3 Short- and long-term outcome of inpatient pulmonary rehabilitation in patients with asthma or chronic obstructive pulmonary disease Part 2: health status and psychosocial functioning

Henk F. van Stel 1 Vivian T. Colland 1 2 Lous H.M. Rijssenbeek-Nouwens 1 Jan M. Bogaard 3 Walter Everaerd 4

1) Asthmacentre Heideheuvel, Hilversum 2) Department of Health Psychology, Utrecht University, Utrecht 3) Department of Lung Diseases, Erasmus Medical Center, Rotterdam 4) Department of Clinical Psychology, University of Amsterdam, Amsterdam Submitted

Everywhere you look now - questionnaires They want every detail of your affairs Oh, so many peoples lives Wrapped up in those statistics Who are you? Give details. Manfred Mann's Earth Band / Andy Qunta Tribal Statistics, 1982

health status outcome of IPR

3.1

61

Abstract

Pulmonary rehabilitation improves health status in patients with stable chronic obstructive pulmonary disease (COPD). However, most programs exclude patients with unstable disease and patients with asthma. We studied the short- and long-term outcome of 3- to 6-month inpatient pulmonary rehabilitation (IPR) on health status and psychosocial functioning in patients with unstable, moderate to severe asthma or COPD. 56 patients with asthma and 84 patients with COPD were included in a prospective, observational study with follow-up at 6 and 12 months post-IPR. All domains of the Quality of Life for Respiratory Illness Questionnaire improved significantly (p from 0.002 to <0.00001) and relevant (effect sizes from 0.46 to 0.99) in both groups of patients. Two-third of the patients improved above the minimal clinically important difference. Emotional well-being and self-assessed health status also improved highly significant. Anxiety and depression improved marginally. In the year post-IPR, patients with COPD deteriorated significantly and relevantly in all HRQL-domains except social relations. Patients with asthma remained improved compared to pre-IPR. There was a high, but non-selective dropout. This IPR program resulted in large, clinically relevant improvements in health status and well-being. Patients with asthma remained improved in the year post-IPR; patients with COPD deteriorated towards the pre-IPR level.

3.2

Introduction

A major goal of pulmonary rehabilitation is to improve health status. Both in- and outpatient pulmonary rehabilitation succeed in improving health status and functional exercise tolerance in patients with chronic obstructive pulmonary disease (COPD). Both patients with asthma and patients with COPD benefit from outpatient pulmonary rehabilitation [1]. Several recent studies showed that patients with COPD remained improved after outpatient pulmonary rehabilitation [2-4]. Long-term outcome of inpatient pulmonary rehabilitation (IPR) is less positive: patients with COPD deteriorate, after initial improvement, to the pre-IPR level [5-7]. Despite this, patients with COPD receiving IPR remain improved in comparison with controls receiving conventional community care [8]. Knowledge about the short- and long-term outcome of IPR in patients with asthma is scarce: we found only two studies on IPR that also included some patients with asthma [9;10]. Furthermore, most programs exclude patients with unstable disease or with comorbid conditions. We studied the short- and long-term outcome of IPR on health status and psychosocial

62

chapter 3

functioning in patients with unstable, moderate to severe asthma or COPD. This study is part of a prospective observational study on multidisciplinary inpatient pulmonary rehabilitation; outcome on lung function, medication usage, hospitalization and functional exercise tolerance are described in the companion article (Part 1). This part contains baseline characteristics; short-term outcome of IPR (i.e. pre/posttreatment change) and long-term outcome of IPR with follow-up measurements at 6 and 12 months post-IPR.

3.3

Methods

The study design and patient selection are described in Part 1 of this study (see chapter 2). 3.3.1 Outcome measures: health status and psychosocial aspects Disease-specific health status was assessed with the Quality of Life for Respiratory Illness Questionnaire (QoLRIQ) [11], an outcome measure for both patients with asthma and patients with COPD. This questionnaire consists of 55 items divided into seven domains: breathing problems (9 items), physical problems (9), emotions (9), situations triggering or enhancing breathing problems (7), general activities (4), daily and domestic activities (10), and social activities, relationships and sexuality (7). The domain on triggering situations was split in allergic triggers and triggers related to weather because of difference in change in these subdomains. The domain on social activities was split in activities and relations/sexuality because of the large amount of missings in the relations/sexuality subdomain, caused by elderly patients without a partner or by patients unwilling to answer on questions related to sexuality. The QoLRIQ uses a 7–point response scale ranging from “not at all” to “very severe” to assess the degree of trouble from symptoms or impediment in carrying out activities. A higher score represents a higher level of impairment. The minimal important difference (MID) has been estimated at 0.5 units (see chapter 6), which is similar to other questionnaires with 7–point response scales such as the Chronic Respiratory Questionnaire (CRQ) [12;13] and the Asthma Quality of Life Questionnaire (AQLQ) [14;15]. Self-perceived health status was assessed with a one-item question (“How would you rate your health status at this moment”: very good, good, fair, moderate, poor). This question was slightly modified from the Netherlands Health Survey Interview [16]: we changed the category “sometimes good, sometimes bad” into “moderate” due to misunderstandings of the former wording in a pilot study. Psychosocial functioning was assessed with the Medical Psychological Questionnaire for Lung Patients (MPQL) [17;18] for emotional well-being and experienced invalidity and the

