Investigation of quality of life mood pain disability and disease status in primary systemic vasculitis.код для вставкиСкачать
Arthritis & Rheumatism (Arthritis Care & Research) Vol. 49, No. 6, December 15, 2003, pp 826 – 837 DOI 10.1002/art.11471 © 2003, American College of Rheumatology ORIGINAL ARTICLE Investigation of Quality of Life, Mood, Pain, Disability, and Disease Status in Primary Systemic Vasculitis MARIA KOUTANTJI,1 EMMA HARROLD,1 SUZANNE E. LANE,2 SHIRLEY PEARCE,1 RICHARD A. WATTS,2 AND DAVID G. I. SCOTT2 Objective. To assess quality of life (QOL) and psychological adjustment in primary systemic vasculitis (PSV), and to assess their relationship to disease-related measures. Methods. Fifty-one PSV patients completed questionnaires assessing QOL (Short Form 36 [SF-36]), disability (Health Assessment Questionnaire [HAQ]), and mood (Hospital Anxiety and Depression Scale [HADS]). Illness-related measures, disease activity, and permanent damage scores (the Birmingham Vasculitis Damage Index [BVDI] modiﬁed, and the Birmingham Vasculitis Activity Scale modiﬁed) were collected. Aspects of the HAQ’s psychometric properties were evaluated. Results. PSV patients’ SF-36 scores, except for mental health, were signiﬁcantly lower than the norms, indicating poorer QOL. Using the HADS, 43.2% of patients reported increased anxiety symptoms and 25.5% increased depressive symptoms. Patients with increased pain when compared with those with little or no pain had signiﬁcantly impaired scores in all SF-36 subscales, except for mental health; and they scored signiﬁcantly worse on depression, fatigue, problems with sleep, and symptom severity. Patients with neuropathic symptoms and those taking high levels of steroids had signiﬁcantly impaired scores on some of the relevant measures. There were no signiﬁcant correlations between the modiﬁed BVDI scores with the SF-36 subscales or with the other self-report disease-related measures. The HAQ showed high internal consistency and high concurrent and discriminant validity. Conclusion. Many aspects of QOL are signiﬁcantly impaired in PSV. Self-reported pain and disease symptoms, SF-36 scores, and depression and anxiety levels are signiﬁcant indicators of the adverse impact of vasculitis on patients’ lives that need to be assessed and managed. KEY WORDS. Quality of life; Vasculitis; Pain; Mood. INTRODUCTION The primary systemic vasculitides (PSVs) are a group of rare, life-threatening, chronic illnesses (1). They are characterized by inﬂammation of blood vessels, which can Supported by the Arthritis Research Campaign, United Kingdom. 1 Maria Koutantji, PhD, Emma Harrold, PhD, Shirley Pearce, PhD: University of East Anglia, Norwich, United Kingdom; 2Suzanne E. Lane, MB ChB, MRCP, Richard A. Watts, BA, BM BCH, MA, DM, FRCP, David G. I. Scott, MB ChB, MD, FRCP: Norfolk & Norwich University Hospital NHS Trust, Norwich, United Kingdom. Address correspondence to Maria Koutantji, PhD, Imperial College London, Department of Surgical Oncology and Technology, 10th Floor QEQM Building, St. Mary’s Hospital, Praed Street, London W2 1NY, UK. E-mail: M.Koutantji@ Imperial.ac.uk. Submitted for publication October 3, 2001; accepted in revised form November 7, 2002. 826 involve and cause damage to a many organ systems, including lungs, kidneys, and skin. This inﬂammation is accompanied by severe and painful symptoms (2–5). The medical treatment of the disease usually comprises highdose steroids and cyclosphosphamide for acute disease episodes, followed by prolonged immunosuppression (e.g., azathiopine, methotrexate), which in itself causes additional symptoms. It is likely that such a serious illness signiﬁcantly impairs a patient’s quality of life (QOL), causes psychological distress, and forces changes in the patient’s and family’s way of life (6). However, the introduction of patient-reported QOL measures to vasculitis research has only just started (7) because it is only recently that progress in medical management has turned these previously fatal illnesses to chronic ones. As a result, there is very little published research assessing the impact of the disease on QOL and psychological adjustment cross-sectionally or longitudinally. Comparative studies with other systemic Quality of Life in Vasculitides diseases are scarce and limited in scope (8,9), as are comparisons between the speciﬁc vasculitides. However, there are now studies in progress evaluating QOL and, to some degree, psychological adjustment in these conditions. Existing research in vasculitis has focused on evaluating QOL using mainly generic instruments, such as the Short Form 36 (SF-36). Hoffman et al’s (10) study is to our knowledge the only relevant study published as a full article where a disease-speciﬁc questionnaire was devised with input from a patient focus group. The questionnaire assessed the effects of Wegener’s granulomatosis (WG) on the health, function, and income of 60 patients. The ﬁndings showed that patients experienced substantial medical and functional morbidity, with 78% requiring long-term immunosuppressive treatment. Daily activities were signiﬁcantly reduced or constrained in 80% of these patients. There was a negative impact of the disease on patients’ normal daily living, employment circumstances, and income and a variable impact on family and close relationships. Boomsma et al (11) translated this questionnaire to Dutch and assessed the impact of WG on 79 Dutch patients. They replicated Hoffman et al’s (10) main ﬁndings: WG was associated with increased medical morbidity and adverse effects on income and physical functioning. However, they reported lower impact of the disease on interpersonal relationships and attributed this ﬁnding to cultural and socioeconomic differences between The Netherlands and the US. A recent review on the socioeconomic impact of vasculitis suggests that its adverse negative effect is greater than anticipated and recommends a full investigation (7). The information from the following studies has been extracted from work presented in conferences and published as abstracts, which allows limited scope