 DOI 10.1007/s12529-010-9124-7

Increasing Physical Activity in Belgian Type 2 Diabetes Patients: a Three-Arm Randomized Controlled Trial Karlijn De Greef & Benedicte Deforche & Catrine Tudor-Locke & Ilse De Bourdeaudhuij

Published online: 30 October 2010 # International Society of Behavioral Medicine 2010

Abstract Background Pedometer-based physical activity programs have been typically delivered in a group format by a behavioral expert. An alternative strategy that builds on existing interactions is delivery through individual consultation by a general practitioner (GP). These two delivery strategies have not been directly compared. Purpose To compare effectiveness of a 12-week physical activity (PA) intervention for type 2 diabetes patients delivered by a trained GP via an individual consultation or as group delivery by a behavioral expert. Method Sixty-seven primary care participants (mean age= 67.4 years, 70% male) from three Belgian general practices were randomized into three different treatment arms: (1) individual consultation (n=22) with three PA contacts with The trial is registered at ClinicalTrials.gov, number NCT00903500. B. Deforche : I. De Bourdeaudhuij Department of Movement and Sports Sciences, Ghent University, Ghent, Belgium B. Deforche Research Foundation Flanders (FWO), Ghent, Belgium B. Deforche Department of Human Biometry and Biomechanics, Vrije Universiteit Brussel, Brussels, Belgium C. Tudor-Locke Walking Behavior Laboratory, Pennington Biomedical Research Center, Baton Rouge, USA K. De Greef : I. De Bourdeaudhuij (*) Department of Movement and Sports Sciences, Ghent University, Watersportlaan 2, 9000 Ghent, Belgium e-mail: [email protected]

the patient's GP; (2) group counseling (n=21) with three PA group sessions delivered by a behavioral expert; and (3) a control arm (n=24) receiving no intervention. Participant inclusion criteria were ≤80 years; 25–35 kg/m²; ≤12% HbA1c and reporting no PA limitations. Outcome measures were pedometer-determined steps/day, self-reported PA, and health parameters (weight, body mass index, waist circumference, total cholesterol, fasting glucose, and HbA1c). Results Group counseling participants increased 1,706 steps/ day over baseline significantly (p≤0.05) more than other treatment arms. Moreover, they increased their self-reported PA (+82 min/day), while control arm participants showed a decrease in PA (p≤0.05). Participants of the individual consultation had a decrease in waist circumference (−1.4 cm) and HbA1c (−0.32%) and a lower increase in total cholesterol (+7.2 mg/dl) compared to the other treatment arms (all p≤0.05). Conclusion Group counseling in type 2 diabetes patients improved PA, whereas individual consultations had an impact on some health outcomes on the short-term. Keywords Physical activity . Pedometer . Lifestyle changes . Behavioral intervention . Physician . Diabetes mellitus type 2

Introduction Despite the evidence that physical activity (PA) is beneficial to health and one of the cornerstones of diabetes management [1–7], it is an often neglected and an underused aspect in diabetes care [8–11]. Moreover, type 2 diabetes patients report that they receive less support, education, and encouragement concerning PA in comparison to any other aspect of their diabetes management [12, 13]. For example, in Belgium, patients only receive financial

Int.J. Behav. Med. (2011) 18:188–198

refund and support for their medication and dietary treatment [14]. There is a dearth of practical interventions achieving sustained, long-term change in PA behavior in type 2 diabetes patients [15], and little is known about the most appropriate way of promoting PA in diabetes care [8]. In an effort to promote sustainable behavior change, lifestyle approaches focused on promoting PA on a selfstructured and frequent basis have become more common [16]. Program delivery modes have included telephone counseling [17, 18], internet-based programming [19, 20], group-based delivery [21–29], or individual consultations, either with a behavioral expert [30–33] or with the patient's general practitioner (GP) [34–39]. Despite the appeal of telephone and internet-based programming, face-to-face delivery by means of group or individual sessions appears to be most effective [21–33]. Unfortunately, delivery via group sessions has some disadvantages: they require additional organization beyond existing infrastructures, a behavioral expert needs to be appointed and paid, and patients have to be brought together and tracked. Diabetes management already requires a typically demanding self-care regimen, and these added group sessions may be considered an additional patient burden. Finally, sustainability can be a concern if no further contact is planned following completion of the intervention. Of all these options, an individual consultation with the patient's GP provides a unique opportunity to build upon an existing relationship and infrastructure (and therefore is seemingly efficient) but also promises the established effectiveness of face-to-face interaction [40]. In Belgium and Europe, most type 2 diabetes patients (80–90%) are seen by their GP [41–43]. GPs play an important role in type 2 diabetes guidance and are presumably in the best position to coordinate individual patient's care [44] since they interact on a face-to-face basis nine times a year [43, 45, 46]. Furthermore, patients perceive their GP as a trusted and preferred source of lifestyle counseling [44, 47], and they rely on him/her to manage their diabetes care [45, 48]. It is therefore possible to conceive of efficient and effective lifestyle PA counseling as an integrated part of patients' usual care [44, 49]. Despite this, GPs do not routinely counsel patients about PA [50], and in particular, little time is spent on PA counseling with diabetes patients [51], and existing lifestyle counseling sessions are often superficial [52, 53]. GPs indicate that the many medical concerns and patients' priorities expressed during the usual care visit make it difficult to address the topic of PA [54]. Moreover, GPs report some important barriers to the promotion of PA in their daily practice such as lack of time, lack of knowledge about PA, lack of training in cognitivebehavioral techniques and communication skills, lack of support for their efforts, and lack of remuneration [44, 46, 48, 55, 56]. To overcome at least some of these barriers, GPs require adequate PA education and training to develop