health status outcome of IPR

63

Symptom Checklist 90 (SCL-90) [19] for anxiety and depression. The MPQL is validated in patients with asthma or COPD, including both outpatients (90%) and inpatients (10%). Raw scores can be recoded to 5 normative categories, ranging from very favorable to very unfavorable. The SCL-90 is validated in both a ‘normal’ population and a group of outpatient psychiatric patients. Raw scores can be recoded to 7 categories. Coping strategies were assessed with the CNSLD Coping Questionnaire (CCQ). The CCQ is modified from the Asthma Coping Questionnaire [20] and suitable for both patients with asthma and patients with COPD. Because the CCQ is not yet fully validated, we computed the internal consistency (Cronbach’s α). The CCQ measures avoidance (α=0.86), rational action (α=0.77) and emotional reaction (α=0.64). The CCQ asks if the respondent showed the behavior described hardly ever, sometimes, often or very often. The values for internal consistency are similar to those in the study by Ketelaars et al. [6]. In addition to the global ratings on clinical features (see chapter 2), patients were asked to rate self-perceived change in performing activities of daily living, performing social activities, and performing leisure activities on a 5–point response scale: “much improved – improved – the same – worse – much worse”. 3.3.2 Statistical Analysis The hypothesis for this study was an pre-posttreatment improvement of 0.5 units (=MID) on the QoLRIQ domains, with a null-hypothesis of no change in health status. Based on standard deviations obtained from a pilot study, a sample size of 80 patients was needed. This number would also be sufficient to detect an effect size of 0.5 in anxiety and depression. Because of the high dropout, the inclusion period was enlarged to two years. The dropout pattern is described in Part 1 (see chapter 2). All statistical analyses were performed with Statistica for Windows 5.1 (StatSoft, Tulsa, OK). 3.3.2.1 Interpretation of change We used three indicators for change: statistical significance of change, effect size and the MID. Statistical significance of pre/post-IPR change was assessed with the Wilcoxon matched pairs test or the paired t-test. Statistical significance was accepted at α=0.05. Long-term change was evaluated with repeated-measures ANOVA [21]. Significant ANOVA’s were further explored with post-hoc testing using Tukey’s HSD (honest significant difference) test. Interpretation of long-term change consists of two questions. The first question is whether the 6- and 12-months follow-up scores differ relevantly from the pre-IPR scores; this was assessed by comparing pre-IPR scores with 6- and 12 month scores (absolute change for the QoLRIQ-domains, effect size for other questionnaires). The second question is how much patients change (i.e. deteriorate) in the follow-up period (post-IPR scores versus 6- and

64

chapter 3

12 month scores, using the baseline standard deviation for allowing comparison with other effect sizes). To assess the relevance of the observed pre/post-IPR change, absolute differences (for the QoLRIQ) or effect sizes (for other questionnaires) were computed. The effect size is computed as the change score divided by the baseline standard deviation [22]. An effect size of 0.2 represents a small change, 0.5 is a moderate change and 0.8 or higher is a large relevant change [23]. We showed in chapter 6 of this dissertation that the MID for the QoLRIQ is 0.5 units on a 7-point response scale. 90% confidence intervals (90%CI) were computed for the pre/post-IPR mean differences [24]. We checked if the lower limit of the 90%CI [25;26] was above the MID of the QoLRIQ. This enables the interpretation of the clinical relevance and statistical probability of the observed changes. The MID also simplifies interpretation of long-term change. We used one norm-level improvement as MID for the SCL-90 and MPVC. MIDs for the SCL-90 and MPVC are unknown and difficult to assess because the normative levels differ in size, which implies that a certain amount of change has a different meaning depending on its place in the score-range. Jacobson et al. [27] suggested improvement back to normal or normative functioning as a clinically relevant change. Because we did not expect such large changes in severely ill patients, our suggestion is to use one norm-level improvement as MCID for these questionnaires and to compute the percentage of patients who improved or deteriorated at least one norm-level.

3.4

Results

3.4.1 Baseline description General characteristics are shown in table 3.1 of Part 1(see chapter 2). The study group consisted of 56 patients with asthma and 84 patients with COPD. Patients with COPD had significantly worse scores that patients with asthma on daily/domestic activities (p<0.0001), social activities (p=0.04) and experienced invalidity (p<0.0001); but better scores on triggering situations (p=0.03) (see table 3.1). Analysis of the MPQL-scores, which were recoded to 5 normative categories, showed 57% to 83% unfavorable and very unfavorable scores on emotional well-being and experienced invalidity (see table 3.2). A similar analysis of the SCL-90 shows that about 50% of the patients had a high or very high score on anxiety and depression, as compared to the general population. Comparing questionnaire scores found in this study with values from other studies in the Netherlands showed that the patients with COPD had a higher level of anxiety (p=0.06) and depression (p=0.04) than outpatient-treated patients with COPD [28], and more