for critical appraisal of design and methods. Nevertheless, there are consistent patterns in the reported ﬁndings, which are relevant to the background and aims of this study. Exley et al (12) used the SF-36 to assess QOL in patients (n ⫽ 30) with active and inactive vasculitis and found impaired levels of QOL. Herlyn et al (13) also used the SF-36 to compare a group of patients with active PSV (n ⫽ 303) with a group of patients in complete remission (n ⫽ 40) and a healthy population (n ⫽ 350), assessing patients at baseline and at 12 months followup. They found limited QOL for both patient groups at both assessments. Raza et al (14) investigated the relationship between vasculitis activity and damage by combining scores in the separate SF-36 subscales into scores for physical and mental components in 83 patients with systemic necrotizing vasculitis. Their patient group reported signiﬁcantly lower SF-36 scores than a healthy population. Interestingly, there were no signiﬁcant correlations between disease activity or damage scores and physical or mental components of the SF-36. Grove et al (15) investigated the relationship between SF-36 physical and mental functioning scores and disease activity and damage scores in 91 PSV patients. They reported that overall physical functioning was signiﬁcantly impaired (no comparisons with norms or reference to cutoff points were reported in the abstract). Mental functioning was signiﬁcantly impaired for patients with active 827 disease when compared with patients with inactive disease, and physical functioning only was signiﬁcantly lower for patients with high damage. Currently there is no vasculitis-speciﬁc measure of physical disability. The Health Assessment Questionnaire (HAQ) has been used as a disability measure in 2 published studies (8,9) that included vasculitis patients in their assessment of physical functioning in patients with different rheumatic diseases. In this study, we addressed generic and speciﬁc aspects of QOL and psychological adjustment in a group of UK vasculitis patients. Detailed evidence from a UK sample allows assessment of 1) impact of vasculitis on QOL in a UK context and 2) generalizability of previous ﬁndings. It also enables the identiﬁcation of key areas of disease impact, as indicated by the patients themselves, that could beneﬁt from future tailored medical or psychoeducational interventions. We conducted an exploratory survey of vasculitis patients from a prospective regional UK register aiming speciﬁcally to 1) evaluate the impact of the disease on PSV patients’ QOL using the SF-36, assessing also neuropathic and steroid treatment status, disability levels, pain, fatigue, and mood using the Hospital Anxiety and Depression Scale (HADS) (16); 2) compare levels of QOL between the different PSV diagnostic groups; 3) assess internal consistency and concurrent and discriminant validity of the HAQ as a measure of disability in PSV; and 4) evaluate the relationship between the clinical marker of permanent disease damage (modiﬁed Birmingham Vasculitis Activity Score [BVDI]) with patients’ self-reported scores on the above measures. PATIENTS AND METHODS Patients. Sixty-two patients diagnosed with PSV between 1988 and 1998 and resident within the county of Norfolk, UK were identiﬁed from a prospective vasculitis register. Patients were classiﬁed according to the following: WG by the 1990 American College of Rheumatology (ACR) criteria (17); microscopic polyangiitis (MPA) by the Chapel Hill Consensus Conference criteria (18); and Churg-Strauss syndrome (CSS) by the 1990 ACR criteria (19). Inclusion criteria were that the participants should have a good understanding and use of the English language. Exclusion criteria were patients unwilling or unable to comply with the requirements of the protocol and presence or history of severe psychiatric illness (e.g., psychosis). Patients were contacted by the department of rheumatology and data for the study were collected via a postal survey using self-report questionnaires. Questionnaires were returned by 51 patients, achieving an 82% response rate (WG: n ⫽ 31; MPA: n ⫽ 9; CSS: n ⫽ 11). Thirty patients (59%) were male, 21 (41%) were female. Fortytwo (82%) were married or cohabiting and 9 (18%) were single, widowed, or divorced. The characteristics of the sample are presented in Table 1. To evaluate the HAQ’s psychometric properties (i.e., discriminant validity), the 51 PSV patients were matched Primary systemic vasculitis Age, years Illness duration, years Rheumatoid arthritis n Mean ⴞ SD n Mean ⴞ SD 51 51 63.06 ⫾ 10.13 4.19 ⫾ 2.89 51 51 63.12 ⫾ 10.01 3.63 ⫾ 2.23 Pain 60.3 67.78 ⫾ 27.97 76.9 ⫾ 24.0 ⫺2.55‡ 44.8 40.10 ⫾ 21.37 61.8 ⫾ 21.2 ⫺6.98† 64.5 71.36 ⫾ 17.28 76.4 ⫾ 18.4 ⫺1.87† 70.9 64.89 ⫾ 28.87 86.2 ⫾ 22.7 ⫺6.23† 64.9 56.67 ⫾ 44.80 84.8 ⫾ 30.6 ⫺5.92† 40.5 30.0 ⫾ 39.12 75.9 ⫾ 37.5 ⫺8.32† 62.7 49.1 ⫾ 28.35 76.2 ⫾ 22.3 ⫺8.08† * Scores are the mean ⫾ SD. t-test compares this study sample and the norm sample. † P ⬍ 0.001. ‡ P ⬍ 0.05. Table 2. Age and duration of illness for vasculitis patients and the matched rheumatoid arthritis patients 303 51 525 Clinical measures. Data on disease-related variables had previously been collected for all participants by interview and hospital case note review by a rheumatologist (SEL) using modiﬁed versions of the Birmingham Vasculitis Activity Score (BVAS) (20) and the BVDI (21). Patients were classiﬁed by a clinician (RAW) who was independent from those caring for the patients. The BVAS (20) is a clinician-derived measure of disease activity in 9 organs based on disease signs and symptoms. The BVDI (21) is a clinician-derived measure of permanent damage and irreversible damage to organs, a result of the disease itself or from treatment. These scores were intended for prospective use in the clinic in the assessment of whether symptoms are due to active vasculitis, chronic grumbling disease, permanent damage, or comorbidity, e.g., infection. In this sample, the BVAS and BVDI proforma were used to record clinical ﬁndings, histology, and radiographic ﬁndings throughout the disease course that