You are here

Article Text

Article menu
Clinical and epidemiological research
Extended report
The characterisation and determinants of quality of life in ANCA associated vasculitis
  1. Neil Basu1,
  2. Andrew McClean2,
  3. Lorraine Harper2,
  4. Esther Nicole Amft3,
  5. Neeraj Dhaun4,
  6. Raashid A Luqmani5,
  7. Mark A Little6,
  8. David RW Jayne7,
  9. Oliver Flossmann8,
  10. John McLaren9,
  11. Vinod Kumar10,
  12. Lars P Erwig1,
  13. David M Reid1,
  14. Gareth T Jones1,
  15. Gary J Macfarlane1
  1. 1Musculoskeletal Collaboration, School of Medicine and Dentistry, University of Aberdeen, Aberdeen, UK
  2. 2School of Immunity and Infection, University of Birmingham, Birmingham, UK
  3. 3Department of Rheumatology, Western General Hospital, Edinburgh, UK
  4. 4Department of Renal Medicine, Royal Infirmary of Edinburgh, Edinburgh, UK
  5. 5Nuffield Department of Orthopaedics, Rheumatology and Musculoskeletal Sciences, University of Oxford, Oxford, UK
  6. 6School of Medicine, Trinity College Dublin, Dublin, Ireland
  7. 7Department of Clinical Medicine, Addenbrooke's Hospital, Cambridge, UK
  8. 8Renal Unit, Royal Berkshire Hospital, Reading, UK
  9. 9Fife Rheumatic Diseases Unit, Whyteman's Brae Hospital, Kirkcaldy, UK
  10. 10Department of Rheumatology, Ninewells Hospital, Dundee, UK
  1. Correspondence to Dr Neil Basu, Musculoskeletal Collaboration (Epidemiology Group), School of Medicine and Dentistry, University of Aberdeen, Polwarth Building, Foresterhill, Aberdeen AB25 2ZD, UK; neilbasu{at}abdn.ac.uk

Abstract

Objectives To contextualise and identify the determinants of poor health related quality of life (QOL) among patients with antineutrophil cytoplasm antibody (ANCA) associated vasculitis (AAV).

Methods A multicentre AAV case–control study was conducted using two matched groups of population and chronic disease controls. Measures of physical and mental QOL as well as putative bio-psychosocial determinants of QOL impairment were collected. Concurrently, putative clinical QOL determinants were recorded. Conditional logistic regression analyses characterised group differences while multivariable logistic regression identified within-case QOL associations which were further quantified using population attributable risks (PAR).

Results Cases (n=410) experienced similar QOL to chronic disease controls (n=318) (physical QOL: OR 0.7, 95% CI 0.4 to 1.1; mental QOL: OR 1.1, 95% CI 0.8 to 1.6). However, they were substantially more likely to report poor QOL compared to general population controls (n=470) (physical QOL: OR 7.0, 95% CI 4.4 to 11.1; mental QOL: OR 2.5, 95% CI 1.7 to 3.6). A few clinical, but many more bio-psychosocial factors were independently associated with poor QOL. In population terms, fatigue was found to be of principal importance (physical QOL: PAR 24.6%; mental QOL: PAR 47.4%).

Conclusions AAV patients experienced significant QOL impairment compared to the general population, but similar to those with other chronic diseases whose considerable needs are already recognised. Potentially modifiable clinical determinants have been identified; however bio-psychosocial interventions are likely to provide the greater QOL gains in this patient population.

  • Systemic Vasculitis
  • Outcomes Research
  • Granulomatosis with Polyangiitis

https://doi.org/10.1136/annrheumdis-2012-202750

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Introduction

    Enhancing health related quality of life (QOL) is a principal objective of clinical care but few studies have investigated this outcome in antineutrophil cytoplasm antibody (ANCA) associated vasculitis (AAV): a group of chronic complex diseases where patients are typically exposed to long term toxic therapies, burdened by accumulated organ damage and at risk of life threatening relapses.

    Impaired QOL is of universal relevance and not just confined to those with disease. Although existing data demonstrate poor QOL among patients with AAV,1 it is essential to contextualise these reports with both general and other diseased populations in order to fully determine the scale of the problem. Thus far, studies have only made retrospective comparisons with unmatched general populations.2 ,3

    Moreover, the determinants of impaired QOL among AAV patients have not been comprehensively assessed. A few studies have described associations between clinical factors such as disease activity4 and bio-psychosocial factors such as fatigue,2 however no study has been sufficiently powered to assess the influence of both clinical and bio-psychosocial factors in order to identify independent determinants of impaired QOL: a key step to informing future interventions directed at improving patient-centred outcomes.