189

their communication skills and learn how to implement cognitive-behavioral change strategies and tools within the context of their consults [47, 53, 57–59]. Few studies have examined the effectiveness of lifestyle PA interventions for type 2 diabetes patients in primary care [34–39], and only three focused on GP delivery of PA consultation. Di Loreto et al. and Christian et al. showed that a GP-based intervention could increase PA, suggesting that a dialogue between patients and their GPs about behavioral goals can lead to increased PA [35, 39]. Van Sluijs et al. reported no intervention-specific effects (compared to a control condition) following a GPdelivered PA [36]. More research is needed to investigate the effects of GP-delivered PA interventions. For example, to date, the efficacy of pedometer use—which has been shown to be effective intervention strategy in group sessions [21–23, 25, 28, 29]—is still untested in GP counseling. Pedometers are considered to be highly useful motivational tools, providing a direct source of feedback, a readily available memory prompt, and in particular, a cue to increase PA [60]. It is an easy to explain self-monitoring tool, which could be a very workable instrument in primary care, both for the patients and the GPs. Therefore, it is plausible that a primary care intervention would be more effective if pedometer-based programming was included. Overall, the studies above indicate that lifestyle-based PA interventions—regardless of their context—show promising short-term results relating to both acceptability and effectiveness. However, it is currently not clear whether group counseling given by a behavioral expert is necessary or whether trained GPs are capable of obtaining changes in PA behavior in patients with type 2 diabetes, by providing individual PA consultation as a part of regular patient visits. The purpose of this controlled comparative effectiveness study in individuals with type 2 diabetes was to investigate whether a 12-week pedometer-based PA intervention individually delivered by a trained GP during patient visits can be as effective as group delivery by a behavioral expert. We hypothesized that both intervention treatments would have more favorable PA changes compared to the nonintervention control condition. Should this be the case, we could conclude that GP delivery would be the most feasible in Belgium, given that diabetes patients are already meeting with their GP on a regular basis as part of their usual care.

Methods Study Design and Sample Recruitment This study took place from March 2007 to June 2007. Three GPs volunteered to incorporate PA counseling as part of their patient visits, and a sampling pool of potential

190

Int.J. Behav. Med. (2011) 18:188–198

participants was generated from their practices (A, B, and C, respectively). Participants' inclusion criteria were: (1) ≥6 months post diagnosis of type 2 diabetes (disease is stabilized); (2) age ≤80 years (if >80, too much risk for injuries); (3) BMI 25–35 kg/m² (obese patients with BMI >35 kg/m² need to be referred to other treatments); (4) pharmaceutically treated for type 2 diabetes; (5) HbA1c ≤12% (if >12%, need to be referred to more intensive treatment); (6) no documented physical or medical PA limitations; (7) Dutch speaking. There were no exclusion criteria. In total, 261 type 2 diabetes patients were treated. Based on the inclusion criteria, each GP electronically selected eligible type 2 diabetes patients from the medical records of the general practice. Together, the GPs identified a pool of 86 individuals as eligible to participate from their practice lists. Sixty-seven individuals responded positively to a letter of invitation, signed by their GP (response rate, 78%). The most commonly reported reasons for refusal to participate included lack of time or interest. Eighteen participants from practice A, 27 participants from practice B, and 22 participants from practice C were included and signed an informed consent form. Within each practice, participants were randomly assigned (via computerized random numbers) by the GP to either an individual consultation (n=22), a group counseling (n=21), or a control condition (n=24). Randomization was concealed in sealed envelopes until the point of allocation. The GP carried out the process of randomizing participants (sequence generation, allocation concealment, and implementation). Blinding to group allocation could not be maintained postrecruitment, as with most behavioral interventions. The GPs did the blinded group allocation and measurements. A behavioral expert delivered the face-to-face group counseling intervention, while the individual face-to-face consultations were delivered by the patients' GP. The statistical analyses were anonymously executed by the behavioral expert. The Ethical Committee of the Ghent University hospital approved the study.

sleeping. This pedometer has a readable display and has been shown to be valid, accurate, and reliable for counting steps in adults [61]. An activity log was used to record the steps taken and the type and duration of non-walking activities [62]. Participants were asked to complete the activity log at the end of each day. Following established guidelines, participants were instructed to add 150 steps to the daily total for every minute actively spent biking or swimming [62]. At the end of each measurement period, all participants gave their activity log (with an anonymous study number) to their GP. Self-reported PA was assessed with the Dutch interview version of the long International PA Questionnaire (IPAQ) (last 7 days). The interview, instead of the self-administered questionnaire, was chosen because adults tend to overreport their PA levels with the latter version [63]. Validity and reliability of the IPAQ has been shown to be acceptable [64, 65]. In the questionnaire, frequency (number of days) and duration (hours and minutes per day) of PA in different domains (work, transportation, leisure time, and housekeeping) were queried.

Measurements: Materials

Intervention Procedures

The GPs performed all measurements (described below) at baseline and immediately following the 12-week intervention. All participants completed a background questionnaire with general demographic and medical information.

Both intervention conditions began by receiving a pedometer. The pedometer and accompanying diary were both used as motivational tools in PA goal-setting. Participants were asked to wear the pedometer and to record their PA type, duration, and number of steps/day in the diary at the end of each day. The pedometer diary was used to track progress and to encourage discussions with the behavioral expert or the GP, depending on assigned treatment condition. Participants assigned to the cognitive-behavioral group sessions were offered three 90-min group counseling sessions over a 12-weeks period (one session every

Pedometer-Based and Self-Reported PA Measurements At the first measurement visit, the GP instructed the participants how to wear and use a pedometer correctly. The pedometer (The Yamax DigiWalker SW200, Tokyo, Japan) was worn at the waist during waking hours for seven consecutive days following the visit. Participants were asked only to remove it for water-based activities and

Health Measurements Body weight was measured to the nearest 0.5 kg wearing light clothing and no shoes using a SECA 813 Robusta balance. Standing height was measured to the nearest 0.1 cm using a wall-mounted Harpenden stadiometer. Standing waist circumference was measured to the nearest 0.1 cm at the narrowest part of the torso between the rib cage and the iliac crest. Biochemical correlates included fasting plasma glucose, HbA1c, and total cholesterol. Blood samples taken in were analyzed in the respective practice-associated laboratories, using the same analyzing methods: the hexokinase method to analyze fasting plasma glucose and the Adams Hemoglobin A1c procedure to analyze HbA1c. Plasma cholesterol was analyzed with the enzymatic colorimetric analysis.