could be attributed to vasculitis. Severity of disease activity at time of patients’ ﬁrst presentation to the clinic was recorded as the number of organ systems affected by active disease at presentation using the modiﬁed BVAS. Disease damage for near the time of ques- Herlyn et al (ref 13) This study Norm 60–64 years t- test for sex, age, and disease duration to patients with rheumatoid arthritis (RA) from a database of 199 RA patients who were participating at the same time in another trial. RA patients were selected of the same sex and within 5 years either side of the age of the PSV patient to be matched. The RA patient with the closest disease duration was then selected from this subset. Thirty-seven (72.5%) of the resulting RA sample were married or cohabiting and 14 (27.5%) were single, widowed, or divorced. The characteristics of the 2 samples are presented in Table 2. Energy/vitality 2.52 ⫾ 1.80 2 Mental health 46 Social functioning 63.06 ⫾ 10.13 4.19 ⫾ 2.89 19.27 ⫾ 7.91 Role emotional 66 3.42 19 Role physical 51 51 51 Physical functioning Mean score ⴞ SD n Median score Source Age, years Illness duration, years Modiﬁed Birmingham Vasculitis Activity Score at presentation Modiﬁed Birmingham Vasculitis Damage Index near study recruitment n Table 3. Short Form 36 subscale scores obtained in this study, a tertiary referral center, and norms taken from the Oxford Healthy Life Survey* Table 1. Descriptive statistics (median, mean, standard deviation) for age, duration of illness, and clinical markers’ scores for the vasculitis patients in this study 46.6 46.27 ⫾ 22.89 68.1 ⫾ 21.9 ⫺6.78† Koutantji et al General health perception 828 Quality of Life in Vasculitides 829 Figure 1. Short Form-36 (SF-36) subscale scored obtained in this study, a tertiary referral center (reference 13), and norms taken from the Oxford Healthy Life Survey, 1991–1992. tionnaire completion was measured as the number of organ systems damaged by disease using the modiﬁed BVDI. Patient self-report measures. The measures of psychological state, QOL, physical functioning, fatigue, and symptoms were collected from self-completed questionnaires. The HADS (16) was used to assess depression and anxiety levels and has been developed speciﬁcally for use in patients with physical health conditions. The measure has been clinically validated in cross-sectional and longitudinal studies (22). The SF-36 UK version (23), a generic measure of health status, was used to assess certain aspects of patients’ QOL that may be affected by illness. The dimensions comprising the measure are physical function, role limitation due to physical and emotional problems, social functioning, mental health, energy and vitality levels, pain, and perceptions of general health. The measure has been assessed for validity and reliability in its application to the general population and patient groups in the UK and US (23). The HAQ (24) was used to measure levels of physical functioning and disability. This is one of the most frequently used measures of physical functioning in randomized clinical trials of RA (25) and has recently been used to assess physical disability in other rheumatic conditions, including the vasculitides (8,9). Pain, fatigue, symptom severity, and sleep levels were measured using a 101-point visual analog scale (VAS) (26,27). Statistical analyses. Data were analyzed using SPSS software (Chicago, IL). The distributions of the response variables were checked for normality and transformations were performed as appropriate. When the pattern of the results observed using the transformed scores was similar to the results obtained from the raw data analysis, the raw data analysis is presented; if different, we report the results obtained using the transformed scores. Nonparametric tests were used as appropriate where the normality assumptions could not be met. The data were analyzed using Mann-Whitney U tests, t-tests, one-way analysis of variance (one-way ANOVA), and correlations (Pearson r or Spearman rho). Internal consistency of the HAQ was evaluated using Cronbach’s ␣ and correlations between HAQ scores and SF-36 physical disability scores were calculated to assess concurrent validity. Discriminant validity of the HAQ was also assessed comparing scores between the vasculitis sample and a matched sample of RA patients using Wilcoxon’s signed rank test. RESULTS Quality of life. SF-36 scores in this patient cohort appear to be lower than those obtained from patients with vasculitis attending a tertiary referral center from the Herlyn et al study (13), and lower than those in a normal population of similar age (Table 3 and Figure 1). Low SF-36 scores indicate poor QOL. Because there were no standard deviations available for 830 Koutantji et al Table 4. Comparisons between vasculitis diagnostic groups for age, symptom severity, disease status scores, fatigue, sleep, pain, functional disability, and mood Wegener’s granulomatosis (n ⴝ 22–31) Churg-Strauss syndrome (n ⴝ 7–11) Microscopic polyangiitis (n ⴝ 5–9) ANOVA F value Age, years Disease duration, years Modiﬁed BVAS Modiﬁed BVDI 59.97 ⫾ 9.84 4.24 ⫾ 2.94 21.84 ⫾ 8.29 2.14 ⫾ 1.33 67.00 ⫾ 7.27 4.83 ⫾ 2.58 17.09 ⫾ 6.27 3.91 ⫾ 2.39 68.89 ⫾ 10.78 3.25 ⫾ 3.14 13.11 ⫾ 3.06 1.86 ⫾ 1.46 4.25† 0.74 5.68‡ 5.19† Symptom severity (VAS) 15.45 ⫾ 17.01 38.57 ⫾ 25.93 72.00 ⫾ 27.75 16.35§ Fatigue (VAS) Fatigue with age, creatinine, and hemoglobin as covariates Sleep (VAS) Pain this week (VAS) Pain today (VAS) Functional disability (HAQ) HAQ (square root) Anxiety (HADS) Depression (HADS) 35.32 ⫾ 23.31 60.57 ⫾ 29.70 70.00 ⫾ 36.74 4.82† 4.42† 33.59 ⫾ 33.77 24.42 ⫾ 31.32 18.77 ⫾ 28.26 0.53 ⫾ 0.78 0.50 ⫾ 0.53 6.90 ⫾ 3.49 5.94 ⫾ 3.77 46.25 ⫾ 25.60 19.91 ⫾ 20.57 14.63 ⫾ 16.29 1.26 ⫾ 1.12 0.95 ⫾ 0.64 7.55 ⫾ 3.27 6.91 ⫾ 4.23 44.00 ⫾ 40.37 27.50 ⫾ 33.27 27.50 ⫾ 33.27 1.13 ⫾ 1.02 0.91 ⫾ 0.58 6.78 ⫾ 3.49 6.67 ⫾ 3.94 0.53 0.16 0.54 3.81† 3.39† 0.17 0.31 Post hoc Scheffé test locating signiﬁcant differences WG–MPA† — WG–MPA† WG–CSS† CSS–MPA† WG–CSS† WG–MPA§ CSS–MPA† WG–MPA† WG–CSS¶ WG–MPA# — — — WG–CSS** WG–CSS†† — — * Scores are the mean ⫾ SD. ANOVA ⫽ one-way