    This study aimed to quantify the QOL of clinic attending AAV patients compared to both the general population and other hospital clinic attendees, and identify independent determinants of poor QOL among AAV patients with a view to informing future interventions.

    Methods

    A multicentre case–control study was conducted. For each case, both population and disease controls were invited to participate. All participants completed a questionnaire comprising a measure of QOL and a number of potential bio-psychosocial determinants of QOL. Concurrently, information on putative clinical determinants of QOL was collected from cases.

    Subjects

    Cases were adult subjects (>16 years) with AAV who fulfilled the European Medicines Agency classification algorithm for either (granulomatosis polyangiitis (GPA), Wegener's), microscopic polyangiitis (MPA) or eosinophilic granulomatosis with polyangiitis (Churg-Strauss, EGPA).5 They were invited from hospital based rheumatology and renal departments across the UK according to consecutive clinic attendance.

    Population controls were identified from a commercial online sampling frame (http://www.192.com), a representative source shown to have >80% population coverage.6 For each case, the online search engine listed the contact details of potential controls residing within the same or similar postcode. The resulting list was consecutively searched and matching persons, in terms of gender and age (±5 years) were selected. Initially, the first five identified controls were invited. If, after 6 weeks, a case remained unmatched to a general population (online) control, further potential controls (up to a maximum of five) were invited.

    Disease controls were matched according to gender, age (±5 years) and recruiting investigator. Cases recruited by rheumatologists and nephrologists were matched to consecutive clinic attendees with inflammatory arthritis and non-inflammatory chronic kidney disease (CKD), respectively, reflecting a typical patient from the corresponding speciality. Up to two disease controls were invited per case.

    Questionnaire

    AAV cases and disease controls were recruited during clinic attendance where they received the questionnaire for completion at home while population controls received the questionnaire by mail.

    The questionnaire comprised several components:

    • Measure of QOL: The Short Form 36 (SF36) has been the most widely used measure of QOL in AAV, displaying good validity,4 reliability7 and responsiveness.4 The measure scores eight domains which are most widely affected by disease and treatment on a 0–100 scale, where 100 represents a perfect QOL. The domains can be further summarised by two distinct, similarly scaled, scores: mental component summary (MCS) and physical component summary (PCS).

    For the purposes of analysis, cut-offs were defined according to the respective summary score general population control means.

    • Measures of putative bio-psychosocial determinants of QOL in AAV:

      • Socio-demographic factors: Closed questions collected information on gender, smoking history and employment status

      • Depression and anxiety: The hospital anxiety and depression scale (HADS) is a brief instrument which was initially developed to identify cases of depression and anxiety. Its 14 items are scored in a Likert response style resulting in scores ranging between 0 and 21 for each domain. In this current study, the commonly cut-off of 8 for ‘caseness’ was adopted.8

      • Sleep quality: The estimation of sleep problems questionnaire concisely quantifies the most common symptoms of sleep dysfunction.9 ,10 Higher scores (range 0–20) indicate greater sleep disturbance.

      • Pain: Participants were asked whether they had experienced pain lasting for 1 day or longer over the past month (yes/no response).

      • Fatigue: The ‘Chalder Fatigue Scale (CFS)’11is a brief but comprehensive tool which was designed to elucidate 11 core symptoms integral to generic fatigue. Higher scores (range 0–11) indicate higher levels of fatigue.

      • Coping: The ‘brief cope’ assesses 14 dimensions of generalised coping relevant to chronic diseases.12 They comprise dysfunctional coping: denial, self-distraction, self-blame, substance use, venting and behavioural disengagement; emotion-focused coping: acceptance, use of emotional support, humour, positive reframing and religion; and problem-focused coping: instrumental support, planning and active coping.13 Ratings are based on a four point ordinal scale ranging from 1 ‘I don't do this at all’ to 4 ‘I have been doing this a lot’.

    With the exception of HADS, where recognised cut-offs exist, all multi-question scores were dichotomised according to the mean score observed within the general population control sample for the purposes of analysis.