health status outcome of IPR

65

complaints on general activities, daily/domestic activities, social relations and QoLRIQ-total score than outpatients (all p<0.005) [29]. Patients with asthma had worse scores than outpatients with asthma on all domains of the QoLRIQ (p<0.05 to p<0.00001) [29]. Scores on the CCQ were similar to those in the IPR-study by Ketelaars and coworkers [6]. 3.4.2 Non-response analysis Dropout rates are described in Part 1. There were no significant differences in dropout rates between patients with asthma and patients with COPD or between levels of severity. There were only a few differences between IPR-completers, treatment dropouts and study dropouts; from which three reached significance at post-hoc testing. Study dropouts had a worse score than completers on triggering situations (asthma only, p=0.03), a worse score on self-assessed health status than treatment dropouts (asthma only, p=0.02), and a better score on experienced invalidity than both treatment dropouts and completers (COPD only, p=0.02 and 0.04). There were no significant differences in health status or psychosocial functioning between study-completers and follow-up dropouts. 3.4.3 Pre-post IPR change All domains of the QoLRIQ, except for the subdomain allergic triggers, showed highly significant and clinically relevant changes, both in patients with asthma and patients with COPD (see table 3.1). Mean change was at or above the MID of 0.5 units in all domains (see figure 3.1). In several domains, the lower limit of the 90%CI was also above the MID (see figure 3.1). About two-third of the patients who completed the program had an improvement in QoLRIQ-total score above the MID (asthma: 64.1%; COPD: 63.2%) (see table 3.1). When all patients who started with IPR are taken into account (intention to treat), about half of the patients improved above the MID (asthma: 44.6%; COPD: 51.2%). This results in a number needed to treat of 2.2 for asthma and 1.95 for COPD.

66

chapter 3

Table 3.1a : Pre-post-treatment change in health status: asthma (N=39) QoLRIQ-domains

pre-IPR

post-IPR

p(I)

p(D)

p value change

ES

Breathing problems

3.5 (1.3)

2.9 (1.1)

56.4%

15.0%

0.004

0.48

Physical problems

3.4 (1.1)

2.8 (1.3)

61.5%

5.1%

<0.0001

0.62

Emotions

3.2 (1.3)

2.2 (1.0)

59.0%

5.1%

<0.0001

0.71

General activities

4.0 (1.4)

2.6 (1.3)

69.2%

7.7%

0.0001

0.94

Triggering situations

3.5 (1.1)

3.0 (1.1)

48.7%

12.8%

0.002

0.46

Triggers: weather

4.2 (1.3)

3.3 (1.3)

64.1%

10.3%

<0.0001

0.76

Triggers: allergic

2.6 (1.6)

2.5 (1.5)

33.3%

28.2%

0.7

-0.03

Daily/domestic activities

3.8 (1.2)

3.0 (1.3)

62.9%

14.3%

0.001

0.58

Social activities *

3.8 (1.7)

2.7 (1.5)

56.5%

8.7%

0.002

0.52

Social: activities

4.5 (1.8)

3.3 (1.7)

69.2%

15.4%

<0.0001

0.68

Social: relations *

3.4 (1.8)

2.1 (1.5)

60.9%

0%

0.0003

0.58

3.6 (1.0)

2.8 (1.0)

64.1%

7.7%

<0.0001

0.83

QoLRIQ-total score

Values are presented as mean (standard deviation). Lower scores indicate better functioning. p(I)= proportion improving; p(D) = proportion deteriorating (i.e. percentage of patients with change score ≥ MID); ES=effect size. * N =23

67

health status outcome of IPR

Table 3.1b : Pre-post-treatment change in health status: COPD (N=68) QoLRIQ-domains

pre-IPR

post-IPR

p(I)

p(D)

p value change

ES

Breathing problems

3.6 (1.0)

2.8 (1.0)

57.4%

11.8%

<0.0001

0.76

Physical problems

3.4 (1.0)

2.6 (1.0)

58.8%

5.9%

<0.0001

0.73

Emotions

3.4 (1.2)

2.6 (1.2)

55.9%

10.3%

<0.0001

0.68

General activities

4.5 (1.4)

3.1 (1.4)

72.1%

8.8%

<0.0001

0.95

Triggering situations

3.4 (1.1)

2.9 (1.2)

45.6%

16.2%

0.0005

0.45

Triggers: weather

4.2 (1.2)

3.5 (1.3)

54.4%

13.2%

0.0002

0.59

Triggers: allergic

2.3 (1.5)

2.2 (1.4)

29.4%

19.1%

0.4

0.08

Daily/domestic activities

5.0 (1.4)

3.9 (1.6)

62.3%

6.6%

<0.0001

0.81

Social activities *

4.5 (1.6)

3.8 (1.7)

55.0%

25.0%

0.008

0.49

Social: activities

4.9 (1.7)

4.1 (1.7)

52.5%

21.3%

0.004

0.48

Social: relations *

4.1 (1.9)

3.4 (1.9)