analysis of variance; WG ⫽ Wegener’s granulomatosis; MPA ⫽ microscopic polyangiitis; BVAS ⫽ Birmingham Vasculitis Activity Score; BVDI ⫽ Birmingham Vasculitis Damage Index; CSS ⫽ Churg-Strauss syndrome; VAS ⫽ 101-point visual analog scale; HAQ ⫽ Health Assessment Questionnaire; HADS ⫽ Hospital Anxiety and Depression Scale. † P ⬍ 0.05. ‡ P ⬍ 0.01. § P ⬍ 0.001. ¶ P ⫽ 0.088. # P ⫽ 0.078. ** P ⫽ 0.079. †† P ⫽ 0.09. the Herlyn et al (13) study that would allow comparisons between all 3 groups, statistical comparisons were only conducted between patients who participated in this study and available norms for each SF-36 subscale using t-tests. The results showed that vasculitis patients in this study were signiﬁcantly impaired in all aspects of QOL when compared with the normal population, except for mental health (see Table 3). However, the scores for depression and anxiety obtained with the HADS present a different picture. HADS scores of 8 –10 indicate possible, scores of 11–14 indicate probable, and scores of 15–21 indicate extreme cases of depression and anxiety (16). Of the 51 patients who completed the depression subscale, 11.8% scored as possible, 9.8% scored as probable, and 3.9% scored as extreme cases of depression. Of the 50 patients who completed the anxiety subscale, 27.5% scored as possible, 15.7% scored as probable cases, and nobody scored as an extreme case of anxiety. Overall in this sample, 25.5% of patients presented with depressive symptoms and 43% with anxiety symptoms to a variable degree, indicating that these diseases are associated not only with physical morbidity but also with signiﬁcant psychological distress for a considerable number of patients. In this study, PSV patients’ mean HAQ score was 0.79 (SD 0.94), although the 3 subgroups scored differently, with signiﬁcant differences found between the WG and CSS groups (Table 4). Fries et al (28) reported mean HAQ scores in a sample of 331 community-based RA patients being 0.80. Differences between PSV diagnostic groups on pain and disability levels and aspects of QOL. One-way ANOVA was conducted to assess differences between diagnostic groups on patient-reported levels of pain, functional disability (HAQ), symptom severity, sleep, fatigue (101-point VAS), and QOL (SF-36). Post-hoc Scheffé tests were used to locate pairs of diagnostic groups that were signiﬁcantly different from each other (Table 4). The diagnostic groups differed signiﬁcantly in terms of age (F[2,50] ⫽ 4.25, P ⬍ 0.05): Those with WG were signiﬁcantly younger than those with MPA. The groups did not differ signiﬁcantly in terms of disease duration. There was a signiﬁcant main effect for diagnosis on HAQ score (F[2,48] ⫽ 3.39, P ⬍ 0.05). A difference approaching signiﬁcance (exact P ⫽ 0.077) was located between the groups diagnosed with WG and those with CSS. Patients with a diagnosis of WG indicated lower levels of functional disability than patients with CSS, as shown by lower HAQ scores. There was a signiﬁcant main effect for diagnosis on symptom severity (F[2,33] ⫽ 16.35, P ⬍ 0.001). The 3 diagnostic groups were signiﬁcantly different from each other. MPA patients had the highest symptom severity scores, followed by CSS patients and then WG patients. There was a signiﬁcant main effect for diagnosis on Quality of Life in Vasculitides 831 Table 5. Short Form-36 subscale scores for vasculitis diagnostic groups* Wegener’s granulomatosis (n ⴝ 22–31) Churg-Strauss syndrome (n ⴝ 7–11) Microscopic Polyangitis (n ⴝ 5–9) ANOVA F value Post hoc Scheffé test Physical functioning 56.77 ⫾ 26.57 38.18 ⫾ 27.68 34.38 ⫾ 28.47 3.32† Role physical Role emotional Social function Pain Mental health Energy/vitality General health perception 30.00 ⫾ 40.15 56.67 ⫾ 44.76 67.78 ⫾ 22.67 67.78 ⫾ 25.50 70.53 ⫾ 15.17 42.26 ⫾ 22.80 48.57 ⫾ 24.27 31.82 ⫾ 40.45 45.45 ⫾ 45.39 62.63 ⫾ 37.93 68.69 ⫾ 27.13 75.27 ⫾ 22.19 38.64 ⫾ 16.29 36.50 ⫾ 19.67 27.78 ⫾ 38.41 70.37 ⫾ 45.47 58.02 ⫾ 36.76 66.67 ⫾ 38.89 69.33 ⫾ 18.76 34.44 ⫾ 22.70 49.88 ⫾ 20.15 0.03 0.76 0.43 0.01 0.37 0.49 1.17 WG–CSS‡ WG–MPA§ — — — — — — — SF-36 Subscales * Scores are the mean ⫾ SD. ANOVA ⫽ one-way analysis of variance; WG ⫽ Wegener’s granulomatosis; CSS ⫽ Churg-Strauss syndrome; MPA ⫽ microscopic polyangiitis. † P ⬍ 0.05. ‡ P ⫽ 0.16. § P ⫽ 0.13. fatigue (F[2,33] ⫽ 4.82, P ⬍ 0.05). Age, creatinine, and hemoglobin levels (the last 2 were used as indicators of renal function), were entered in the analysis as covariates to evaluate their possible effects on fatigue. The main effect of diagnosis remained signiﬁcant (F[2,30] ⫽ 4.42, P ⬍ 0.05) and none of the covariates were signiﬁcant. In this analysis, differences were located between WG and both MPA (P ⫽ 0.078) and CSS groups (P ⫽ 0.088), with WG patients reporting lower levels of fatigue than the other 2 groups. The 3 groups did not differ signiﬁcantly in terms of reported QOL on any of the subscales of the SF-36, except for physical functioning (F[2,48] ⫽ 3.32, P ⬍ 0.05). The difference was located between WG and MPA patients, with WG patients reporting higher levels of physical functioning without limitations due to health. There were no signiﬁcant main effects of diagnosis on levels of anxiety and depression (HADS), pain, or impact on sleep in the previous week (using a 101-point VAS) (Table 5). Differences on QOL in PSV according to level of pain, neuropathic symptoms, and steroid treatment. We were particularly interested in evaluating the impact of levels of pain as part of the experience of living with vasculitis on patients’ QOL. Pain can be a result of inﬂammation and damage caused by the disease itself or procedures and treatment-related side effects. We operationally divided the whole sample into 2 groups (little or no pain versus moderate to severe pain) on the basis of the amount of pain experienced over the past week (101-point VAS) using a median split (median ⫽ 10). Patients with pain when compared with those with little or no pain, using Mann-Whitney U tests, had signiﬁcantly impaired scores in all aspects of QOL (SF-36), except for mental health (Tables 6 and 7 and Figure 2). They also scored signiﬁcantly