Ó Springer 2006

Quality of Life Research (2006) 15: 249–257 DOI 10.1007/s11136-005-0890-7

Changes in health status, psychological distress, and quality of life in COPD patients after hospitalization Randi Andenæs1, Torbjørn Moum2, Mary H. Kalfoss3 & Astrid K. Wahl4 1 Faculty of Nursing Education, Oslo University College, P.O. Box 4, St. Olavs plass, Oslo, Norway (E-mail: [email protected]); 2Department of Behavioural Sciences in Medicine, Medical Faculty, University of Oslo, P.O. Box 1111, Blindern, Oslo, Norway; 3Faculty of Nursing Research, Diakonova University College, Linstows gt. 5, Oslo, Norway; 4Faculty of Nursing Education, Oslo University College, P.O. Box 4, St Olavs plass, Oslo, Norway Accepted in revised form 8 July 2005

Abstract The purpose of this paper was to describe quality of life (QoL) following an acute exacerbation of chronic obstructive pulmonary disease (COPD), and to examine possible relationships between QoL, health status, psychological distress and QoL. This prospective longitudinal study examined data from hospitalization and two subsequent phases, collected over a 9-month period. The sample consisted of 51 COPD patients aged 48–87 years. Health status was assessed with the St George’s Respiratory Questionnaire (SGRQ); psychological distress with the Hopkins Symptom Checklist (HSCL); and quality of life using the World Health Organization Quality of Life – Bref (WHOQOL – Bref). Health status improved significantly over the 9 months; from 65.95 to 59.40 (p=0.001) in the SGRQ total score. Psychological distress improved significantly from hospitalization to the 1-month assessment (T2) (p=0.001). QoL remained stable except for a significant increase in the physical domain from hospitalization to T2, and in the environmental domain from hospitalization to T2 and T3. Test–retest correlations of the WHOQOL – Bref were high. The results suggest an improvement in the health status over a 9-month period from being discharged after an exacerbation of COPD in spite of high levels of psychological distress and reduced QoL. Key words: Chronic obstructive pulmonary disease, COPD, Health status, Psychological distress, Quality of life Abbreviations: COPD – Chronic obstructive pulmonary disease; HSCL – The Hopkins Symptom Checklist; Predicted FEV1% – Forced expiratory volume produced in first second, compared with the predicted normal value based on age, height and gender; QoL – Quality of life; SGRQ – St George’s Respiratory Questionnaire; WHOQOL – Bref: The World Health Organization Quality of Life – Bref Version

Introduction Measurement of quality of life (QoL), and in particular its health-related components, has been increasingly used in patients with Chronic obstructive pulmonary disease (COPD). This recognizes the greater emphasis now being placed on impaired quality of life and non-clinical aspects of treatment and the illness situation [1].

Research has shown that the symptoms associated with COPD impair physical and psychosocial functioning [2]. However, one study found that patients rated their well-being as fairly good [3]. Patients with COPD are prone to exacerbations which may lead to hospitalization, which may represent additional stress and may reduce their QoL [4]. Moreover, attacks of dyspnea in COPD exacerbations have been found to elicit feelings of

250 threat and anxiety [5]. Psychological distress has also been documented as an important predictor of mortality in hospitalized and home-based COPD patients [6]. Only a few studies have examined changes in health status after an exacerbation of COPD. These studies show fast improvement in physical and emotional functioning over the first weeks, followed by a slower phase that continues for months [7]. Because previous studies have been confined to outpatients, and evaluating only limited aspects of quality of life, we wanted to (1) describe changes in health status, psychological distress, and QoL following hospitalization with an acute exacerbation, and (2) to uncover possible relationships between these variables over time.