Int.J. Behav. Med. (2011) 18:188–198

3 weeks). The sessions were delivered by a behavioral expert, more specifically, a clinical psychologist with a background in behavior change strategies. All sessions took place in the waiting room of the GPs when the practice was closed for other patients. The group sessions were based on cognitive-behavioral therapy [66], the Diabetes Prevention Program [67], the First Step Program [68], and motivational interviewing [69]. Table 1 gives an overview of the content of the different sessions. The group sessions started with a motivational interviewing phase. In the first session, each participant received a pedometer. The aim of this session was to increase knowledge about the benefits of PA. During this session, discussions with the participants about benefits and risks took place. Commitment to a lifestyle change plan was strengthened by learning time management skills. The participants together with the behavioral expert developed a lifestyle change plan in which “where, when, and how” of the planned behavior changes was explained. Patients were encouraged to take responsibility for their own goal-setting. In a second session, the first experiences of the participants with the pedometers as a motivation tool and goal-setting were exchanged. Benefits and barriers of PA as well as strategies and tips on how to overcome the barriers were discussed. In the last session, the focus was on learning relapse prevention skills. The group format was highly interactive; ideas were introduced by the participants and discussed within the group. During the group sessions, the participants received personal goal-setting sheets and a decisional balance table. The participants allocated to the GP-delivered intervention received three individual 15-min face-to-face consultations. The content of this intervention was similar as this of the group intervention (Table 1). Since it was not possible for the GPs to quote all topics in the limited time they had for each patient (15 min vs. 90 min of group session), they were asked to focus on pedometer use, goal-setting, decisional balance, and relapse prevention. The three GPs received an intensive 10-h train-thetrainer session from the behavioral expert (who was in charge of the group counseling). Training topics included time management, goal-setting, decisional balance, selfefficacy, and relapse prevention. Motivational interviewing techniques were presented, and role-playing was practiced. In addition, the behavioral expert modeled the techniques of the group sessions before the GPs started with their own individual consultations. Furthermore, after the train-the-trainer session, a rehearsal with the three GPs and the behavioral expert was organized to discuss how the future PA consultations would be conducted. Consequently, GPs had enough background and experience in the key topics and techniques that had to be discussed. Moreover, they had a basis of the techniques of motivational interviewing and PA consultation. Finally, the GPs

191 Table 1 Intervention content Session Session content

Behavioral/cognitive and self-regulation methods

Session • Welcome • Information 1 • Increasing knowledge of the • Consciousness raising benefits of physical activity (PA) • Mastery experience • Time management (skills) • Explanation of the benefits • Self monitoring of pedometer use (pedometer) Session • Experiences with pedometers • Activity reminders 2 and goal-setting (skills) • Self monitoring of behavior (pedometer) • Increasing self-efficacy to set • Coaching in realistic and up an action plan measurable goal-setting • Listing personal benefits and • Decisional balance barriers (changing attitudes) • Counter-conditioning • Increasing self-efficacy to • Discussion of problemovercome barriers solving approach to address behaviors Session • Increasing self-efficacy to • Reattribution training 3 change habits • Cognitive restructuring • Planning coping responses • Relapse prevention (skills) • Coaching in realistic and measurable goal-setting • Discussing action plan and • Feedback on performance increasing self-efficacy to set up a long-term action plan The taxonomy of behavior change techniques was used to define the behavioral/cognitive and self-regulation methods [80]

were contacted after their first consultation to discuss problems or questions raised. The control group only received general care from their GP. Data Analyses Data were analyzed using SPSS 15 for Windows and were expressed as mean±standard deviations (SD). A p value of ≤0.05 was interpreted as statistically significant, and a p value between 0.5 and 0.1 was considered indicative of a trend towards significance. One-way analyses of variance (ANOVA) showed that the three treatment arms were comparable on most, but not all descriptive characteristics, at baseline. Those characteristics (BMI, waist circumference, and diastolic blood pressure) that differed between treatment arms were included as covariates in the analyses.

192

Intervention effects were analyzed using repeated-measures analyses of covariance (ANCOVA) (with treatment arm as the between-subjects factor and time as the within-subjects factor). Separate post hoc analyses (repeated-measures ANCOVA) were executed to examine the mutual differences among the three treatment arms. Data were analyzed using intention-to-treat and involved all participants who were randomly assigned to the different treatment arms. Data of the participants who dropped out (three patients) were included in analyses; baseline scores were substituted for missing data for week 13. From the activity logs, pedometer data from seven consecutive days were extracted, and 150 steps were added to the daily total for every minute of reported biking and/or swimming [61]. The amount of PA in each IPAQ domain was calculated by multiplying the frequency (number of days) of the activity by the duration (minutes per day) of the activity. To control for overestimations of PA by the IPAQ, 80% of the value of actual scores was reported [63, 65]. A score for total housekeeping and gardening was obtained by summing up moderate-intensity housekeeping and moderate- to vigorous-intensity (MVPA) gardening. A score for walking during leisure time was obtained by summing up walking at light-intensity and walking at moderate-intensity during leisure time. A total PA score included PA during working hours, transportation, leisure time, and housekeeping and gardening. Total MVPA consisted of MVPA identified during working hours, transportation, leisure time, housekeeping, and gardening. As the self-reported PA data were positively skewed, the statistical analyses were performed on the log-transformed IPAQ data [62]. For reasons of clarity and comparability, means and SD of the original IPAQ data are reported in the results. A priori power analyses were conducted using anticipated intervention effects reported in previous research [18]. Based on a 0.80 power to detect an expected increase of 2,000 steps/day [70] (p=0.05, two-sided), 25 patients were required for each study group.

Int.J. Behav. Med. (2011) 18:188–198

self-reported PA and health outcomes (fasting glucose, HbA1c, total cholesterol)). Exceptions included BMI (F=3.2, p≤0.05), waist circumference (F=5.7, p≤0.01), and diastolic blood pressure (F=4.6, p≤0.05). The group counseling participant had a lower BMI and waist circumference, and the individual consultation participants had a higher diastolic blood pressure compared to the others. Dropout during the 12-week intervention was 4.5%. Two participants from general practice B (one form the control condition, one from the group counseling) and one person from general practice C (control condition) dropped out, all because of medical reasons. Effect on PA Outcomes Pedometer-based PA Outcomes An overall time-by-group interaction effect (F=3.68, p≤0.05) was found for daily step counts. The participants of the group counseling increased their steps/day by 1,706 ± 698, which was significantly more than those of the control group (+313± 493 steps/day) (F=5.08, p≤0.05) and those of the individual consultation group (+837±688 steps/day) (F=4.31, p≤ 0.05). No significant time-by-group interaction effect was found for the individual consultation and the control group (F=1.42, p≥0.1) (see Table 3). Self-reported PA Outcomes Overall analyses of selfreported PA revealed time-by-condition interaction effects for housekeeping and gardening (F=6.05, p≤0.01), walking during leisure time (F=5.06, p≤0.01), total PA (F= 4.76, p≤0.05), and MVPA (F=5.67, p≤0.01). Post hoc analyses showed that these differences were between the group counseling and the control condition (all p≤0.05). After 12 weeks, the group-counseling participants showed an increase in self-reported PA while the control condition showed a decrease. Furthermore, group-counseling participants showed an increase in MVPA, while MVPA decreased in the individual consultation participants (F=4.15, p≤0.05). Overall, no differences over time were found between the individual consultation and the control condition for self-reported PA (see Table 3).