higher on scores of depression (HADS), on symptoms of fatigue, problems with sleep, and perceived symptom severity using 101-point VAS scores (associated P levels ranged from ⬍0.001 to 0.04). The 2 Table 6. Short Form-36 subscale scores for vasculitis patients according to levels of pain experienced* Pain this week (>10 VAS) Little or no pain this week (<10 VAS) Subscale n Mean n Mean Mann-Whitney U score Physical function Role physical Role emotional Social function Pain Mental health Energy/vitality General health perception 22 21 21 21 21 22 22 22 36.14 ⫾ 26.23 15.48 ⫾ 30.08 42.86 ⫾ 44.90 50.79 ⫾ 28.45 42.86 ⫾ 23.12 67.45 ⫾ 18.53 29.09 ⫾ 19.43 35.59 ⫾ 20.92 27 28 28 28 28 27 28 25 60.37 ⫾ 25.90 40.18 ⫾ 42.68 69.05 ⫾ 41.50 74.60 ⫾ 25.28 85.32 ⫾ 13.54 74.81 ⫾ 16.07 48.75 ⫾ 19.37 55.32 ⫾ 21.21 153.5† 195.5‡ 199.5‡ 150.5† 38.5§ 211.0 143.0† 140.5† * Scores are the mean ⫾ SD. Mann-Whitney U values refer to comparisons between groups. VAS ⫽ visual analog scale. † P ⱕ 0.01. ‡ P ⬍ 0.05. § P ⬍ 0.001. 832 Koutantji et al Table 7. Mood, disability, and disease-related symptom scores for the vasculitis patients according to levels of pain experienced* Pain this week (>10 VAS) HADs-depression HADs-anxiety HAQ Symptom severity in the last month (VAS) Problem of sleep (VAS) Problem of fatigue (VAS) Little or no pain this week (<10 VAS) n Mean n Mean Mann-Whitney U score 22 22 21 16 16 16 7.59 ⫾ 3.97 7.91 ⫾ 3.61 1.15 ⫾ 1.08 41.25 ⫾ 28.66 53.69 ⫾ 33.57 57.75 ⫾ 31.25 28 27 28 17 18 17 5.18 ⫾ 3.15 6.33 ⫾ 3.14 0.51 ⫾ 0.73 16.47 ⫾ 24.64 25.55 ⫾ 26.34 33.35 ⫾ 23.99 177† 224 190.5‡ 61† 75.5‡ 75‡ * Scores are the mean ⫾ SD. Mann-Whitney U values refer to comparisons between groups. VAS ⫽ 101-point visual analog scale; HADS ⫽ Hospital Anxiety and Depression Scale; HAQ ⫽ Health Assessment Questionnaire. † P ⱕ 0.01. ‡ P ⬍ 0.05. groups did not differ in age, duration of illness, BVAS score, or BVDI score. The impact of permanent neuropathic symptoms (present at interview versus absent) and steroid treatment (collected from medical notes) on relevant QOL and selfreport measures was also investigated using Mann-Whitney U tests. Patients with neuropathic symptoms had signiﬁcantly impaired scores on symptom severity (101-point VAS), disability (HAQ), and SF-36 physical functioning (P ⬍ 0.05) (Table 8). Steroid treatment (⬍10 mg prednisolone versus ⱖ10 mg prednisolone) when used as a grouping factor showed that patients on higher dose had signiﬁcantly impaired scores for pain this week (101-point VAS), depression (HADS), SF-36 social function, SF-36 energy and vitality, and SF-36 pain (P ⬍ 0.05; Table 9). Assessment of aspects of validity of the HAQ as a disability measure for PSV. Internal consistency, a test of the homogeneity of the scale, was assessed with calculation of Cronbach’s ␣ values for the HAQ overall and for each subscale against the overall score. Cronbach’s ␣ values ranged from 0.91 to 0.93 (Table 10), these are considered to be high (29), indicating high internal consistency of the measure when used with PSV patients. Concurrent validity was evaluated by correlating the HAQ total score with the SF-36 physical functioning subscale score. There was a signiﬁcant negative correlation between the 2 measures ( ⫽ – 0.80, P ⬍ 0.01) which indicate that high HAQ scores are associated with low SF-36 physical functioning scores (Table 11), a strong correlation in the expected direction, which provides Figure 2. Short Form-36 (SF-36) subscale scores for the primary systemic vasculitis patients according to levels of pain experienced (little or no pain group versus moderate to severe pain group). Quality of Life in Vasculitides 833 Table 8. Signiﬁcant differences observed on quality of life measures according to neuropathy status* SF-36 physical functioning HAQ Symptom severity in the last month (101-point VAS) Neuropathy present (n ⴝ 12) No neuropathy (n ⴝ 38) Mann-Whitney U score 31.36 ⫾ 24.81 1.31 ⫾ 1.46 42.86 ⫾ 20.38 54.21 ⫾ 27.89 0.13 ⫾ 0.57 24.81 ⫾ 29.81 112† 122† 46† * Scores are the mean ⫾ SD. Mann-Whitney U values refer to comparisons between groups. SF-36 ⫽ short form 36; HAQ ⫽ Health Assessment Questionnaire; VAS ⫽ visual analog scale. † P ⬍ 0.05. support for concurrent validity of the HAQ used in this population. Discriminant validity was assessed as the degree to which the HAQ was able to distinguish between 2 patient groups (30). Wilcoxon’s signed rank tests were performed on the HAQ subscales to compare disability levels for PSV patients and a matched sample of RA patients. One of the study rheumatologists predicted a priori that RA patients would be more disabled overall. One of the psychologists also predicted that for the 2 component dimensions of the HAQ, the RA patients would be more disabled in ﬁne motor skills (i.e., gripping, dressing and grooming) and that there would be no difference in global disability (i.e., walking and activities) between the 2 patient groups. The matched pairs of patients were compared for pain this week, pain today, and HAQ scores using t-tests. There were signiﬁcant differences between the pairs on each of these scores, these differences were maintained under square root transformations (Table 12). RA patients reported signiﬁcantly higher levels of pain this week and pain today than the PSV patients. The RA patients reported signiﬁcantly higher levels of disability on the HAQ than PSV patients according to the prediction. The 2 groups were also compared using Wilcoxon’s signed rank test on the subscale scores of the HAQ. There were statistically signiﬁcant differences between the 2 patient groups on all subscales except for walking and activities (Table 13). RA patients reported levels of impairment to physical functioning that were signiﬁcantly greater than those reported by the PSV patients. The exceptions were the walking and the activities subscales, where no signif- icant differences were found; RA patients still reported slightly greater, although comparable, levels of impairment on walking and activities to those of the PSV patients; this ﬁnding also supports the relevant prediction. The relationships