Method Data collection and assessment Between September 1997 and February 2000, COPD patients were recruited during their hospital stay at the Pulmonary Department at a university hospital in Oslo, Norway. They met the following inclusion criteria: (1) a diagnosis of COPD; (2) aged 45 or above; (3) freedom from chronic conditions that might confound the results, e.g., serious cardiac deficiency, psychiatric illness treated during the last 5 years, known alcohol abuse, and active cancer treatment; and (4) the ability to understand and express themselves orally in Norwegian. Written informed consent was obtained from all participants. The investigator collected data in personal interviews conducted during the first days after admission. Follow-up interviews were then conducted at two points: 1 month after hospital discharge (Time 2), and from 6 to 9 months after hospital discharge (Time 3). The 1-month point was chosen because it is a recommended time for followup assessment after discharge from a hospital [8]. The 6- to 9-month follow-up point was considered to allow sufficient time for the exacerbation to have stabilized. The last interview was conducted at a point within this 3-month period because, in severe COPD, the patients’ physical condition can fluctuate. Patients who agreed by telephone to participate in the Time 2 and Time 3 assessments were interviewed in the patient’s home, or at the

interviewer’s office or another meeting place. Respondents were regarded as missing when they could not complete all of the follow-up interviews because of ill health, death, study refusal, or they could not be reached. The Regional National Committee for Medical Research Ethics in Norway reviewed the study. Medical assessments Medical variables included the duration of a patient’s COPD, comorbidity, lung function, days spent in hospital, the number of hospital admissions over the previous year prior to hospitalization, and readmissions during the 9-month follow-up period. Comorbidity was defined as the number of health conditions in addition to COPD at discharge, as reported in the patients’ medical records. Lung function was measured by predicted FEV1% (forced expiratory volume produced in the first second, compared with the predicted normal value based on age, height and gender) [9]. The numbers of hospitalizations during the previous year were recorded, and the numbers of readmissions in the follow-up period were ascertained from yes/no responses in the follow-up interviews. Health status measurement Health status was assessed with the St George’s Respiratory Questionnaire (SGRQ), which is a questionnaire designed for measuring subjective health status in patients with chronic lung disease [10]. The questionnaire includes 50 items with 76 weighted responses, providing a total score and three component scores: symptoms, activity, and impact. A low score indicates good health status, and an increase in score indicates a worsening health status. The recall period for symptoms was 1 year at Time 1, and 6–9 months at Time 3. The SGRQ is the most widely used measure during acute exacerbations [11]. Previous studies show satisfactory results regarding the validity and sensitivity in various types of lung disease [12, 13]. Psychological distress Psychological distress was assessed with the Hopkins Symptom Checklist-25 (HSCL-25), which

251

The sociodemographic variables, age, sex, and cohabitation status, were assessed during the initial interview. Information where patients were discharged after hospitalization was recorded.

health status, psychological distress and QoL. Group comparisons between the interview group and the missing group were performed on the baseline disease and sociodemographic variables. Differences in mean scores were compared using t-tests for continuous variables, and chi-squares for discrete variables. To analyse the differences between the scores in health status, psychological distress, and QoL at the three assessments, the mean scores were calculated for subscale scores of the SGRQ, the HSCL and for the domains of the WHOQOL – Bref, and compared using paired t-statistics. Further, we performed bivariate correlation analyses to study the associations between the study variables. Cronbach’s a coefficients for the HSCL, subscales of the SGRQ, and the domains of the WHOQOL – Bref were calculated [21]. In our model, we treated health status (SGRQ) and psychological distress (HSCL) as predictor variables and QoL (WHOQOL – Bref) as the outcome variable [22]. To address the question of which factors influence the QoL and to identify possible causal relationships, a crosslagged model was used [23] that compared the regression coefficients (b) for Variable A/Times 1–3 and Variable B/Times 2–3 with the regression coefficients between Variable B/Times 1–3 and Variable A/Times 2–3. Analyses were performed for the same variable measured at different times (bA1A2) and cross-lagged analyses were performed between different variables measured at different times (bA1B1). This is illustrated in Figure 1. Regression analyses were performed for the variables in the model to determine which variables remained significant predictors of outcome, controlling for the initial score. The crosslagged model was estimated for each time-period separately by using these procedures. Regression analyses were also performed to identify the variables explaining QoL at Times 1 and 3. The significance level for the standardized beta coefficients was set at 0.05.