    Clinical assessment

    Information regarding potential clinical determinants of QOL was collected at recruitment. This included measures of disease activity (Birmingham Vasculitis Activity Score 3, BVAS),14 where active disease was defined as BVAS>0, and damage (Vasculitis Damage Index (VDI))15 categorised as absent (VDI=0), mild/moderate (VDI=1–4) and severe VDI>4. In addition, the Charlson index (CI)16 was employed to quantify pre-diagnosis co-morbidity, where for analysis the scale was dichotomised into no previous co-morbidity=0 and previous co-morbidity >0. Diagnostic status, history of system involvement, body mass index, past and present immunological status (indirect immunofluorescent ANCA, ANCA antigens), estimated glomerular filtration rate (eGFR17), haemoglobin, lymphocytes, albumin and C-reactive protein (CRP) were recorded and analysed according to recognised clinically relevant categories and disease duration and immunosuppressant exposure according to tertiles (where data not heavily skewed).

    Statistical analysis

    Characteristics of the recruited populations were expressed using simple descriptive statistics. SF36 population differences were determined using Mann–Whitney tests with differences between cases and controls further quantified using conditional logistic regression. Individual associations between putative determinants and the SF36 summary scores within cases were examined using χ2 tests and then quantified using logistic regression. Independent associations for each of the SF36 summary scores were then assessed using forward stepwise logistic regression where all biologically plausible determinants with moderate to strong individual associations (defined as p<0.2) and with a population prevalence >10% (in order to target only those factors with greatest potential population impact) were offered to the model. Factors were included in the final models at p≤0.10 and excluded at p≥0.15. All effect sizes were expressed as ORs and 95% CIs.

    To reduce the impact of cumulative missing data on effect size precision, standard logistic regression models were then developed for each summary score considering only those variables identified to be important by the forward stepwise technique.

    In order to estimate the overall impact of each retained variable, population attributable risks (PAR) were calculated.18 All analyses were conducted using STATA V.11.2.

    Ethics approval

    Ethics approval for the study was obtained from the North of Scotland Research Ethics Committee (ref: 09/S0801/83) and from the University of Aberdeen (ref: CERB/2010/1/493). Written informed consent was obtained from participants.

    Results

    Study population

    A total of 486 cases were approached from 11 clinics, of which 410 (84.4%) completed and returned the questionnaire.

    The clinical characteristics of the cases (table 1) indicated a typical middle aged to elderly patient group (mean 61.4 (SD 14.7) years) of almost equal gender distribution (49.0% male). The majority were classified as GPA (64.6%) followed by MPA (23.2%) and then EGPA (10.7%). Overall, cases were established (median 5.1, IQR 2.2–10.0 years’ disease duration) with largely stable disease: only 19.3% had any evidence of disease activity.

    View this table:
    Table 1

    Case clinical characteristics*

    Of the 2729 general population controls invited, 470 responded (17.2%): mean age 61.5 (SD 13.6) years, 47.5% male. We approached 564 disease controls, of which 318 (56.3%) returned a completed questionnaire (59.4% inflammatory arthritis, 36.8 non-dialysis CKD, 3.8% dialysis CKD; see online supplementary table S1): mean age 62.2 (SD 13.6) years, 48.1% male.

    Contextualising QOL (SF36)

    Across almost all domains, general population controls reported better QOL than the AAV cases (figure 1). The average PCS and MCS scores reported by cases were 16 and 5.7 points less, respectively (Mann–Whitney test, both comparisons p<0.00001). In comparison to population controls, the odds of cases reporting below average physical (OR 7.0, 4.4–11.1) and mental (OR 2.5, 1.7–3.6) related QOL was significantly increased.

    Figure 1
    Figure 1

    Median SF36 scores. PF, Physical Function; RP, Role Physical; BP, Bodily Pain; GH, General Health; VT, Vitality; SF, Social Functioning; RE, Role Emotional; MH, Mental health.

    In contrast there was no excess odds of cases reporting below average QOL compared to disease controls PCS (OR 0.7, 0.4–1.1) or MCS (OR 1.1, 0.8–1.6) (figure 1).