between the clinical marker of permanent disease damage (modiﬁed BVDI) and self-report measures of QOL. No signiﬁcant correlations were found between the modiﬁed BVDI scores with any of the SF-36 subscales or with the other self-report disease-related measures (Table 14). However, there were signiﬁcant positive correlations between the HAQ total scores and scores on pain, symptom severity, problems of fatigue, and sleep (Table 14), where high HAQ scores were associated with high scores on the above variables. There were also significant negative correlations between scores on pain this week, pain today, symptom severity, fatigue, and sleep with all SF-36 subscales’ scores, except for scores on symptom severity and sleep with SF-36 role reduced due to emotional problems, and symptom severity scores with SF-36 mental health scores (Table 14). High scores on the above variables were associated with low SF-36 scores, which indicate poorer QOL. DISCUSSION In this study we evaluated the impact of PSV on patients’ QOL, psychological adjustment, and physical functioning. The results of the assessment of QOL using the SF-36 showed that this vasculitis group demonstrated impaired Table 9. Signiﬁcant differences observed on quality of life measures according to steroid treatment status* Pain this week (101-point VAS) HADS depression SF-36 social function SF-36 energy/vitality SF-36 pain <10 mg prednisolone (n ⴝ 42) > 10 mg prednisolone (n ⴝ 8) Mann-Whitney U score 18.55 ⫾ 23.18 6⫾3 69.05 ⫾ 27.39 43.26 ⫾ 27.39 71.69 ⫾ 25.17 52.13 ⫾ 41.56 9⫾5 43.06 ⫾ 28.13 43.06 ⫾ 28.13 47.22 ⫾ 34.50 81.5† 96.5† 76† 84† 94† * Scores are the mean ⫾ SD. Mann-Whitney U values refer to comparisons between groups. VAS ⫽ visual analog scale; HADS ⫽ Hospital Anxiety and Depression Scale; SF-36 ⫽ Short Form 36. † P ⬍ 0.05. 834 Koutantji et al Table 10. Cronbach’s ␣ coefﬁcients for HAQ subscales and HAQ overall score as indicators of internal consistency* HAQ subscales Cronbach’s ␣ Dressing and grooming Rising Eating Walking Hygiene Reach Grip Activities HAQ overall score 0.91 0.92 0.93 0.92 0.91 0.91 0.93 0.92 0.93 * HAQ ⫽ Health Assessment Questionnaire. levels on all aspects of QOL, except for mental health, when compared with normative data. In addition, this group had lower QOL scores than a German vasculitis group (13). These results support previous ﬁndings in PSV when the SF-36 was used to assess QOL from work currently published as abstracts (e.g., 12–15); and they support ﬁndings from 2 published studies that used a vasculitis-speciﬁc assessment of QOL (10,11). The SF-36 results from this study showed that patients reported signiﬁcantly impaired levels of physical and social functioning, energy and vitality, and role limitation due to physical and emotional problems; they also reported increased levels of pain and lower perceptions of general health when compared with the norms. Given the severity of the illness itself, the unpredictability of its course, and the signiﬁcant side effects of its treatment, the observed low levels of QOL correspond well with the clinical features of the illness and clinical experience. However, no statistically signiﬁcant differences were observed in this study between the vasculitis group and normative data for the SF-36 mental health subscale. Using the HADS, however, we found that 25.5% of the whole vasculitis sample reported high depressive symptoms and 43.2% reported high anxiety symptoms to a variable degree. Prevalence levels obtained with the HADS in a healthy population (22) were 5% for depression (cut off 8), and 7% for anxiety (cut off 10). In a study investigating mood levels in cancer patients (31) using the HADS, 8.7% of patients scored within the range for possible clinical disorder for depression, and 27% did so for anxiety. In this study, we observed 5.1 times the depressive symptoms and 6.17 times the anxiety symptoms reported in the healthy population (22), and obtained 2.93 times the incidence of depressive symptoms and 1.6 times the incidence of anxiety symptoms when compared with the cancer patients’ mood levels (31). Although the generalizability of these results needs to be assessed, they indicate that psychological distress is high in this disease group, and this is likely to be a major issue for many patients. It is reasonable to suggest that psychological distress should be assessed and addressed as part of the disease management efforts. Our ﬁndings also indicate that there might be a discrepancy in levels of psychological distress obtained by the SF-36 mental health subscale and the HADS. This requires further investigation to determine the most psychometrically sound way to evaluate distress in this population. As part of the aims of this study, we also investigated differences in QOL and psychological adjustment between the 3 different diagnostic subgroups of PSV. Despite the small power for the MPA and CSS groups, statistically signiﬁcant differences were observed for some of the outcome variables, providing some preliminary information on areas of differentiation between the diagnoses that could guide future research. Patients with moderate to severe pain showed signiﬁcantly impaired scores in all aspects of QOL as assessed by the SF-36, except for mental health, although they still scored lower than the little or no pain group in this subscale. Furthermore, patients with pain reported signiﬁcantly higher depression scores and higher anxiety scores (using the HADS), although not signiﬁcantly so. They also scored signiﬁcantly higher on symptom severity over the past month, levels of fatigue, and problems with sleep. These ﬁndings suggest that pain, whatever its cause, is a major determinant of QOL in a similar way in PSV as in most other diseases. It is reasonable to suggest that appropriate and adequate pain management should be a high priority of clinicians treating PSV patients, given its signiﬁcant association with impaired levels of QOL in a range of dimensions. Patients with neuropathic symptoms and patients on higher levels of steroid treatment had signiﬁcantly impaired scores on some, but