Statistical analyses

Results

Statistics analyses were performed with SPSS version 11. Descriptive statistics such as means, standard deviations, and 95% confidence intervals were used for presenting the measurements in

Description of the sample

was originally developed from subscale scores of the SCL-90 [14, 15]. A 23-item version of the HSCL-25 was used in this study. Two items (suicidal ideation and the loss of sexual interest) were excluded. The HSCL has been widely used in health surveys [16], in the elderly [17], and in COPD patients [18]. Previous studies have demonstrated satisfactory validity and reliability of the HSCL as a measure of psychological distress [14, 19]. Quality of life QoL was assessed by the WHOQOL – Bref which is a 26-item generic questionnaire that provides a short-form QoL assessment and was developed from the original 100-item version [20]. This instrument focuses upon respondents ‘perceived’ quality of life. There are two items on general well-being: One yields information about an individual’s global quality of life and one about satisfaction with health in general. The WHOQOL – Bref contains 26 items which produces four domains related to quality of life: physical (health), psychological, social relationships and environmental. Each domain includes 3–8 items. Each item is scored on a scale from 1 to 5, higher scores indicating better QoL. Responses are elicited with reference to the previous 2 weeks. The WHOQOL – Bref was chosen for several reasons; test and retest correlations are needed in Norwegian validity testing, population norms and criterion related information has been recently published, and the instrument contains an environmental domain which is absent in many other QoL assessments. Sociodemographic and other variables

Of the 107 patients invited, 97 agreed to participate in the study, representing a response rate of

252 Cross-lagged model

A1

a

A2

b

b

B1

a

B2

Figure 1. A1, Variable1/Time 1; A2, Variable1/Time 3; B1, Variable2/Time 1; B2, Variable2/Time 3. Pathways ‘a’ denote test–retest correlations; pathways ‘b’ denote cross-lagged correlations.

91%. Two died during hospitalization and three were excluded because they suffered from serious illness. Of the 92 patients in the initial study, 10 died during the study, and 51 (55.4%) completed all three study points. Comparisons showed that the respondents who completed the study were younger than those who were missing; 67.6 and 71 years, respectively (p=0.009). However, no statistically significant differences were found in sex, predicted FEV1%, health status, psychological distress, and QoL. Sociodemographic and clinical data are presented in Table 1. Most respondents were women (58.8%) and most respondents lived alone (64.7%). Of the 49 patients with pulmonary function test, the mean predicted FEV1% was 39.8 (SD 16.65), indicating impaired pulmonary function typical of severe COPD [8]. A large majority of the sample (80.4%) had comorbidity. Cardiovascular disease was reported in 41.2%, pneumonia 39.2%, osteoporosis 11.8%, rheumatic arthritis 2%, and 33% had other different medical diseases (e.g., sequellae tuberculosis, treated cancer, hypothyreosis, diabetes mellitus). Readmissions were common. By Time 3, this included 33 (64.7%) of the patients. Changes in health status, psychological distress, and quality of life The results of the SGRQ scores are displayed in Table 2. Except for the symptom scores, the

impact and activity subscales improved significantly over the 9 months. The greatest difference was in the activity subscale, which changed by 9.3 points. Improvement in the total score was 6.55; indicating a clinically significant change [10]. Psychological distress, as measured by the HSCL (Table 2), showed a significantly reduction from Times 1 to 2 (p=0.001). At Time 1, 12 (23.5%) of the patients rated their QoL in general (WHOQOL – Bref) at least as good. At Time 2, the number was 19 (37.3%) and, at Time 3, it had risen to 21 (41.2%). On the Table 1. Baseline characteristics of COPD patients who completed three interview points (n=51) Age, mean (SD) Sex, female/male, n (%) Living alone, n (%) Disease duration, n (%) 0–5 years 6–15 years >16 years Comorbidity, n (%) No comorbidity >One comorbid condition FEV1% pred, mean (SD) Duration of hosp. stay, mean (SD) Hospitalizations previous year, n (%) Readmissions during follow-up, n (%) After 1 month (T2) After 6–9 months (T3)

67.6 (9.5) 30/21 (58.8/41.2) 33 (64.7) 23 (45.1) 17 (33.3) 11 (21.6) 10 (19.6) 41 (80.4) 39.8 (16.5) 12.8 (15.0) 40 (78.1) 20 (39.2) 33 (64.7)