    Individual associations of QOL

    In univariable analyses, SF36 PCS scores were associated with a number of clinical and bio-psychosocial factors (see online supplementary tables S2 and S3). Clinical factors were especially related to PCS: BVAS>0 (OR 3.6, 1.5–8.6), raised CRP (OR, 1.7–7.2), anaemia (OR 2.5, 1.3–5.0) and nervous system involvement (OR 2.5, 1.3–4.8) were all strongly associated (p<0.01) with poor PCS while a high cumulative exposure to azathioprine was protective against poor PCS (OR 0.3, 0.2–0.6). However, other clinical measures did not significantly relate to PCS, including diagnostic label, ANCA status, VDI, abnormal eGFR, and ear, nose and throat involvement. The only clinical factors significantly associated with MCS were hypoalbuminaemia (OR 2.1, 1.2–3.7), raised CRP (OR 1.8, 1.1–2.9) and exposure to mycophenolate (OR 1.6,1.1–2.6).

    Most bio-psychosocial factors were strongly associated with MCS and/or PCS. These included unemployment (MCS: OR 6.5, 2.9–14.8; PCS: OR 8.9, 2.1–37.4), anxiety (MCS: OR 12.9, 6.2–26.7), depression (MCS: OR 16.1, 6.8–38.1; PCS: OR 6.1, 2.4–15.7), pain (MCS: OR 2.0, 1.3–3.0; PCS: OR 4.0, 2.4–6.7), sleep disturbance (MCS: OR 3.8, 2.4–5.8; PCS: OR 4.9, 2.8–8.4) and fatigue (MCS: OR 6.5, 3.7–11.4; PCS: OR 10.1, 5.6–18.2) as well as dysfunctional coping strategies such as denial (MCS: OR 3.3, 1.8–6.0; PCS: OR 2.6, 1.2–5.7) and behavioural disengagement (MCS: OR 8.3, 4.7–14.7; PCS: OR 2.0, 1.1–3.6).

    Independent associations of QOL

    PCS multivariable model

    The forward stepwise technique retained eight variables: older age, current prednisolone dose above 5 mg, raised CRP, high fatigue, sleep disturbance, a history of nervous system involvement and a tendency towards denial as a coping style (table 2). On evaluation of PARs, fatigue was the strongest associate of PCS within the final model, with almost double the influence of any other considered variable. Sleep disturbance and pain ranked second and third, respectively (table 2).

    View this table:
    Table 2

    Multivariable explanatory models of poor quality of life (QOL) among ANCA associated vasculitis patients

    MCS multivariable model

    The stepwise model retained 10 variables in the final model. These included depression, fatigue, anxiety and hypoalbuminaemia in addition to the coping styles of self blame, substance use, self distraction and low positive reframing (table 2). The strongest association was again fatigue: high fatigue was associated with a PAR of 47.4%, while the next nearest ranking variable, self distraction, recorded a PAR of 19.3% (table 2). With the exception of hypoalbuminaemia, the model was notable for the absence of clinical variables.

    Discussion

    This study, the largest investigation of QOL among patients with AAV, is the first to contextualise QOL: patients reported significantly worse QOL compared to the general population but similar levels to that reported by patients with chronic conditions whose substantial needs are already recognised. Furthermore, this study is the first to adopt a comprehensive approach to the assessment of possible determinants of QOL impairment and, by doing so, has identified a number of independent clinical and bio-psychosocial associations which may be amenable to modification. Of particular note, fatigue was the most important contributor to poor QOL.

    The QOL results, using SF36, are consistent with a previous cross-sectional AAV study from the UK region of Norfolk.1 In contrast, discrepancies are observed with data from other countries.3 ,4 Such distinctions probably relate to differing population characteristics such as age, disease damage and activity, and emphasise the importance of accurately contextualising scores.

    Walsh and colleagues were the first to examine the independent associations of QOL in AAV.19 Although their analysis was restricted to newly diagnosed patients and assessed only clinical factors, they also identified old age and neurological involvement as potentially important determinants of poor physical related QOL. Neither this nor our study has delineated the precise type of neurologic involvement, however it is known that peripheral neuropathies constitute the commonest AAV neurological manifestation and that these result in pain, sensory disturbance and functional disability.1 ,20 Its negative impact on physical related QOL is therefore certainly plausible and so more aggressive treatment of this system may improve this outcome.