Results Participants' Characteristics

Effect on Health Outcomes

At baseline, mean age of the participants was 67.4 (±9.3) years and mean weight was 88.5 (±16.4)kg. Additional demographic characteristics can be found in Table 2. At baseline, no differences were noted between the three conditions for most demographic characteristics (age, education, civil state, employment status, smoking behavior, duration of type 2 diabetes, health problems, perceived health status, use of diabetic medications) and descriptive (steps/day,

Overall analyses on health outcomes showed significant time-by-condition interaction effects on waist circumference (F=3.56, p≤0.05) and a trend towards significance for BMI and HbA1C (see Table 3). The change over time was different for the individual consultation participants and those of the control condition for waist circumference (F= 6.88, p≤0.05) and HbA1c (F=4.77, p≤0.05). There was a decrease in waist circumference (3 cm) and HbA1C

Int.J. Behav. Med. (2011) 18:188–198 Table 2 Demographic characteristics of the study sample

193 Total (n=67)

Age (years) (mean±SD) Gender (%) Female Male Education (%) Primary Secondary College/University Civil state (%) Married Single Employment status (%) Employed Not employed Retired Smoking (%) Non-smoker Smoker Diabetes diagnosis (%) <5 years >5 years Health problems (%) None Problems Health status (%) Average or good Weak Medication use (%) Insulin Oral medication Combination

Control group (n=24)

Group counseling (n=21)

Individual consultation (n=22) F 1.16

67.4±9.3

66.0±11.1

70.0±6.3

66.6±9.5 χ² 1.2

29.9 70.1

29.2 70.8

38.1 61.9

22.7 77.3

20.3 51.6 28.1

26.1 56.5 17.4

25.0 35.0 40.0

47.6 23.8 28.6

74.2 25.8

79.2 20.8

81 19

61.9 38.1

7.1

2.5

9.2 13.4 13.4 73.1

25.0 16.7 58.3

0.0 4.8 95.2

13.6 18.2 68.2

34.4

39.1

25.0

38.1

65.6

60.1

75.0

69.9

64.5 35.5

59.1 40.9

68.4 31.6

66.7 33.3

9.4 90.6

13.0 87.0

10.0 90.0

4.8 95.2

90.6 9.4

82.6 17.4

90.0 10.0

0.0 100.0

1.6 90.3 8.1

4.5 77.3 18.2

0.0 95.0 5.0

0.0 100.0 0.0

1.1

0.5

0.9

3.9

9.1

(0.32%) in the individual consultation participants, whereas waist circumference remained the same, and HbA1C decreased slightly in the control condition. No differences in changes over time were found for the group counseling and the control condition or for the group counseling and the individual consultation participants (see Table 3).

Discussion The purpose of this study was to investigate whether a 12week PA intervention delivered to type 2 diabetes patients as an individual consultation by a trained GP can be as

effective as group counseling delivered by a behavioral expert. Therefore, the effects of an individual PA consultation and a PA cognitive-behavioral group counseling were evaluated in a three-arm randomized controlled trial. It was hypothesized that both intervention groups would have more favorable PA changes compared to the nonintervention control group. If so, the PA consultations by the GP are the most feasible, as diabetes patients are meeting their GP on a regular basis as part of their usual care. Results demonstrated that the group counseling was more effective in increasing pedometer-determined PA compared to both the individual consultation and the

194 Table 3 Overall effects of the interventions on patient's outcomes (repeated-measures ANCOVA) and post hoc analyses (mutual differences between the group counseling (n=21), individual consultation (n=22), and control group (n=24))

Int.J. Behav. Med. (2011) 18:188–198 Measurements (mean±SD)

Baseline

Physical activity (PA outcomes) Steps/day Control group 4,831±2,601 Group counseling 5,924±3,725 Individual consultation 5,114±2,503 Min/day PA for housekeeping and gardening Control group 31±50 Group counseling 44±41 Individual consultation 59±66 Min/day walking during leisure time Control group 17±29 Group counseling 30±38 Individual consultation 16±22 Min/day total PA Control group 86±72 Group counseling 113±74 Individual consultation 126±86 Min/day moderate-to-vigorous PA (MVPA) Control group 44±52 Group counseling 46±43 Individual consultation 60±60 Health outcomes Body mass index (BMI) (kg/m²) Control group 31.5±5.6 Group counseling 28.2±3.4 Individual consultation 31.2±5.3 Waist circumference (cm) Control group 110.1±12.5 Group counseling 100.0±8.4 Individual consultation 109.9±12.2 Fasting glucose (mg/dl) Control group 130.0±40.2 Group counseling 134.9±30.2

A: Group condition differs from control condition B: Individual condition differs from control condition C: Group condition differs from individual condition *p≤0.10; *p≤0.05; **p≤0.01

Individual consultation HbA1C (%) Control group Group counseling Individual consultation Total Cholesterol (mg/dl) Control group Group counseling Individual consultation

control condition. Group counseling participants had a defined increase in steps/day (+1,706) compared to those of the other conditions. This magnitude of increase with only three sessions over 12 weeks is promising since it has been shown that an increase of 2,000–2,500 steps/day can improve health outcomes over the longer term [70–72].