not as many, of the relevant measures. Presence of neuropathy was associated with impaired scores on SF-36 physical functioning, disability (HAQ), and symptom severity; treatment with higher steroid dose was associated with signiﬁcantly higher levels of pain, depression (HADS), and impaired levels of SF-36 scores for social function and energy and vitality. In addition to the disability subscale of the SF-36, we Table 11. Spearman’s rho correlation coefﬁcients between the Health Assessment Questionnaire (HAQ) scores and Short Form-36 subscale scores HAQ * P ⬍ 0.001. † P ⬍ 0.01. ‡ P ⬍ 0.05. Physical functioning n ⴝ 48 Role physical n ⴝ 48 Role emotional n ⴝ 48 Social functioning n ⴝ 48 Mental health n ⴝ 48 Energy n ⴝ 49 Pain n ⴝ 48 Health perception n ⴝ 46 ⫺0.80* ⫺0.37† ⫺0.18 ⫺0.50* ⫺0.37† ⫺0.36‡ ⫺0.43† ⫺0.49† Quality of Life in Vasculitides 835 Table 12. Pain and functional disability measures for the vasculitis patients and the matched rheumatoid arthritis patients* Primary systemic vasculitis Pain this week (VAS) Pain today (VAS) Pain today (square root transformation) HAQ HAQ (square root transformation) Rheumatoid arthritis n Mean n Mean 49 49 49 49 49 24.40 ⫾ 29.27 19.65 ⫾ 26.91 3.07 ⫾ 3.23 0.79 ⫾ 0.94 0.67 ⫾ 0.59 49 49 49 49 49 45.71 ⫾ 24.81 38.24 ⫾ 26.08 5.71 ⫾ 2.40 1.46 ⫾ 0.97 1.09 ⫾ 0.50 t ⫺3.42† ⫺3.04‡ ⫺4.19† ⫺3.26‡ ⫺3.48† * Scores are the mean ⫾ SD. VAS ⫽ 101-point visual analog scale; HAQ ⫽ Health Assessment Questionnaire. † P ⱕ 0.001. ‡ P ⬍ 0.01. used the HAQ to assess physical functioning and evaluated its psychometric properties when used in PSV. Overall, there was evidence of high internal consistency and concurrent and discriminant validity of the HAQ as a measure of functional disability in PSV. Future research should assess its reliability. The HAQ scores were signiﬁcantly correlated with patient-reported illness symptoms (pain today and pain this week, fatigue, sleep, and symptom severity) and with the SF-36 subscale scores (physical functioning, social functioning, energy and vitality, pain, mental health, and general health perception), suggesting that these were the areas of difﬁculty for patients experiencing adverse effects on their physical functioning. The exception was SF-36 scores for impact on role due to emotional problems, where no signiﬁcant correlation with the HAQ was observed. However, the scores derived from the modiﬁed BVDI clinical measure were weakly and not signiﬁcantly correlated with patient reports of illness symptoms or with QOL scores. Studies of QOL in other chronic rheumatic illnesses have shown that patients’ mood is signiﬁcantly related to the way they perceive their illness and not to markers of disease activity or damage (32,33). It seems that self-report measures provide better information for the experience of living with a systemic illness than disease markers, and simple VAS illness-related measures (e.g., for pain, sleep, fatigue) could easily be incorporated in clinical assessment to capture this. Pincus et al (9) described a modiﬁed version of the HAQ (the Multidimensional Health Assessment Questionnaire [MDHAQ]) and suggested it may be suitable for use with a range of rheumatic diseases in addition to RA. Their ﬁndings using the MDHAQ, and the results of this study, render some support for this. Future research should assess validity and reliability of different methods of assessing disability and psychological distress in PSV. On the basis of the preliminary ﬁndings from this study and previous cross-sectional research that needs to be published in full, it seems that many aspects of QOL are signiﬁcantly impaired in PSV. Additional longitudinal research is essential to establish causal pathways and assess the impact of the disease over time. Future psychological work could 1) investigate how patients and their caretakers think and cope with their illness, and how these affect their QOL; and 2) design, implement, and evaluate psychoeducational interventions that aim to address their psychosocial needs. Although a clinical marker of disease damage serves as an indicator of physical status of vasculitis, it is not necessarily related to measures of QOL or Table 13. Comparison of the vasculitis patients and the matched rheumatoid arthritis patients on subscales of the HAQ* Primary systemic vasculitis Rheumatoid arthritis HAQ subscale n Mean ⴞ SD n Mean ⴞ SD Wilcoxon’s signed rank test (z score) Dressing and grooming Rising Eating Walking Hygiene Reach Grip Activities HAQ overall score 51 51 51 51 49 50 50 50 49 0.84 ⫾ 1.22 0.47 ⫾ 0.83 0.39 ⫾ 0.80 1.02 ⫾ 1.26 0.65 ⫾ 1.19 0.94 ⫾ 1.19 0.84 ⫾ 1.28 1.22 ⫾ 1.33 0.79 ⫾ 0.94 51 51 51 51 49 50 50 50 49 1.53 ⫾ 1.24 1.25 ⫾ 1.63 1.14 ⫾ 1.02 1.18 ⫾ 1.24 1.51 ⫾ 1.30 1.68 ⫾ 1.32 1.67 ⫾ 1.30 1.29 ⫾ 1.17 1.41 ⫾ 0.98 ⫺2.34† ⫺3.41‡ ⫺3.69‡ ⫺0.62 ⫺3.09§ ⫺2.83§ ⫺2.91§ ⫺0.36 ⫺3.16§ * HAQ ⫽ Health Assessment Questionnaire. † P ⬍ 0.05. ‡ P ⱕ 0.001. § P ⬍ 0.01. 836 Koutantji et al Table 14. Spearman rho correlation coefﬁcients between HAQ scores and modiﬁed BVDI scores, and scores of pain, symptom severity, problems of fatigue, and sleep* Modiﬁed BVDI Pain this week (VAS) Pain today (VAS) Symptom severity in the last month (VAS) Problem of fatigue (VAS) Problem of sleep (VAS) SF-36 pain SF-36 mental health SF-36 energy vitality SF-36 general health perception ⫺0.26 ⫺0.46† ⫺0.45† ⫺0.48† ⫺0.25 ⫺0.87§ ⫺0.82§ ⫺0.33¶ ⫺0.14 ⫺0.45† ⫺0.38† ⫺0.28 ⫺0.28 ⫺0.54§ ⫺0.51§ ⫺0.52† ⫺0.25 ⫺0.44† ⫺0.43† ⫺0.57§ ⫺0.39‡ ⫺0.55*† ⫺0.37‡ ⫺0.50† ⫺0.40‡ ⫺0.63§ ⫺0.68§ ⫺0.56§ ⫺0.32¶ ⫺0.63§ HAQ SF-36 physical function SF-36 role physical SF-36 role emotional SF-36 social function 0.15 0.42† 0.47† 0.63§ ⫺0.15 ⫺0.46† ⫺0.46† ⫺0.69§ ⫺0.04 ⫺0.45† ⫺0.41† ⫺0.40‡ ⫺0.14 ⫺0.34‡ ⫺0.29‡ ⫺0.18 0.45† 0.60§ ⫺0.53† ⫺0.55† ⫺0.44† ⫺0.41‡ ⫺0.41‡ ⫺0.26 * HAQ ⫽ Health Assessment Questionnaire; BVDI ⫽ Birmingham Vasculitis Damage Index; SF-36 ⫽ short form-36; VAS ⫽ 101-point visual analog scale. † P ⱕ 0.01. ‡ P ⬍ 0.05. § P ⬍ 0.001. ¶ P ⬍ 0.07. physical and emotional functioning. Self-reported pain and disability, presence of neuropathy, treatment with steroids, depression and anxiety scores, scores on the SF36, and self-report measures of disease symptoms are signiﬁcant indicators of the impact of vasculitis on the patient’s life and need to be assessed and addressed in overall patient management. ACKNOWLEDGMENTS We would like to thank all participants and staff involved in the study for their time, and Dr. R. Luqmani for his helpful comments on a conference presentation of this work. REFERENCES 1. Watts R, Scott D. Classiﬁcation and epidemiology of the vasculitides. Bailliere’s Clin Rheumatol 1997;11:191–218. 