65.0%

20.0%

0.008

0.36

3.9 (0.8)

3.1 (1.0)

63.2%

8.8%

<0.0001

0.99

QoLRIQ-total score

Values are presented as mean (standard deviation). Lower scores indicate better functioning. p(I)= proportion improving; p(D) = proportion deteriorating (i.e. percentage of patients with change score ≥ MID); ES=effect size. * N =40

Both groups improved significantly and clinically relevant in emotional well-being and selfassessed health status (see table 3.2), while patients with COPD also improved in experienced invalidity. Both groups improved non-significantly in anxiety and depression. Rating of self-perceived change showed improvement in performance of activities (daily living, social, leisure) in 53% to 62% of the patients. Both groups showed significant and relevant changes in coping behavior: less avoidance and more emotional reaction (see table 3.3). There was no change in rational action.

68

chapter 3

Table 3.2: Pre-post-treatment change in psychological domains Asthma (n=39)

before

% high score a

after

p value change

ES b

improved ≥1 level

emotional well-being

18 (12)

73.3%

28 (12)

0.007

0.73

42.2%

experienced invalidity

27 (4)

57.7%

26 (6)

0.1

0.46

28.9%

anxiety

20 (12)

51.9%

16 (10)

0.04

0.22

31.8%

depression

33 (19)

51.9%

25 (14)

0.04

0.37

50.0%

self-assessed health status

4 (1)

3 (2)

<0.0001

1.11

COPD (n=68) emotional well-being

19 (8)

83.1%

31 (16)

<0.0001

1.44

55.8%

experienced invalidity

31 (5)

81.8%

29 (7)

0.0004

0.81

36.4%

anxiety

17 (8)

44.4%

15 (9)

0.1

0.08

37.1%

depression

27 (12)

60.5%

24 (13)

0.07

0.18

45.2%

self-assessed health status

4 (1)

3 (1)

<0.0001

1.29

Values are presented as median (interquartile range) a % of patients in two highest norm levels. Thresholds for high score: well-being: unfavourable ≤26; experienced invalidity: unfavourable ≥26; anxiety: high≥15 (men) or ≥18 (women); depression: high≥23 (men) or ≥28 (women) b ES = effect size (mean change divided by baseline standard deviation) Lower scores indicate better functioning, except for emotional well-being where higher scores indicate better functioning

69

health status outcome of IPR

Table 3.3: Pre-post-treatment change in coping domains Asthma (n=39)

before

after

p value change

ES

avoidance

2.6 (0.5)

2.3 (0.6)

0.0008

0.58

rational action

2.5 (0.4)

2.6 (0.3)

0.3

0.26

emotional reaction

1.7 (0.5)

2.8 (0.5)

<0.00001

2.24

avoidance

2.4 (0.5)

2.2 (0.5)

0.0001

0.5

rational action

2.6 (0.4)

2.6 (0.3)

0.5

0.08

emotional reaction

1.8 (0.5)

2.9 (0.5)

<0.00001

2.33

COPD (n=68)

Values are presented as mean (standard deviation). Range: 1= hardly ever, 4=very often. ES=effect size

2.0

asthma COPD

1.5

1.0

0.5

0.0

breathing problems

physical problems

emotions

general activities

trigger situations

daily activities

social activities

QoLRIQ total

Figure 3.1: Mean pre-posttreatment improvement (box) with 90% confidence interval (whiskers) in health status domains. The dashed line represents the minimal important difference.

70

chapter 3

3.4.4 One-year follow-up Repeated measures-ANOVA’s gave significant time effects for all domains and total score of the QoLRIQ, both with or without the 12-month assessment (see table 3.4 and 3.5). The only exception was triggering situations at 12 months in patients with asthma. In patients with asthma, three domains (emotions, general activities, social relations) and the total score were both at 6 and 12 months follow-up significantly and/or clinically relevant improved compared to pre-treatment scores (see table 3.4). Triggering situations and daily/domestic activities were still better than pre-IPR at 6 months, but not at 12 months. There were three domains showing clinically relevant but non-significant deterioration in the 12-month follow-up period in the asthma group (breathing problems, physical problems and general activities). In patients with COPD, the 6 and 12-month follow-up scores of 4 domains (emotions, general activities, daily/domestic activities, social relations) still differ significantly and/or relevantly from the pre-IPR score (see table 3.5). All domains (except social relations) show significant and/or clinically relevant deterioration when comparing follow-up with post-treatment scores. There was a similar pattern in emotional well-being and experienced invalidity (see table 3.6). Emotional well-being in patients with asthma was at both 6 and 12 months follow-up significantly and relevantly better than pre-IPR, without deterioration. Emotional well-being in patients with COPD was still better than pre-IPR at 6 months follow-up, but not at 12 months follow-up. Furthermore, in the follow-up period patients with COPD showed significant and relevant deterioration both in emotional well-being and experienced invalidity. Follow-up scores for experienced invalidity from patients with COPD did not differ from the pre-IPR scores. Repeated-measures Anova’s for anxiety and depression were not significant, except for depression in patients with asthma (F=3.4, p=0.02) with a marginally significant (HSD, p=0.04) effect size of 0.44 between pre-treatment and 12month scores. The initial change in the coping style ‘emotional reaction’ disappeared completely in both groups of patients. Post-hoc analysis of the highly significant repeatedmeasures Anova’s (all p<0.00001) showed that post-treatment scores differed from pretreatment and follow-up scores (HSD, all p=0.0001), in both groups. The coping style ‘avoidance’ showed a similar pattern that was only significant in patients with COPD.