Post-intervention

F (time×group)

5,173±3,094 7,630±4,423 5,951±3,191

3.68*

A* C*

19±40 65±66 60±73

6.05**

A**

11±26 40±44 23±28

5.06**

A*

4.76*

A*

5.67**

A** C*

65±68 195±106 158±110 27±39 75±64 88±100

31.5±5.3 27.3±3.4 30.9±5.3

2.83*

110.4±11.7 98.6±7.8 106.6±12.4

3.56*

131.2±26.8 131.7±33.7

B*

1.14

143.1±54.4

129.7±33.9

7.00±0.87 7.12±1.35 7.23±0.71

6.93±0.84 6.94±1.19 6.91±0.60

2.59*

B*

183.5±43.6 184.7±33.2 193.8±38.8

192.4±46.3 192.0±28.9 196.2±46.5

2.30

B*

However, in this short-term study, no improvements in health outcomes were observed for the patients assigned to the group counseling intervention. There are several facts that might explain this; (1) an increase of 1,706 steps/day might have been too small, (2) there were only measurements on a short-term basis, (3) the sessions only focused

Int.J. Behav. Med. (2011) 18:188–198

on PA. Compared to the participants of the individual consultation and the control group, the participants of the group counseling reported also a greater increase in walking during leisure time. The most remarkable effect of the group counseling was thus seen on walking behavior (steps/day and reported walking during leisure time), which is an activity that can be performed safely and that is easily incorporated into daily life. As a consequence, it can be recommended as an adjunct therapy to diet treatment in type 2 diabetes patients [73]. Also for all other self-reported PA variables (housekeeping and gardening, MVPA, and total PA), an increase was found in the participants of the group counseling compared to those of the other groups. A first explanation for the behavioral differences between the group counseling and the individual consultation could be that the group structure enables participants to exchange their knowledge, beliefs, and experiences [74]. A second explanation could be that peer support and interactive group visits favor learning, behavior change, and maintenance in diabetes patients [75]. Thirdly, the skills and motivation of the behavioral expert could have been an important factor. Even with intervention-specific behavioral training, it is plausible that GPs may still have had some problems with communication skills and that they are still not as well founded in cognitive-behavioral techniques and tools as a behavioral expert. Fourthly, the intensity of the sessions could also be an important factor [76, 77]. The participants of the group counseling had a longer exposure for PA counseling than those of the individual consultation. GPs reported lack of time: it was not always easy for them to make extra time in their daily practices for the study patients; 15 min was the maximum time they could spend on each PA consultation. This was also supported by other studies in GPs [44, 46, 56]. A patient-centered approach, leading to increased PA knowledge, self-monitoring, self-efficacy, and eventually PA increase, almost inevitably requires long and repeated consultations [78]. So it is possible that to increase the program impact of the individual consultation; program exposure and treatment intensity need to be increased [78]. However, it must be kept in mind that the individual guidance in the individual consultation of 15 min is as long as or even longer than the individual guidance of the participants in the group counseling (six to nine participants in a session of 90 min). In addition, more information about the exact content of the consultations after the train-thetrainer sessions is necessary, as we did not gather such information. Data of quality of treatment delivery by the general practitioners are also lacking. In future research, both participants and GPs should be questioned on this topic. Finally, it is important to notice that the intervention framework, used in the different intervention studies, enclosed many cognitive-behavioral techniques. The be-

195

havioral expert had no problem implementing all these techniques. However, the difficulties that non-psychologists might experience with implementing the intervention framework in daily practice were not considered. For example, the GPs did not have the same amount of time, resources, or abilities as the behavioral expert to deliver such intensive lifestyle counseling. Moreover, it is important to notice that since it was not possible for the GPs to quote all topics in the limited time they had for each patient, they focused on pedometer use, goal-setting, decisional balance, and relapse prevention. The lack of behavioral effects in their consultations might be the result of skipping important factors responsible for increases in physical activity. Although the fact that it is questionable that there would be already effects of this short-term intervention on health outcomes, individual consultation participants showed significant improvements in health outcomes compared to those of the group condition and the control condition. The participants of the individual consultation decreased in HbA1c and waist circumferences and had a smaller increase in total cholesterol compared to the other conditions. The group counseling had no effects on health outcomes. A possible explanation might be that the GPs also emphasized health parameters (nutrition, weight, glycemic control) in their consultations because they are more familiar with them compared to the behavioral expert (although this was not the purpose of the study). Consequently, patients could have incorporated not only the aspect of PA, but also for example, the aspect of healthy nutrition in their daily life. On the contrary, as the group counseling only focused on PA, the topic “food” was not discussed during the sessions. Secondly, from conversations with the participants of both intervention groups, it emerged that they had a tendency to compensate their increase in PA with increases in food intake. Unfortunately, as it was not our aim to incorporate food and medication use in our intervention, we had no detailed information about adjustments in medication use and dietary behavior of the patients. Further research is necessary on the compliance of the GPs with the intervention guidelines, as we did not gather such information. Earlier studies [21–29] found similar results concerning the behavioral effects of group counseling. Furthermore, some authors did not find effects on health outcomes after group counseling [22, 25, 29]. Those three interventions had a similar duration as ours and focused on PA. In contrast, the study of Johnson et al. [28] showed clear effects on weight, BMI, blood pressure, and resting heart rate after 24 weeks. However, it is important to bear in mind that the latter study had a duration of 24 weeks, whereas ours only had a duration of 12 weeks. The lack of behavioral effect in the individual consultations delivered by the GPs was not found in other GP-based interventions

196

[35, 36, 39]. In the study of Van Sluijs et al. [36], the participants of the intervention as well as the control group had improvements in health outcomes and PA on the longer term [36]. In contrast, the study of Di Loreto et al. [35] and Christian et al. [39] found clear effects on both health and PA after the intervention. Again, these studies had a duration of respectively 12 and 24 weeks [35, 39]. Several limitations must be acknowledged in this study. First, it should be emphasized that it cannot be concluded whether the effects are due to the delivery mode (individual versus group) or to the deliverer (GP versus behavioral expert). However, this was not the purpose of the present study. It is not possible for a GP to give group sessions for all his/her type 2 diabetes patients just like it is not feasible (from financial point of view) to appoint a behavioral expert to give individual PA consultation to type 2 diabetes patients in general practices. Another important limitation of this study is the fact that the two intervention groups not only differed in mode of deliverer (GP versus behavioral expert), but also in mode of delivery (individual versus group sessions) with different duration and content of consultations and opportunities for social comparisons and learning from group dynamics in the group counseling. Therefore, the conclusions drawn from this study need to be cautious. Secondly, it has to be taken into account that this trial had no longer-term follow-up period. It is possible that the group counseling and the individual consultation have a positive impact on motivation to change behavior in the short-term but not over the long-term. On the contrary, it is possible that the short-term changes, found in PA for the group counseling did not yet turn over in effects on health outcomes in such a short timeframe. Thirdly, this study was limited by the duration of the intervention. It was a relatively short intervention period, and it is possible that the participants of the GP consultation group, who had fewer changes in PA, needed more PA consultation time to embed those changes in their daily life. A fourth limitation was that the study was somewhat underpowered and that there were only three, relatively young, GPs included, which limits the opportunity to generalize the results. Moreover, as the GPs were enrolled in a course for training, they might not be representative of the broader body of GPs in Belgium. Furthermore, reactivity could have taken place as the pedometer was used as both a measurement and an intervention instrument [79]. However, the use of a control group does solve some problems of reactivity. Finally, it is important to notice that some behavior change interventions among type 2 diabetes patients targeted both PA and dietary behaviors. In the present study, it was decided not to combine both behaviors but to put a strong and sole emphasis on PA in the intervention.