253 Table 2. Means and changes in study variables for Times 1, 2, and 3 (n=51) Variables

Time 1

HSCL SGRQ Symptoms Impact Activity Total WHOQOL – Bref Phys dom Psych dom Social dom Envir dom

Time 2

2.05 (0.45)

1.84 (0.45)

75.38 59.36 72.69 65.95

(15.55) (14.21) (13.63) (10.89)

– – – –

10.59 13.06 14.77 13.51

(2.11) (2.54) (1.96) (1.89)

11.67 13.45 14.71 13.52

(2.87) (2.58) (1.95) (2.08)

Time 3 1.86 (0.50)

pa

pb

pc

0.001

0.018

0.670

75.80 52.04 63.38 59.40

(17.88) (17.60) (17.38) (15.67)

– – – –

0.824 0.002 0.001 0.001

– – – –

11.13 13.58 14.92 14.14

(3.13) (2.96) (2.27) (2.53)

0.003 0.259 0.725 0.945

0.207 0.245 0.699 0.017

0.120 0.683 0.378 0.003

Means at each time-point, with SDs (Standard Deviation) and t-tests. Differences were measured by a paired t-test between Times 1 and 2 (pa), Times 1 and 3 (pb), and between Times 2 and 3 (pc).

other hand, 23 (45.1%) reported their QoL as bad or very bad at Time 1. At Time 2, the number was 17 (33.3%), and at Time 3 it had fallen to 12 (23.6%). The results of the repeated measures of the WHOQOL – Bref domain scores are shown in Table 2. As can be seen, the mean scores were relatively stable. The scores of the physical domain improved significantly from Times 1 to 2. Likewise, the scores in the environmental domain improved significantly from Times 2 to 3. Correlations between study variables and their effects on the QoL Table 3 presents the correlations between the study variables at Times 1 and 3. At Time 1, the

HSCL showed a weak correlation with the SGRQ impact (r=0.279, p=0.05). At Time 3, the correlations with all the SGRQ subscales were statistically strong. Likewise, the initial correlations between the HSCL and the WHOQOL physical and psychological domains at Time 1 showed a stronger correlation at Time 3. Most study variables were correlated at Time 3. Regression analyses according to the cross-lagged model (Figure 1) were performed between study variables. The WHOQOL physical domain at Time 1 (independent variable) and the HSCL at Time 2 (dependent) demonstrated statistical significance (r=)0.330, p=0.007). The finding indicates that high scores of QoL (physical domain) at hospitalization predict improvement in psychological distress after 1 month. Significant

Table 3. Bivariate correlations between study variables at T1 and T3 (n=51) Time 1

Time 3 Symptoms Impact

Symptoms Impact Activity Psych distr Phys dom Psych dom Social dom Envir dom Cronbach’s a

0.684*** 0.329* 0.261 )0.079 )0.182 )0.154 0.078 )0.014 0.68

Activity

0.628*** 0.623*** 0.504*** 0.684*** 0.293* 0.344* 0.279* )0.138 )0.392** )0.329* )0.189 )0.101 0.023 0.131 )0.104 0.205 0.43 0.65

Psych distr Phys dom 0.504*** 0.645*** 0.506*** 0.336* )0.355* )0.435** )0.184 )0.157 0.82

)0.507* )0.734*** )0.536*** )0.638*** 0.394** 0.362** 0.182 0.148 0.50

Psych dom Social dom Envir dom Chronbach’s a )0.313* )0.498*** )0.378** )0.622*** 0.553*** 0.339** 0.061 0.179 0.61

)0.092 )0.127 )0.233 )0.225 0.094 0.442** 0.477** 0.313* 0.03

)0.316* )0.457** )0.411** )0.405** 0.363** 0.633*** 0.393** 0.616*** 0.53

0.71 0.70 0.72 0.89 0.79 0.85 0.44 0.66

Pearson’s correlation and Cronbach’s alpha coefficient (a). Time 1 (at hospitalization) values are below the numbers in bold, Time 3 (at 6–9 months) values are above the numbers in bold. Test–retest correlations (Times 1–3) are in bold. * p< 0.05; **p< 0.01; *** p< 0.001.