2. Chakravarty K. Vasculitis by organ system. Bailliere’s Clin Rheumatol 1997;11:357–93. 3. Valente RM, Conn DL. Polyarteritis: polyarteritis nodosa and microscopic polyangitis. In: Klippel JH, Dieppe PA, editors. Rheumatology. 2nd ed. London: Mosby; 1998. 4. Hoffman GS. Wegener’s granulomatosis. In: Klippel J, Dieppe P, editors. Rheumatology. 2nd ed. London: Mosby; 1998. 5. Guillem LP Churg J. Chrug-Strauss syndrome. In: Klippel J, Dieppe P, editors. Rheumatology. 2nd edition. London: Mosby; 1998. 6. Koutantji M, Pearce S, Harrold E. Psychological aspects of vasculitis. Rheumatology 2000;39:1173–9. 7. Cotch MF. The socioeconomic impact of vasculitis. Curr Opin Rheumatol 2000;12:20 –3. 8. Abu-Shakra M, Smythe H, Lewtas J, Badley E, Weber D, Keystone E. Outcome of polyarteritis nodosa and Churg-Strauss syndrome. Arthritis Rheum 1994;12:1798 –1803. 9. Pincus T, Swearingen C, Wolfe F. Toward a multidimensional health assessment questionnaire (MDHAQ): assessment of advanced activities of daily living and psychological status in the patient-friendly health assessment questionnaire format. Arthritis Rheum 1999;42:2220 –30. 10. Hoffman GS, Drucker Y, Cotch MF, Locker GA, Easley K, Kwoh K. Wegener’s granulomatosis (WG): patient-reported effects of disease on health, function and income. Arthritis Rheum 1998;41:2257– 62. 11. Boomsma MM, Stegeman CA, Cohen Tervaert JW. Comparison of Dutch and US patients’ perceptions of the effects of Wegener’s granulomatosis on health, function, income and interpersonal relationships: comment on the article by Hoffman et al. Arthritis Rheum 1999;42:2495– 6. 12. Exley AR, Moots RM, Carruthers D, Bacon P. Short-form-36 in vasculitis. Clin Exp Immunol 1995;63 Suppl 1:101. 13. Herlyn K, Reinhold-Keller E, Zeidler A, Raspe H, Gutﬂeisch J, Peter HH, et al. Health-related quality of life in primary systemic vasculitides. Arthritis Rheum 1998;41 Suppl 9:S122. 14. Raza K, Wilson A, Carruthers DM, Reay C, Amft N, Bacon PA. Impaired SF36 in patients with primary systemic necrotizing vascultis (1oSNV): inﬂuence on factors besides disease activity and damage. Arthritis Rheum 1999;42 Suppl 9:S315. 15. Grove ML, Wilson A, Raza K, Carruthers DM, Bacon PA. Patients with damage from primary systemic vasculitis experience impaired physical health as measured by the SF36. Rheumatology 2001;40 Suppl:125. 16. Zigmond AS, Snaith RP. The hospital anxiety and depression scale. Acta Psychiatr Scand 1983;67:361–70. 17. Leavitt RY, Fauci AS, Bloch DA, Michel BA, Hunder GG, Arend WP, et al. The American College of Rheumatology 1990 criteria for the classiﬁcation of Wegener’s granulomatosis. Arthritis Rheum 1990;33:1101–7. 18. Jennette J, Falk R, Andrassy K, Bacon PA, Churg J, Gross WL, et al. Nomenclature of systemic vaculitides: proposal of an international consensus conference. Arthritis Rheum 1994; 37:187–92. 19. Masi AT, Hunder GG, Lie JT, Michel BA, Bloch DA, Arend WP, et al. The American College of Rheumatology 1990 criteria for the classiﬁcation of Churg Strauss syndrome (allergic granulomatosis and angiitis). Arthritis Rheum 1990;33:1094 – 100. 20. Luqmani RA, Bacon PA, Moots RJ, Janssen BA, Pall A, Emery P, et al. Birmingham Vasculitis Activity Score (BVAS) in systemic necrotising vasculitis. QJM 1994;87:671– 8. 21. Exley AR, Bacon PA, Luqmani R, Kitas GD, Gordon C, Savage CO, et al. Development and initial validation of the vasculitis damage index (VDI) for the standardized clinical assessment of damage in the systemic vasculitides. Arthritis Rheum 1997; 40:371– 80. 22. Herrmann C. International experiences with the hospital anxiety and depression scale: a review of validation data and clinical results. J Psychosom Res 1997;42:17– 41. 23. Jenkinson C, Wright L, Coulter A. Quality of life measurement in health care: a review of measures and population norms for the UK SF-36. Oxford: Health Services Research Unit; 1996. 24. Fries J, Spitz P, Young D. The dimensions of health outcomes: Quality of Life in Vasculitides 25. 26. 27. 28. 29. the Health Assessment Questionnaire disability and pain scales. J Rheumatol 1982;9:789 –93. Fransen J, Stucki G. Current use of health status instruments in randomised controlled trials on patients with rheumatoid arthritis. Dis Manage Health Outcomes 1998;3:271–7. Jensen MP, Karoly P, Braver S. The measurement of clinical pain intensity: a comparison of six methods. Pain 1986;27: 117–26. Jensen MP, Karoly P. Self-report scales and procedures for assessing pain in adults. In: Turk CD, Melzack R, editors. Handbook of pain assessment. London: The Guildford Press; 1992. p. 135–51. Fries JF, Spitz PW, Kraines RG, Holman HR. Measurement of patient outcome in arthritis. Arthritis Rheum 1980;23:137– 45. Bowling A. Measuring disease. Philadelphia: Open University Press; 1995. 837 30. Horne R, Weinman J, Hankins M. The beliefs about medicines questionnaire: the development and evaluation of a new method for assessing the cognitive representation of medication. Psychol Health 1999;14:1–24. 31. Moorey S, Greer S, Watson M, Gorman C, Rowden L, Tunmore R, et al. The factor structure and factor stability of the Hospital Anxiety and Depression Scale in patients with cancer. Br J Psychiatry 1991;158:255–9. 32. Burckhardt C, Archenholtz B, Mannerkorpi K, Bjelle A. Quality of life of Swedish women with ﬁbromyalgia syndrome, rheumatoid arthritis or systemic lupus erythematosus. J Musculoskel Pain 1993;1:199 –207. 33. Drent M, Wirnsberger R, Breteler M, Kock L, Vries J, Wouters E. Quality of life and depressive symptoms in patients suffering from sarcoidosis. Sarcoidosis Vasc Diffuse Lung Dis 1998; 15:59 – 66.