71

health status outcome of IPR

Table 3.4: Long-term change in health status, asthma domain

follow-up period

Fvalue

p-value

pre-IPR / follow-up

post-IPR / follow-up

Diff

p HSD

Diff

p HSD

breathing problems

6 months

3.6

0.03

0.24

0.6

-0.4

0.2

12 months

3.0

0.04

0.14

0.9

-0.54

0.1

physical problems

6 months

7.0

0.002

0.46

0.08

-0.31

0.03

12 months

4.8

0.006

0.36

0.4

-0.5

0.2

emotions

6 months

16.5

<0.001

0.93

<0.001

-0.13

0.8

12 months

11.2

<0.001

0.85

<0.001

-0.38

0.3

general activities

6 months

17.4

<0.001

1.21

<0.001

-0.17

0.8

12 months

9.6

<0.001

0.79

0.04

-0.67

0.1

triggering situations

6 months

5.9

0.004

0.49

0.03

-0.11

0.8

12 months

1.3

0.3

daily/domestic activities

6 months

9.0

0.001

0.79

0.001

0.06

0.9

12 months

6.1

0.001

0.37

0.3

-0.36

0.3

social relations

6 months

2.9

0.07

0.68

0.08

0.13

0.9

12 months

3.6

0.03

0.89

0.03

0.16

0.9

QoLRIQ-total

6 months

15.8

<0.001

0.67

<0.001

-0.16

0.6

12 months

9.3

0.001

0.5

0.02

-0.35

0.9

N at 6 months: 29; N at 12 months: 17. Repeated measurements ANOVA with post-hoc significance testing (p-value of HSD, Tukey’s honest significant difference). The MID for difference (Diff) is 0.5 units. Minus sign indicates deterioration.

72

chapter 3

Table 3.5: Long-term change in health status, COPD domain

follow-up period

Fvalue

p-value

pre-IPR / follow-up

post-IPR / follow-up

Diff

p HSD

Diff

p HSD

breathing problems

6 months

17.8

<0.001

0.38

0.08

-0.6

0.001

12 months

7.8

<0.001

0.16

0.8

-0.73

0.002

physical problems

6 months

15.8

<0.001

0.21

0.4

-0.61

<0.001

12 months

9.3

<0.001

0.1

0.9

-0.77

<0.001

emotions

6 months

10.6

<0.001

0.46

0.06

-0.45

0.06

12 months

6.4

0.001

0.65

0.07

-0.45

0.3

general activities

6 months

24.3

<0.001

0.7

0.002

-0.7

0.002

12 months

13.4

<0.001

0.56

0.2

-1.13

<0.001

triggering situations

6 months

9.9

<0.001

0.13

0.7

-0.49

0.004

12 months

4.3

0.007

-0.03

0.9

-0.62

0.01

daily/domestic activities

6 months

20.7

<0.001

0.6

0.007

-0.65

0.004

12 months

13.4

<0.001

0.7

0.03

-0.83

0.006

social relations

6 months

3.9

0.03

0.68

0.2

-0.3

0.7

12 months

2.8

0.05

0.8

0.3

-0.32

0.9

6 months

22.7

<0.001

0.4

0.2

-0.56

<0.001

12 months

12.2

<0.001

0.33

0.3

-0.74

<0.001

QoLRIQ-total

N at 6 months: 49; N at 12 months: 31. Repeated measurements ANOVA with post-hoc significance testing (p-value of HSD, Tukey’s honest significant difference). The MID for difference (Diff) is 0.5 units. Minus sign indicates deterioration

73

health status outcome of IPR

Table 3.6: Long-term change in psychosocial functioning. domain

emotional wellbeing

experienced invalidity

diagnosis

followup

N

Fvalue

p-value

pre-IPR vs. follow-up

post-IPR vs. follow-up

ES

HSD

ES

HSD

asthma

6

29

9.56

<0.001

0.75

0.02

-0.27

0.4

asthma

12

18

6.21

0.001

0.91

0

-0.14

0.9

COPD

6

49

21.86

<0.001

0.6

0.02

-0.84

<0.001

COPD

12

32

11.86

<0.001

0.5

0.1

-0.78

<0.001

asthma

6

29

3.27

0.04

0.3

0.5

-0.4

0.3

asthma

12

18

1.47

0.2

COPD

6

49

8.73

<0.001

0.26

0.7

-0.89

0.006

COPD

12

32

3.77

0.01

0.25

0.1

-0.78

0.003

Repeated measures Anova with post-hoc significance testing (p-value of HSD, Tukey’s honest significant difference). Minus sign indicates deterioration. ES=effect size.