254 associations were also found from the SGRQ impact subscale to the physical and psychological domains of the WHOQOL Time 3, (r=)0.534, p < 0.001) and (r = )0.367, p = 0.006), respectively. Further regression analyses were performed to control for possible confounders. The analyses reveal that the initial correlation between psychological distress (HSCL) and the WHOQOL (physical and psychological domains) were reduced when controlling for the SGRQ impact, indicating that this relationship was spurious. The SGRQ impact seems to be a confounding variable in our study, indicating that this variable had a strong effect on both on psychological distress and the WHOQOL (physical and psychological domains). The cross-lagged analyses reveal a strong test– retest correlation for the WHOQOL physical domain from Times 1 to 2 (r=0.557, p < 0.001), and from Times 2 to 3 (r=0.662, p < 0.001). Similarly, the correlations are significant from Times 1 to 2, (r=0.546, p < 0.001), and from Times 2 to 3 (r=0.671, p<0.001) in the psychological domain. Evaluating the test–retest reliability, these findings indicate relatively strong stability in the measurements over time. The Chronbach’s a coefficients for the scales are presented in Table 3. At T3, the a coefficients are higher for all scales. Regression analyses (Table 4) show a statistically significant association between the SGRQ impact and the WHOQOL physical domain at T1 (b=)0.32, p=0.03), and at T3 (b=)0.67, p <0.001) and psychological domain (b=)0.46, p=0.02). Health status (SGRQ) explained 51% of the variance in the physical domain of QoL, and

20% of the psychological domain at T3. The findings are indicative of a relationship where the SGRQ impact is a mediating variable, influenced by symptoms and activity, on the physical domain of QoL. Regression analyses were also performed controlling for comorbidity without influencing the overall results.

Discussion Study results demonstrate that hospitalization with an exacerbation of COPD represents impaired health status and increased psychological distress supporting findings from other studies [24–26]. Although improvements in health status and psychological distress seemed to occur during the 9-month follow-up period, QoL remained relatively stable. Findings also suggest that the psychosocial consequences of the COPD seem to be the most predominating factor influencing QoL. Changes in health status Hospitalization for an acute exacerbation of COPD identifies the subgroup of the COPD patients most severely affected. These assumptions are confirmed in the sample in the present study, consisting of elderly patients with a low predicted FEV1%, unchanged symptom scores (SGRQ) between Times 1 and 3, and frequent readmissions. An exacerbation represents an acute illness phase, influencing the patients both physically and psychologically. Over time, some improvements in the COPD patients’ physical condition would be expected. Study results indicate an improvement in

Table 4. Controlled effects of health status on QoL domains at T1 and T3 (n=51) WHOQOL – Bref Phys dom

Symptoms Impact Activity R2 adj.

Psych dom

Social dom

Envir dom

T1

T3

T1

T3

T1

T3

T1

T3

)0.02 )0.32* )0.23 15%

)0.06 )0.67*** )0.04 51%

)0.10 )0.15 )0.03 0%

0.02 )0.46* )0.08 20%

0.07 )0.03 0.11 0%

0.07 0.10 )0.31 0%

)0.03 )0.17 0.26 0%

0.01 )0.34 )0.19 18%

Presented as standardized beta weights and adjusted R2 from multiple linear regression analyses. * p< 0.05; **p< 0.01; *** p< 0.001.

255 health status, except for symptoms, over a 9 months period lending support to Spencer and Jones findings [27]. Their study differed from ours in some respects. They recorded SGRQ scores four times over a period of 26 weeks, whereas we did not assess health status between hospitalization and Time 3. Changes in psychological distress With respect to psychological distress, the scores remained high at all three time points, although an improvement from Time 1 to Time 2 indicates a stress recovery period the first month after discharge from hospital. The HSCL includes somatic symptoms, as well as psychological symptoms. Reduced psychological distress could be explained by an improvement in the patients’ physical condition and/or the reduction of stressful feelings connected with hospitalization per se. Nevertheless, the psychological distress scores remained high over time, indicating a persistent instability in the patients’ psychological condition during both acute and stable phases of the illness. These findings has been supported by other findings [28, 29]. Changes in QoL Further, the present study revealed low QoL. Compared to Norwegian population norms (mean age 45.7 years), the differences between the groups are considerably large in the physical and psychological domains [30]. The healthy population scored 15.78 compared with 11.13 in the COPD patients (Time 3). In the psychological domain, the population sample scored 15.16 and our COPD patients 13.58. Almost one-half (51%) of the variance in the physical domain was explained by the SGRQ impact, indicating that the psychosocial consequences of living with COPD are associated with physical health symptoms. With few exceptions, the mean WHOQOL – Bref domain scores were relatively stable over time. COPD is a progressive disease. Medical treatment during an exacerbation is primarily palliative and represents no fundamental change in the underlying, chronic condition of the patient. This may explain why the scores in the psychosocial domains remain unchanged, despite improvement in health status. Findings could also suggest that perceptions