    Raised CRP and high current prednisolone dose represent further potentially modifiable clinical variables which are independently associated with poor PCS. To some extent, they are both likely to reflect poor disease control and, as per standard clinical guidance, the treating physician should target disease remission. Certainly BVAS was significantly associated with poor PCS on univariable analysis in keeping with Tommasson and colleagues’ observations from two US cohorts.21 However, BVAS was absent from our study's final model, suggesting that these factors are likely to make contributions to the reporting of low PCS over and above disease activity alone. CRP is a non-specific measure of inflammation and other triggers, especially infection, should be considered and addressed before assuming underlying disease activity; high dose glucocorticoids are not only a consequence of disease activity but are directly associated with co-morbidities such as depression. Ultimately, clinical factors such as BVAS, VDI, eGFR and anaemia may have failed to have entered the final model due to the relative stability of the cohort, for example only 7% were classified as having end stage renal disease. Therefore, despite its large sample size, this study was insufficiently powered to detect the impact of severe kidney disease on QOL.

    In keeping with observations elsewhere, the greatest contributors to poor physical and mental related QOL were bio-psychosocial rather than clinical factors.1 ,2 Koutanji and colleagues recorded fatigue, sleep problems and pain visual analogue scale scores to be universally and significantly correlated with the sub-domains of SF36.1 It is understandable why living with a chronic disease, such as AAV, may result in such symptoms: the relapsing nature of a life-threatening disease leads to uncertainty regarding the future and associated fear.22 Such factors are best addressed via a non-pharmacological multidisciplinary approach. For example, sleep hygiene promotion is preferred to prescription of hypnotics in most cases of sleep disturbance,23 similarly behaviour interventions for anxiety and depression are now preferred to psychotropic treatments which are reserved for more severe problems.24 ,25

    This study has identified a number of dysfunctional coping styles in AAV which have previously been noted in related conditions. In systemic lupus erythematosus patients, for example, behavioural disengagement has negatively impacted on both SF36 summary scores and venting has been associated with low MCS.26 Again, non-pharmacological interventions incorporating coping skills training have been effective in improving QOL outcomes in other chronic rheumatic conditions27 ,28 and should be tested for patients with AAV.

    Fatigue was the only factor to be independently associated with both PCS and MCS and was also, by far, the greatest contributor to poor QOL in population terms across both models. This is in keeping with our previous work in the UK2 and international studies29 which have identified fatigue as a principal burden in the lives of these patients. Unfortunately its underlying mechanisms are poorly understood and therapeutic options are currently limited.

    There are a number of methodological issues to consider in the interpretation of these findings. First, cross-sectional analyses are unable to determine causality since they typically do not disassociate exposure and outcome in time. Thus, although the within-case analysis found many variables to be strongly associated with impaired QOL (all considered to be potential determinants), the true direction of causality was difficult to infer. Second, more than a quarter of the cases were not analysed within the final forward stepwise models because of missing data. Assessment of individual variables revealed average missing data rates compared to other studies (0–11% for questionnaire measures and 0–14% for clinical measures).30 Thus the large volume of absent data observed in the final multivariable models suggests a cumulative loss rather than certain patients (with certain characteristics) systematically providing incomplete data for numerous variables and so introducing response bias. Third, despite fatigue's large contribution to the final models, it is recognised that this symptom in itself could subsume a number of clinical factors which may otherwise have been independently associated.

    Finally, the poor population control group response rates are recognised. Unfortunately, these are reflective of the internationally observed downward trend in participation rates among epidemiological studies.31 There is the concern that persons with better health and QOL would be more likely to respond, although the impact of health status on non-response within general population surveys is mixed.32 ,33 Furthermore, this study's matched analysis has adjusted for the major demographic factors recognised to be associated with non-response and so significant bias was not anticipated.

    In summary, QOL among AAV patients is poor and determined by multiple clinical and bio-psychosocial factors, of which fatigue appears to be of principal importance. Clinically, optimal control of underlying inflammation and neurological manifestations are likely to improve certain aspects of QOL. However multidisciplinary non-pharmacological interventions targeting bio-psychosocial determinants may offer even greater gains. Therapeutic advancements have transformed AAV outcomes in terms of mortality and morbidity. In line with other chronic diseases, it is now essential to further develop treatment strategies aimed at improving person centred outcomes such as QOL and its key determinant fatigue.

    Acknowledgments

    The authors wish to thank Jennie King, Julia Wilkinson, Jennifer O'Donoghue and Denise Brown for supporting patient recruitment and Nabi Moaven-Hashemi, Jennifer Banister, Marcus Beasley and Leyla Swafe who assisted with the mailing and data entry for the study.