3.5

Discussion

The unfavorable scores on health status and psychosocial functioning show that the patients in this study are highly impaired, in addition to the clinical severity of illness and instability found in Part 1 (see chapter 2). A major goal of the IPR-programme is to improve daily functioning. IPR resulted in improvements in all HRQL-domains, both in patients with asthma and patients with COPD. The majority of the patients improved above the MID [30], indicating clinically relevant changes. Intention-to-treat analysis gave a number needed to treat of about 2. Large improvements were also seen in self-assessed health status and emotional well-being. The improvements in anxiety and depression were smaller than expected and in contrast to the large changes in the emotions-domain of the QoLRIQ and in emotional well-being. The generic nature of the questionnaire we used (SCL-90) may be an explanation for the small changes. Coping behavior also changed: there was less avoidance and more use of the coping style emotional reaction. There was an important difference in long-term change between the two groups of patients. Patients with COPD deteriorated significantly and clinically relevant in the follow-up period in most HRQL-domains, emotional well-being and experienced invalidity. Only domains related to daily functioning remained improved in comparison to the pre-IPR level. Patients with asthma showed some non-significant deterioration in health status and emotional well-

74

chapter 3

being in the follow-up period, but remained significantly and clinically relevant improved in comparison to their pre-IPR status. The initial changes in coping behavior disappeared in both groups. Non-response analysis gave no clear indication of selective dropout, as was found by Ketelaars and coworkers [6]. There were some significant differences, but not in the expected direction, i.e. worse scores for treatment/illness related dropouts. We observed an interesting difference in between clinical/physiological and HRQL/psychosocial outcome measures. Significant differences between completers and dropouts in clinical/physiological variables were only found in the follow-up phase of the study, while differences in HRQL/psychosocial variables were only found in the IPR-phase. The positive effects of outpatient and home-based pulmonary rehabilitation on health status and exercise tolerance of patients with COPD are well documented [26]. Pulmonary rehabilitation in patients with asthma and the results of inpatient pulmonary rehabilitation have received far less attention. One study on physiotherapy practice-based pulmonary rehabilitation showed that patients with asthma show similar improvements as patients with COPD [31]. Studies on IPR with patients with COPD have consistently shown improvements in health status and exercise tolerance [5-7;32-37]. The only randomized controlled trial on IPR [8] showed that an 8-week IPR programme for patients with stable COPD results in sustained improvement in comparison with conventional care. IPR-studies with follow-up [5-7;35] showed that patients with COPD deteriorate to the pre-IPR level, which is similar to the results of the present study. A major finding of our study is that patients with asthma showed lasting improvements in health status and emotional well-being. Another interesting finding is the change in coping behavior, which was not found in another study on IPR with the same questionnaire [6]. This change may be explained by the focus on training of self-management skills and modification of disease behavior. The control-group problem is not unique for this study: only one randomized trial of IPR versus standard care has been published [8]. That study was feasible because of a 3 to 6 month waiting list and by excluding unstable patients [38]. The severity of illness of patients referred for IPR has been used as an argument against randomization [39]. Because our study had no control group, we searched for alternative methods to assess the statistical probability and clinical relevance of the observed changes. The first method is to control for multiple testing by decreasing the level for accepting significance to a very stringent level. A p-value of 0.001 provides reasonable evidence against the null-hypothesis [40]. Self-assessed health status, most QoLRIQ-domains,

health status outcome of IPR

75

emotional reaction and in patients with COPD also emotional well-being and experienced invalidity, improved with p-values of 0.001 or lower. A second method is to check if the lower limit of the confidence interval is above the MID of 0.5 units, as suggested by Jones [25]. This was true for several QoLRIQ-domains. This suggests that there is a clinically relevant improvement in almost all patients, which also shows from the high proportion of patients benefitting. A third method, related to the large number of dropouts, is to perform imputation of missing data. Imputation is used to check if something changes in the magnitude and significance of the observed difference when all data are used [41]. A combined approach of sensitivity analysis and imputation, which is described in chapter 4, shows that the dropout did not distort the study findings. Even when assuming a worst case scenario, i.e. deterioration for the treatment dropouts and no improvement for the study dropouts, the short-term improvement in QoLRIQ-total score remained significant in both groups. The change was also clinically relevant (above the MID) in all scenarios, except for the worst case scenario in patients with asthma. We found a contradiction between the diminishing of the initial treatment effect on health status and psychosocial functioning and the large decrease in hospital admissions post-IPR in patients with COPD. It should be studied why patients report deterioration in quality of life and well-being while they are clearly improved on clinical parameters. The quick deterioration of health status in patients with COPD in the follow-up period is worrying, both from the perspective of the patients involved and from the perspective of third-payer parties. Therefore, the effectiveness of methods for retaining the initial treatment effect, such as booster sessions, after-care programs [6] or repeating pulmonary rehabilitation [42], should be studied.