of health status and quality of life represent two separate perceptual and cognitive processes in the minds of individuals, with separate factors influencing their conditions. Or perhaps health status and QoL evaluations are representative of more enduring stable personality characteristics. For example, self-ratings of health have been shown to be relatively stable, indicating a subjective state with its own health consequences. The low correlations between SGRQ and the WHOQOL – Bref domains at Time 1 support these suggestions. Further, the stability of the QoL scores could be attributed to the phenomenon termed response shift as proposed by Schwartz and Sprangers [31, 32]. They propose that when individuals report stable QoL in the face of major changes in health or life circumstances (=response shift), these responses may be due to changes occurring in one’s self-evaluation of QoL. Changes in selfevaluation are said to be influenced by changes in personal values, internal standards of measurement (i.e., scale recalibration) or by redefining of QoL. Thus, stable QoL scores could reflect changes in how the patients’ viewed their own situation, due to adaptation processes over time. For example, changes in a coefficients, seen in the SGRQ subscales and the WHOQOL – Bref, could be indicative of a scale recalibration. The low internal consistency at the initial assessment may reflect a considerable impact of random error variance (‘noise’). Further, learning effects may have occurred with subsequent assessments [33]. In our study, instruments were offered in the same sequence during all study points. The low correlations between the SGRQ subscales and the WHOQOL – Bref domain scores at Time 1 compared with the high correlations at Time 3 is difficult to explain. Perhaps peoples’ conception of their QoL is an independent process and not influenced by their present health status to the extent we usually presume. Or perhaps intervening variables like self-assessed health is operating latently between objective health symptoms and QoL [34]. Limitations Only two other studies have presented longitudinal data of COPD patients after a hospitalization [24, 25]. Taking into consideration the severity of the

256 COPD patients’ condition, the response rate of 55.4% in our study is acceptable. However, it should be kept in mind that a large number of patients were lost to follow-up, in spite of similar baseline characteristics. Because those who completed the study did not differ significantly from those who withdrew or dropped out, there are reasons to believe that the findings are representative of COPD patients being hospitalized for an exacerbation of COPD. With respect to the instruments, the responsiveness of the WHOQOL – Bref, i.e., the instrument’s ability to detect changes [1], needs further investigation. Our study, however, documented proof of its test–retest reliability, especially in the physical and psychological domains. The social domain of the WHOQOL – Bref include only three items and has demonstrated low a reliability in other studies [20, 35]. Future researchers are recommended to use complimentary measures in assessing social factors. Which factors influenced QoL during the study period? With respect to our model, attempts at causal modelling frequently seem to rest on dubious and conflicting assumptions. Various studies use the concepts of health status, psychological distress and QoL interchangeably as independent, moderating and dependent variables [11, 36]. Thus, one can question the logic of our causal model. Our results indicate that the impact subscale of the SGRQ influences both psychological distress and the physical and psychological domains of QoL. Both SGRQ impact and HSCL include questions concerning psychological distress, which may influence the relationship between them and their relationships with other dependent variables. This may represent a methodological tautology problem as well. The SGRQ impact subscale may be operationally redundant to the HSCL operationalizations and the physical and psychological domains of the WHOQOL – Bref, thus making interpretations of statistical associations between these scales ambiguous.