Int.J. Behav. Med. (2011) 18:188–198

Despite these limitations, this is the first study that compared both methods: group counseling by a behavioral expert and an individual consultation by a GP. The GPs received an intensive train-the-trainer session. Moreover, this is the first time that a pedometer approach was used in primary care. Additional studies are required to examine why behavioral effects were missing in the individual consultations and why health effects were lacking in the group counseling sessions. It is possible that our train-thetrainer session or the individual consultation intervention were not intense enough. Future research might consider the possibility of giving an initial lifestyle PA intervention for type 2 diabetes patients by means of group sessions given by a behavioral expert and afterwards continue the care by GPs. The long-term feasibility and possibilities of this kind of interventions require further attention. Among other things, attendance rates at sessions and consultations were high and dropout rate was low. Participants of both intervention conditions were satisfied with the kind of intervention they received. The GPs reported, just like in earlier studies: lack of time and no extra remuneration as barriers [44, 46, 56]. They found it very difficult to incorporate lifestyle advice in their consultations and proposed to incorporate a behavioral (PA) expert for the future care of Belgian type 2 diabetes patients next to GPs, endocrinologists, and dieticians. It can be concluded that a lifestyle PA intervention combined with pedometer use has some promising effects in type 2 diabetes patients in primary care. From this study, it can be inferred that group counseling in type 2 diabetes patients improved PA, whereas individual consultations had an impact on some health outcomes on the short-term. Continued support for a longer time than 3 months is necessary. Acknowledgements We are much indebted to Stephanie Degomme, Evelyne Courteyn, and Ellen Van Puyvelde for their collaboration in this study. They followed the train-the-trainer session, gave the GP interventions, and assisted with the data collection. Further, we thank the patients who participated in this study. Conflict of Interest We confirm that we have no conflict of interest. We have full control of all primary data and we agree to allow the journal to review our data if requested. All patient/personal identifier have been removed or disguised so the patient/person(s) described are not identifiable and cannot be identified through the details of the story.

References 1. Albright A, Franz M, Hornsby G, Kriska A, Marrero D, Ullrich I, et al. American College of Sports Medicine position stand. Exercise and type 2 diabetes. Med Sci Sports Exerc. 2000;32:1345–60. 2. Toledo FGS, Menshikova EV, Ritov VB, Azuma K, Radikova Z, DeLany J, et al. Effects of physical activity and weight loss on

Int.J. Behav. Med. (2011) 18:188–198

3.

4.

5.

6.

7. 8.

9.

10. 11.

12.

13.

14. 15.

16.

17.

18.

19.

20.

21.

22.

skeletal muscle mitochondria and relationship with glucose control in type 2 diabetes. Diabetes. 2007;56:2142–7. Thomas DR, Elliott EJ, Naughton GA. Exercise for type 2 diabetes mellitus. Cochrane Database Syst Rev. 2006;2:1–41 (Online: Updated Software). Hu FB, Stampfer MJ, Solomon C, Liu SM, Colditz GA, Speizer FE, et al. Physical activity and risk for cardiovascular events in diabetic women. Ann Int Med. 2001;134:96–105. Gregg EW, Gerzoff RB, Caspersen CJ, Williamson DF, Narayan KMV. Relationship of walking to mortality among US adults with diabetes. Arch Int Med. 2003;163:1440–7. Department of Health, Physical Activity, Health Improvement, & Prevention. At least five a week: evidence of the impact of physical activity and its relationship to health. A report from the chief medical officer. London: Department of Health; 2004. LaMonte MJ, Blair SN, Church TS. Physical activity and diabetes prevention. J Appl Physiol. 2005;99:1205–13. Kirk AF, Barnett J, Mutrie N. Physical activity consultation for people with type 2 diabetes. Evidence and guidelines. Diabet Med. 2007;24:809–16. Dutton GR, Johnson J, Whitehead D, Bodenlos JS, Brantley PJ. Barriers to physical activity among predominantly low-income African-American patients with type 2 diabetes. Diab Care. 2005;28:1209–10. Constans T, Lecomte P. Non pharmacological treatments in elderly diabetics. Diab Metab. 2007;33:S79–86. Pham DT, Fortin F, Thibaudeau MF. The role of the health belief model in amputees' self-evaluation of adherence to diabetes selfcare behaviors. Diab Educ. 1996;22:126–32. Wilson W, Ary DV, Biglan A, Glasgow RE, Toobert DJ, Campbell DR. Psychosocial predictors of self-care behaviors (compliance) and glycemic control in non-insulin-dependent diabetes-mellitus. Diab Care. 1986;9:614–22. Ary DV, Toobert D, Wilson W, Glasgow RE. Patient perspective on factors contributing to nonadherence to diabetes regimen. Diab Care. 1986;9:168–72. INAMI—RIZIV. Zorgtrajecten, of hoe de aanpak van chronische ziekten verbeteren. Available at:http://www.zorgtraject.be. 2009. Harris SB, Petrella RJ, Leadbetter W. Lifestyle interventions for type 2 diabetes—relevance for clinical practice. Can Fam Physician. 2003;49:1618–25. Dunn AL, Marcus BH, Kampert JB, Garcia ME, Kohl HW, Blair SN. Comparison of lifestyle and structured interventions to increase physical activity and cardiorespiratory fitness: a randomized trial. JAMA. 1999;281:327–34. Richter H, Kraft K, Kleinwechter H, Demandt N, Meincke G, Dabelstein A, et al. Effects of a telephone intervention in patients with type 2 diabetes: increase of physical activity and reduction of cardiovascular risk factors. Dtsch Med Wochenschr. 2008;133:2203–8. Eakin E, Reeves M, Lawler S, Graves N, Oldenburg B, Del Mar C, et al. Telephone counseling for physical activity and diet in primary care patients. Am J Prev Med. 2009;36:142–9. Liebreich T, Plotnikoff RC, Courneya KS, Boule N. Diabetes NetPLAY: a physical activity website and linked email counselling randomized intervention for individuals with type 2 diabetes. Int J Behav Nutr Phys Act. 2009;6:18. Mckay HG, King D, Eakin EG, Seeley JR, Glasgow RE. The diabetes network internet-based physical activity intervention—a randomized pilot study. Diab Care. 2001;24:1328–34. Tudor-Locke C, Lauzon N, Myers AM, Bell RC, Chan CB, McCargar L, et al. Effectiveness of the first step program delivered by professionals versus peers. J Phys Act Health. 2009;6:456–62. Tudor-Locke C, Bell RC, Myers AM, Harris SB, Ecclestone NA, Lauzon N, et al. Controlled outcome evaluation of the first step