76

3.6

chapter 3

Conclusion

The patients referred for IPR showed highly impaired daily and psychosocial functioning. The IPR-programma resulted in large, clinically relevant improvements in health status, selfassessed health status and emotional well-being in both patients with asthma and patients with COPD. Patients with asthma remained improved in the year after IPR. Patients with COPD deteriorated in all domains, often to the pre-IPR level. Only daily functioning remained partially improved in patients with COPD. Two-third of the patients who completed the program and half of the patients who started IPR had an improvement in QoLRIQ-total score above the MID, which results in a number needed to treat of 2.

3.7

Overall conclusion part 1 and 2

This comprehensive study reported on baseline characteristics, pre/posttreatment change and long-term outcome of inpatient pulmonary rehabilitation in both patients with asthma and patients with COPD. The study group consisted of clinically unstable and highly impaired patients with a very high pre-IPR hospitalization rate. There was a high, but nonselective dropout. The major short-term findings of this study are the large, clinically relevant improvements in health status and psychosocial functioning, and the decreased use of oral corticosteroids. Patients reported improved exercise tolerance, which was not accompanied by a mean change in walking distance. The major long-term findings of this study are the four- to six-fold decrease in hospitalization in the follow-up period and the lasting improvements in health status and emotional well-being in patients with asthma in the year after IPR. Health status in patients with COPD deteriorated to the pre-IPR level, as was found in other studies.

health status outcome of IPR

3.8 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11.

12. 13. 14. 15. 16. 17. 18. 19. 20.

77

Reference List Cambach W, Wagenaar RC, Koelman TW, van Keimpema AR, Kemper HC. The long-term effects of pulmonary rehabilitation in patients with asthma and chronic obstructive pulmonary disease: a research synthesis. Arch Phys Med Rehabil 1999; 80:103-111. Griffiths TL, Burr ML, Campbell IA, Lewis-Jenkins V, Mullins J, Shiels K et al. Results at 1 year of outpatient multidisciplinary pulmonary rehabilitation: a randomised controlled trial. Lancet 2000; 355:362-368. Troosters T, Gosselink R, Decramer M. Short- and long-term effects of outpatient rehabilitation inpatients with chronic obstructive pulmonary disease: a randomized trial. Am J Med 2000; 109:207-212. Guell R, Casan P, Belda J, Sangenis M, Morante F, Guyatt GH et al. Long-term effects of outpatient rehabilitation of COPD: A randomized trial. Chest 2000; 117:976-983. Guyatt GH, Berman LB, Townsend M. Long-term outcome after respiratory rehabilitation. CMAJ 1987; 137:1089-1095. Ketelaars CA, Abu-Saad HH, Schlösser MAG, Mostert R, Wouters EFM. Long-term outcome of pulmonary rehabilitation in patients with COPD. Chest 1997; 112:363-369. Buchi S, Villiger B, Sensky T, Schwarz F, Wolf C, Buddeberg C. Psychosocial predictors of long-term success of in-patient pulmonary rehabilitation of patients with COPD. Eur Respir J 1997; 10:12721277. Goldstein RS, Gort EH, Stubbing D, Avendano MA, Guyatt GH. Randomised controlled trial of respiratory rehabilitation. Lancet 1994; 344:1394-1397. van der Schoot TAW, de Weerdt I, Kaptein AA, Dekker FW, Deenen ThAM, Speelberg B. Favorable effects of a stay in the Dutch Asthma Center Davos on medical consumption and quality of life in CNSLD patients. [Dutch]. Ned Tijdschr Geneeskd 1993; 137:197-201. Bijl D, Speelberg B, Folgering HThM. Pulmonary rehabilitation at moderate altitude: a 1-year followup. Netherlands Journal of Medicine 1994; 45:154-161. Maillé AR, Koning CJM, Zwinderman AH, Willems LN, Dijkman JH, Kaptein AA. The development of the 'Quality-of-life for Respiratory Illness Questionnaire (QOL-RIQ)': a disease-specific quality-oflife questionnaire for patients with mild to moderate chronic non- specific lung disease. Respir Med 1997; 91:297-309. Guyatt GH, Berman LB, Townsend M, Pugsley SO, Chambers LW. A measure of quality of life for clinical trials in chronic lung disease. Thorax 1987; 42:773-778. Jaeschke R, Singer J, Guyatt GH. Measurement of health status. Ascertaining the minimal clinically important difference. Control Clin Trials 1989; 10:407-415. Juniper EF, Guyatt GH, Epstein RS, Ferrie PJ, Jaeschke R, Hiller TK. Evaluation of impairment of health related quality of life in asthma: development of a questionnaire for use in clinical trials. Thorax 1992; 47:76-83. Juniper EF, Guyatt GH, Willan A, Griffith LE. Determining a minimal important change in a diseasespecific Quality of Life Questionnaire. J Clin Epidemiol 1994; 47 :81-87. König-Zahn C, Furer JW, Tax B. Health status measurement: description and evaluation of questionnaires. Part 1: General Health (Het meten van de gezondheidstoestand: beschrijving en evaluatie van vragenlijsten. 1: Algemene gezondheid). Assen: van Gorcum, 1993. Erdman RAM, Cox NJM, Duivenvoorden HJ. Handleiding Medisch Psychologische Vragenlijst voor CARA-patienten [User manual Medical Psychological Questionnaire for Lung patients]. Lisse: Swets & Zeitlinger, 1992. Cox NJM, Hendriks JC, Dijkhuizen R, Binkhorst RA, van Herwaarden CLA. Usefulness of a medicopsychological questionnaire for lung patients. Int J Rehabil Res 1991; 14:267-272. Arrindell WA, Ettema JHM. SCL-90 Manual for a multidimensional indicator for psychopathology [handleiding bij een multidimensionele psychopathologie-indicator]. Lisse: Swets & Zeitlinger bv., 1986. Schlösser MAG. Self-management and Asthma. Thesis. 1 ed. Leiden: DWSO Press, 1992.