Conclusions Despite these limitations, the results from the present study suggest an improvement in the health status over a 9-month period from being

discharged after an exacerbation of COPD in spite of high levels of psychological distress and reduced QoL. Health care professionals might well accord greater attention to these elements of QoL. Acknowledgements A special thank to Petter Giæver, Chief Physician at the Pulmonary Department, Ulleva˚l University Hospital, for providing medical information concerning the participating patients. References 1. Fayers PM, Machin D. Quality of Life Assessment, Analysis and Interpretation. Chichester, West Sussex, England: John Wiley & Sons Ltd., 2000. 2. Williams SJ. Assessing the consequences of chronic respiratory disease: A critical review. Int Disabil Stud 1989; 11(4): 161–166. 3. Parsons EJ. Coping and well-being strategies in individuals with COPD. Health Val 1990; 14(3): 17–23. 4. Seemungal TA, Donaldson GC, Paul EA, Bestall JC, Jeffries DJ, Wedzicha JA. Effect of exacerbation on quality of life in patients with chronic obstructive pulmonary disease. Am J Respir Crit Care Med 1998; 157(5 Pt 1): 1418–1422. 5. Nicolson P, Anderson P. The patient’s burden: Physical and psychological effects of acute exacerbations of chronic bronchitis. J Antimicrob Chemother 2000; 45: 25–32. 6. Mikkelsen RL, Middelboe T, Pisinger C, Stage KB. Anxiety and depression in patients with chronic obstructive pulmonary disease (COPD). A review. Nord J Psychiatry 2004; 58(1): 65–70. 7. Aaron SD, Vandemheen KL, Clinch JJ, Ahuja J, Brison RJ, Dickinson G, Hebert PC. Measurement of short-term changes in dyspnea and disease-specific quality of life following an acute COPD exacerbation. Chest 2002; 121(3): 688–696. 8. Pauwels RA, Buist AS, Calverley PM, Jenkins CR, Hurd SS. Global strategy for the diagnosis, management, and prevention of chronic obstructive pulmonary disease. NHLBI/WHO Global Initiative for Chronic Obstructive Lung Disease (GOLD) Workshop summary. Am J Respir Crit Care Med 2001; 163(5): 1256–1276. 9. British Thoracic Society. BTS guidelines for the management of chronic obstructive pulmonary disease. The COPD Guidelines Group of the Standards of Care Committee of the BTS. Thorax 1997; 52(Suppl 5): S1–28. 10. Jones PW. Interpreting thresholds for a clinically significant change in health status in asthma and COPD. Eur Respir J 2002; 19(3): 398–404. 11. Doll H, Miravitlles M. Health-related QOL in acute exacerbations of chronic bronchitis and chronic obstructive pulmonary disease: A review of the literature. Pharmaco Economics 2005; 23(4): 345–363.