197

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

39.

program: a daily physical activity intervention for individuals with type II diabetes. Int J Obes. 2004;28:113–9. Furber S, Monger C, Franco L, Mayne D, Jones LA, Laws R, et al. The effectiveness of a brief intervention using a pedometer and step-recording diary in promoting physical activity in people diagnosed with type 2 diabetes or impaired glucose tolerance. Health Prom J Aus. 2008;19:189–95. Toobert DJ, Strycker LA, Glasgow RE, Barrera M, Angell K. Effects of the Mediterranean lifestyle program on multiple risk behaviors and psychosocial outcomes among women at risk for heart disease. Ann Behav Med. 2005;29:128–37. Cheong SH, Hanson J, Fitzpatrick JJ, McCargar L, Paty B, TudorLocke C, et al. First step first bite program: increasing physical activity and low glycemic index foods among individuals with type 2 diabetes. Diabetes. 2005;54:A561. Davies MJ, Heller S, Skinner TC, Campbell MJ, Carey ME, Cradock S, et al. Effectiveness of the diabetes education and selfmanagement for ongoing and newly diagnosed (DESMOND) programme for people with newly diagnosed type 2 diabetes: cluster randomised controlled trial. Br Med J. 2008;336:491–5. Vincent D. Culturally tailored education to promote lifestyle change in Mexican Americans with type 2 diabetes. J Am Acad Nurse Pract. 2009;21:520–7. Johnson ST, Bell GJ, McCargar LJ, Welsh RS, Bell RC. Improved cardiovascular health following a progressive walking and dietary intervention for type 2 diabetes. Diab Obes Metab. 2009;11:836– 43. De Greef K, Deforche B, Tudor-Locke C, De Bourdeaudhij I. A cognitive-behavioural pedometer-based group intervention on physical activity and sedentary behaviour in individuals with type 2 diabetes. Health Educ Res. 2010; doi: 10.1093/her/cyq017 Kirk A, Mutrie N, MacIntyre P, Fisher M. Effects of a 12-month physical activity counselling intervention on glycaemic control and on the status of cardiovascular risk factors in people with type 2 diabetes. Diabetologia. 2004;47:821–32. Kim CJ, Hwang AR, Yoo JS. The impact of a stage-matched intervention to promote exercise behavior in participants with type 2 diabetes. Int J Nurs Stud. 2004;41:833–41. Kirk A, Mutrie N, MacIntyre P, Fisher M. Increasing physical activity in people with type 2 diabetes. Diab Care. 2003;26:1186– 92. Clark M, Hampson SE, Avery L, Simpson R. Effects of a tailored lifestyle self-management intervention in patients with type 2 diabetes. Br J Health Psychol. 2004;9:365–79. Gleeson-Kreig JM. Self-monitoring of physical activity—effects on self-efficacy and behavior in people with type 2 diabetes. Diab Educ. 2006;32:69–77. Di Loreto C, Fanelli C, Lucidi P, Murdolo G, De CA, Parlanti N, et al. Validation of a counseling strategy to promote the adoption and the maintenance of physical activity by type 2 diabetic subjects. Diab Care. 2003;26:404–8. van Sluijs EMF, van Poppel MNM, Twisk JWR, Paw MJCA, Calfas KJ, van Mechelen W. Effect of a tailored physical activity intervention delivered in general practice settings: results of a randomized controlled trial. Am J Public Health. 2005;95:1825– 31. Jackson R, Asimakopoulou K, Scammell A. Assessment of the transtheoretical model as used by dietitians in promoting physical activity in people with type 2 diabetes. J Hum Nutr Diet. 2007;20:27–36. Bjorgaas MR, Vik JT, Stolen T, Lydersen S, Grill V. Regular use of pedometer does not enhance beneficial outcomes in a physical activity intervention study in type 2 diabetes mellitus. Metabolism. 2008;57:605–11. Christian JG, Bessesen DH, Byers TE, Christian KK, Goldstein MG, Bock BC. Clinic-based support to help overweight patients

198

40.

41. 42.

43.

44.

45.

46.

47.

48.

49.

50.

51.

52.

53.

54.

55.

56.

57.

58. 59.

60.