78 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42.

chapter 3 Stevens J. Applied multivariate statistics for the social sciences. 3 ed. Mahwah, NJ: Lawrence Erlbaum Associates, 1996. Kazis LE, Anderson JJ, Meenan RF. Effect sizes for interpreting changes in health status. Med Care 1989; 27:S178-S189. Cohen J. Statistical power analysis for the behavioral sciences. 2 ed. Hillsdale: Lawrence Erlbaum Associates, 1988. Altman DG, Machin D, Bryant TN, Gardner MJ. Statistics with confidence. 2 ed. London: BMJ Books, 2000. Jones PW. Health status measurement in chronic obstructive pulmonary disease. Thorax 2001; 56:880-887. Lacasse Y, Wong E, Guyatt GH, King D, Cook DJ, Goldstein RS. Meta-analysis of respiratory rehabilitation in chronic obstructive pulmonary disease. Lancet 1996; 348 :1115-1119. Jacobson NS, Follette WC, Revenstorf D. Psychotherapy outcome research: methods for reporting variability and evaluating clinical significance. Behavior Therapy 1984; 15:336-352. Kaptein AA, Dekker FW, Dekhuijzen PNR, Wagenaar JPM, Janssen PJ. Patiënten met chronische luchtwegobstructie - Scores op NPV en SCL-90 en hun relaties met functionele capaciteit en belemmeringen in dagelijkse activiteiten. Gezondheid & Samenleving 1986; 7:10-19. Maillé AR. Quality of Life in Asthma and COPD. Development of a disease-specific questionnaire (thesis). University of Amsterdam, Amsterdam: University of Amsterdam, 2000. van Stel HF, Maillé AR, Colland VT, Everaerd W. Interpretation of change and longitudinal validity of the Quality of Life for Respiratory Illness Questionnaire (QoLRIQ) in pulmonary rehabilitation. Quality of Life Research, in press. Cambach W. Rehabilitation of patients with asthmaand mild to moderate chronic obstructive pulmonary disease in local physiotherapy practices. A study on the efficacy of physiotherapy practices-based pulmonary rehabilitation in Amsterdam. 1999. Cox NJM. Effects of a pulmonary rehabilitation programme in patients with obstructive lung diseases (in comparison with a control group). Dissertation. Nijmegen: Katholieke Universiteit Nijmegen, 1990. Votto JJ, Bowen J, Scalise P, Wollschlager C, ZuWallack RL. Short-stay comprehensive inpatient pulmonary rehabilitation for advanced chronic obstructive pulmonary disease. Arch Phys Med Rehabil 1996; 77:1115-1118. van der Schoot TAW. Kwaliteit van leven van CARA-patiënten in het Nederlands Astmacentrum Davos. Beloop en determinanten (Quality of life of patients with CNSLD in the Dutch Asthmacentre in Davos. Course and determinants). Universitaire Pers Maastricht, 1996. Rooyackers JM. Pulmonary rehabilitation in patients with severe chronic obstructive pulmonary disease. Nijmegen: Katholieke Universiteit Nijmegen, 1997. Buchi S, Brandli O, Klingler K, Klaghofer R, Buddeberg C. [Inpatient rehabilitation in inpatients with chronic obstructive lung diseases (COPD): effect on physical capacity for work, psychological wellbeing and quality of life]. Schweiz Med Wochenschr 2000; 130:135-142. Stewart DG, Drake DF, Robertson C, Marwitz JH, Kreutzer JS, Cifu DX. Benefits of an inpatient pulmonary rehabilitation program: a prospective analysis. Arch Phys Med Rehabil 2001; 82:347352. Goldstein RS. personal communication. 2001. Cox NJM, Hendriks JC, Binkhorst RA, van Herwaarden CLA. A pulmonary rehabilitation program for patients with asthma and mild chronic obstructive pulmonary diseases (COPD). Lung 1993; 171:235-244. Sterne JA, Davey SG. Sifting the evidence-what's wrong with significance tests? BMJ 2001; 322:226231. Acock AC. Working with missing values. Family Science Review 1997; 10:76-102. Foglio K, Bianchi L, Ambrosino N. Is it really useful to repeat outpatient pulmonary rehabilitation programs in patients with chronic airway obstruction? A 2-year controlled study. Chest 2001; 119:1696-1704.