257 12. Jones PW, Quirk FH, Baveystock CM, Littlejohns P. A self-complete measure of health status for chronic airflow limitation. The St. George’s Respiratory Questionnaire. Am Rev Respir Dis 1992; 145(6): 1321–1327. 13. Jones PW, Bosh TK. Quality of life changes in COPD patients treated with salmeterol. Am J Respir Crit Care Med 1997; 155(4): 1283–1289. 14. Derogatis LR, Lipman RS, Rickels K, Uhlenhuth EH, Covi L. The Hopkins Symptom Checklist (HSCL): A selfreport symptom inventory. Behav Sci 1974; 19(1): 1–15. 15. Derogatis LR, Rickels K, Rock AF. The SCL-90 and the MMPI: A step in the validation of a new self-report scale. Br J Psychiatr 1976; 128: 280–289. 16. Roll-Hansen D. Samordnet leveka˚rsundersøkelse 1998 – tverrsnittsundersøkelsen. Dokumentasjonsrapport [Co-ordinated study of the living conditions 1998 – the cross-sectional study. Documentation report. Oslo: Statistisk sentralbyra˚ Statistics Norway, 1999. 17. Frojdh K, Hakansson A, Karlsson I, Molarius A. Deceased, disabled or depressed – A population-based 6-year follow-up study of elderly people with depression. Soc Psychiatr Psychiatr Epidemiol 2003; 38(10): 557–562. 18. Kaptein AA, Brand PL, Dekker FW, Kerstjens HA, Postma DS, Sluiter HJ. Quality-of-life in a long-term multicentre trial in chronic nonspecific lung disease: Assessment at baseline. The Dutch CNSLD Study Group. Eur Respir J 1993; 6(10): 1479–1484. 19. Sandanger I, Moum T, Ingebrigtsen G, Sorensen T, Dalgard OS, Bruusgaard D. The meaning and significance of caseness: The Hopkins Symptom Checklist-25 and the Composite International Diagnostic Interview. II. Soc Psychiatry Psychiatr Epidemiol 1999; 34(1): 53–59. 20. The WHOQOL Group. Development of the World Health Organization WHOQOL-BREF quality of life assessment. Psychol Med 1998; 28(3): 551–558. 21. Cronbach LJ. Coefficient alpha and the internal structure of test. Psychometrika 1951; 16: 297–334. 22. Wilson IB, Cleary PD. Linking clinical variables with health-related quality of life. A conceptual model of patient outcomes. JAMA 1995; 273(1): 59–65. 23. Finkel SE. Causal analysis with panel data. Thousand Oaks Calif: Sage, 1995. 24. Almagro P, Calbo E, Ochoa dE, Barreiro B, Quintana S, Heredia JL, Garau J. Mortality after hospitalization for COPD. Chest 2002; 121(5): 1441–1448. 25. Connors AFJ, Dawson NV, Thomas C, et al. Outcomes following acute exacerbation of severe chronic obstructive lung disease. The SUPPORT investigators (Study to Understand Prognoses and Preferences for Outcomes and Risks of Treatments) [published erratum appears in Am J

26.

27.

28.

29.

30.

31.

32.

33.

34. 35.

36.

Respir Crit Care Med 1997 Jan;155(1): 386]. Am J Respir Crit Care Med 1996; 154(4 Pt 1): 959–967. Seneff MG, Wagner DP, Wagner RP, Zimmerman JE, Knaus WA. Hospital and 1-year survival of patients admitted to intensive care units with acute exacerbation of chronic obstructive pulmonary disease. JAMA 1995; 20; 274(23): 1852–1857. Spencer S, Jones PW. Time course of recovery of health status following an infective exacerbation of chronic bronchitis. Thorax 2003; 58(7): 589–593. Gift AG, Shepard CE. Fatigue and other symptoms in patients with chronic obstructive pulmonary disease: Do women and men differ. J Obstet Gynecol Neonatal Nurs 1999; 28(2): 201–208. Yohannes AM. Depression and COPD in older people: A review and discussion. Br J Commun Nurs 2005; 10(1): 42–46. Hanestad BR, Rustoen T, Knudsen Ø, Lerdal A, Wahl A. Psychometric properties of the WHOQOL-BREF questionnaire for the Norwegian general population, (in press). Schwartz CE, Sprangers MA. Methodological approaches for assessing response shift in longitudinal health-related quality-of-life research. Soc Sci Med 1999; 48(11): 1531– 1548. Sprangers MA, Schwartz CE. Integrating response shift into health-related quality of life research: A theoretical model. Soc Sci Med 1999; 48(11): 1507–1515. Jansen SJ, Stiggelbout AM, Nooij MA, Noordijk EM, Kievit J. Response shift in quality of life measurement in early-stage breast cancer patients undergoing radiotherapy. Qual Life Res 2000; 9(6): 603–615. Moum T. Self-assessed health among Norwegian adults. Soc Sci Med 1992; 35(7): 935–947. O’Carroll RE, Smith K, Couston M, Cossar JA, Hayes PC. A comparison of the WHOQOL-100 and the WHOQOLBREF in detecting change in quality of life following liver transplantation. Qual Life Res 2000; 9(1): 121–124. Schmier JK, Halpern MT, Higashi MK, Bakst A. The quality of life impact of acute exacerbations of chronic bronchitis (AECB): A literature review. Qual Life Res 2005; 14(2): 329–347.

Address for Correspondence: Randi Andenæs, Faculty of Nursing Education, Oslo University College, P.O. Box 4, St. Olavs plass, N-0310 Oslo, Norway Phone: +47-22453768; Fax: +47-22453799 E-mail: [email protected]