Int.J. Behav. Med. (2011) 18:188–198 with type 2 diabetes increase physical activity and lose weight. Arch Intern Med. 2008;168:141–6. Mehuys E, De Bolle L, Van Bortel L. Behandeling van diabetes type 2 in Vlaanderen. Available at: www.hospitals.be/pdf/ vol6n2p30.pdf. 2008. Jaffiol C. Current management of type 2 diabetes is France. Bull Acad Natl Méd. 2009;193:16450–1661. Ashenden R, Silagy C, Weller D. A systematic review of the effectiveness of promoting lifestyle change in general practice. Fam Prac. 1997;14:160–75. Van Den Berg MJ, Kolthof ED, De Bakker DH. Tweede nationale studie naar ziekten en verrichting in de huisartspraktijk. De werkbelasting van de huisartsen. Nivel: Utrecht; 2004. Harris SB, Petrella RJ, Lambert-Lanning A, Leadbetter W, Cranston L. Lifestyle management for type 2 diabetes—are family physicians ready and willing? Can Fam Physician. 2004;50:1235–43. Harris SB, Stewart M, Brown JB, Wetmore S, Faulds C, WebsterBogaert S, et al. Type 2 diabetes in family practice—room for improvement. Can Fam Physician. 2003;49:778–85. Larme AC, Pugh JA. Attitudes of primary care providers toward diabetes—barriers to guideline implementation. Diab Care. 1998;21:1391–6. Pinto BM, Goldstein MG, Ashba J, Sciamanna CN, Jette A. Randomized controlled trial of physical activity counseling for older primary care patients. Am J Prev Med. 2005;29:247–55. Brown JB, Harris SB, Webster-Bogaert S, Wetmore S, Faulds C, Stewart M. The role of patient, physician and systemic factors in the management of type 2 diabetes mellitus. Fam Prac. 2002;19:344–9. Lawlor DA, Keen S, Neal RD. Increasing population levels of physical activity through primary care: GPs' knowledge, attitudes and self-reported practice. Fam Prac. 1999;16:250–4. Lobelo F, Duperly J, Frank E. Physical activity habits of doctors and medical students influence their counselling practices. Br J Sports Med. 2009;43:89–92. Clark M, Hampson SE. Implementing a psychological intervention to improve lifestyle self-management in patients with type 2 diabetes. Patient Educ Couns. 2001;42:247–56. Wolpert HA, Anderson BJ. Management of diabetes: are doctors framing the benefits from the wrong perspective? BMJ. 2001;323:994–6. Poskiparta M, Kasila K, Kiuru P. Dietary and physical activity counselling on type 2 diabetes and impaired glucose tolerance by physicians and nurses in primary healthcare in Finland. Scand J Prim Health Care. 2006;24:206–10. Huang J, Yu H, Marin E, Brock S, Carden D, Davis T. Physicians' weight loss counseling in two public hospital primary care clinics. Acad Med. 2004;79:156–61. Kenny SJ, Smith PJ, Goldschmid MG, Newman JM, Herman WH. Survey of physician practice behaviors related to diabetesmellitus in the United States. Diab Care. 1993;16:1507–10. McKenna J, Naylor PJ, McDowell N. Barriers to physical activity promotion by general practitioners and practice nurses. Br J Sports Med. 1998;32:242–7. Burke LE, Fair J. Promoting prevention: skill sets and attributes of health care providers who deliver behavioral interventions. J Cardiovasc Nurs. 2003;18:256–66. Phillips LS, Branch WT, Cook CB, Doyle JP, El-Kebbi IM, Gallina DL, et al. Clinical inertia. Ann Intern Med. 2001;135:825–34. Wens J, Vermeire E, Royen PV, Sabbe B, Denekens J. GPs' perspectives of type 2 diabetes patients' adherence to treatment: a qualitative analysis of barriers and solutions. BMC Fam Pract. 2005;6:20. Tudor-Locke CE, Bell RC, Myers AM, Harris SB, Lauzon N, Rodger NW. Pedometer-determined ambulatory activity in individuals with type 2 diabetes. Diab Res Clin Pract. 2002;55:191–9.

61. Crouter SE, Schneider PL, Karabulut M, Bassett DR. Validity of ten electronic pedometers for measuring steps, distance, and energy cost. Med Sci Sports Exerc. 2003;35:1455–60. 62. De Cocker K, Cardon G, De Bourdeaudhuij I. Pedometerdetermined physical activity and its comparison with the international physical activity questionnaire in a sample of Belgian adults. Res Q Exerc Sport. 2007;78:429–37. 63. Rzewnicki R, Auweele YV, De Bourdeaudhij I. Addressing overreporting on the International Physical Activity Questionnaire (IPAQ) telephone survey with a population sample. Public Health Nutr. 2003;6:299–305. 64. Craig CL, Marshall AL, Sjostrom M, Bauman AE, Booth ML, Ainsworth BE, et al. International physical activity questionnaire: 12-country reliability and validity. Med Sci Sports Exerc. 2003;35:1381–95. 65. Vandelanotte C, De Bourdeaudhuij I, Sallis JF, Philippaerts R, Sjostrom M. Reliability and validity of a computerised International Physical Activity Questionnaire (IPAQ). JPAH. 2005;2:63–75. 66. Welschen LMC, van Oppen P, Dekker JM, Bouter LM, Stalman WAB, Nijpels G. The effectiveness of adding cognitive behavioural therapy aimed at changing lifestyle to managed diabetes care for patients with type 2 diabetes: design of a randomised controlled trial. BMC. 2007;7:74. 67. The Diabetes Prevention Program (DPP). Description of lifestyle intervention. Diab Care. 2002;25:2165–71. 68. Tudor-Locke CE, Myers AM, Rodger NW. Development of a theory-based daily activity intervention for individuals with type 2 diabetes. Diab Educ. 2001;27:85–93. 69. Resnicow K, Dilorio C, Soet JE, Borrelli B, Hecht J, Ernst D. Motivational interviewing in health promotion: it sounds like something is changing. Health Psychol. 2002;21:444–51. 70. Bravata DM, Smith-Spangler C, Sundaram V, Gienger AL, Lin N, Lewis R, et al. Using pedometers to increase physical activity and improve health—a systematic review. JAMA. 2007;298:2296–304. 71. Tudor-Locke CE. Taking steps toward increased physical activity: using pedometer to measure and motivate. Presidents Counc Phys Fit Sports Ser. 2002;3:17. 72. Hill JO, Wyatt HR, Reed GW, Peters JC. Obesity and the environment: where do we go from here? Science. 2003;299:853–5. 73. Yamanouchi K, Shinozaki T, Chikada K, Nishikawa T, Ito K, Shimizu S, et al. Daily walking combined with diet therapy is a useful means for obese NIDDM patients not only to reduce bodyweight but also to improve insulin sensitivity. Diab Care. 1995;18:775–8. 74. Maldonato A, Bloise D, Ceci M, Fraticelli E, Fallucca F. Diabetesmellitus—lessons from patient education. Patient Educ Couns. 1995;26:57–66. 75. van Dam HA, van der Horst FG, Knoops L, Ryckman RM, Crebolder HFJM, van den Borne BHW. Social support in diabetes: a systematic review of controlled intervention studies. Patient Educ Couns. 2005;59:1–12. 76. Wee CC. Physical activity counseling in primary care—the challenge of effecting behavioral change. JAMA. 2001;286:717–9. 77. Simons-Morton DG, Blair SN, King AC, Morgan TM, Applegate WB, O’Toole M, et al. Effects of physical activity counseling in primary care—the activity counseling trial: a randomized controlled trial. JAMA. 2001;286:677–87. 78. Absetz P, Valve R, Oldenburg B, Heinonen H, Nissinen A, Fogelholm M, et al. Type 2 diabetes prevention in the “Real world”—1-year results of the GOAL implementation trail. Diab Care. 2007;30:2465–70. 79. Clemes SA, Parker RA. Increasing our understanding of reactivity to pedometers in adults. Med Sci Sports Exerc. 2009;41:674–80. 80. Abraham C, Michie S. A taxonomy of behavior change techniques used in interventions. Health Psychol. 2008;27:379– 87.