    References

    1. ↵
      1. Koutantji M,
      2. Harrold E,
      3. Lane SE,
      4. et al
      . Investigation of quality of life, mood, pain, disability, and disease status in primary systemic vasculitis. Arthritis Rheum-Arthritis Care Res 2003;49:826–37.
      OpenUrlCrossRefWeb of Science
    2. ↵
      1. Basu N,
      2. Jones GT,
      3. Fluck N,
      4. et al
      . Fatigue: a principal contributor to impaired quality of life in ANCA-associated vasculitis. Rheumatology 2010;49:1383–90.
      OpenUrlAbstract/FREE Full Text
    3. ↵
      1. Faurschou M,
      2. Sigaard L,
      3. Bjorner JB,
      4. et al
      . Impaired health-related quality of life in patients treated for Wegener's granulomatosis. J Rheumatol 2010;37:2081–5.
      OpenUrlAbstract/FREE Full Text
    4. ↵
      1. Tomasson G,
      2. Boers M,
      3. Walsh M,
      4. et al
      . Assessment of health-related quality of life as an outcome measure in granulomatosis with polyangiitis (Wegener's). Arthritis Care Res (Hoboken) 2012;64:273–9.
      OpenUrlCrossRefPubMedWeb of Science
    5. ↵
      1. Watts R,
      2. Lane S,
      3. Hanslik T,
      4. et al
      . Development and validation of a consensus methodology for the classification of the ANCA-associated vasculitides and polyarteritis nodosa for epidemiological studies. Ann Rheum Dis 2007;66:222–7.
      OpenUrlAbstract/FREE Full Text
    6. ↵
      1. Basu N,
      2. Swafe L,
      3. Reid D,
      4. et al
      . The evaluation of a novel general population sampling frame: an online solution for a persisting problem? J Epidemiol Community Health 2011;65:A175–6.
      OpenUrlAbstract/FREE Full Text
    7. ↵
      1. Herlyn K,
      2. Reinhold-Keller E,
      3. Zeidler A,
      4. et al
      . Health-related quality of life in primary systemic vasculitides. Arthritis Rheum 1998;41:S122–2.
      OpenUrlCrossRef
    8. ↵
      1. Snaith RP
      . The hospital anxiety and depression scale. Health Qual Life Outcomes 2003;1:29.
      OpenUrlCrossRefPubMed
    9. ↵
      1. Jenkins CD,
      2. Stanton BA,
      3. Niemcryk SJ,
      4. et al
      . A scale for the estimation of sleep problems in clinical research. J Clin Epidemiol 1988;41:313–21.
      OpenUrlCrossRefPubMedWeb of Science
    10. ↵
      1. Deodhar A,
      2. Braun J,
      3. Inman RD,
      4. et al
      . Golimumab reduces sleep disturbance in patients with active ankylosing spondylitis: results from a randomized, placebo-controlled trial. Arthritis Care Res (Hoboken) 2010;62:1266–71.
      OpenUrlCrossRefPubMedWeb of Science
    11. ↵
      1. Chalder T,
      2. Berelowitz G,
      3. Pawlikowska T,
      4. et al
      . Development of a fatigue scale. J Psychosom Res 1993;37:147–53.
      OpenUrlCrossRefPubMedWeb of Science
    12. ↵
      1. Carver CS
      . You want to measure coping but your protocol's too long: consider the brief COPE. Int J Behav Med 1997;4:92–100.
      OpenUrlCrossRefPubMedWeb of Science
    13. ↵
      1. Campos RP,
      2. Vazquez MI
      . Health-related quality of life in women with fibromyalgia: clinical and psychological factors associated. Clin Rheumatol 2012;31:347–55.
      OpenUrlCrossRefPubMed
    14. ↵
      1. Mukhtyar C,
      2. Lee R,
      3. Brown D,
      4. et al
      . Modification and validation of the Birmingham vasculitis activity score (version 3). Ann Rheum Dis 2009;68:1827–32.
      OpenUrlAbstract/FREE Full Text
    15. ↵
      1. Exley A,
      2. Banbury P,
      3. Carruthers D,
      4. et al
      . The Vasculitis Damage Index (Vdi). Arthritis Rheum 1995;38:1124–4.
      OpenUrlWeb of Science
    16. ↵
      1. Charlson M,
      2. Szatrowski TP,
      3. Peterson J,
      4. et al
      . Validation of a combined comorbidity index. J Clin Epidemiol 1994;47:1245–51.
      OpenUrlCrossRefPubMedWeb of Science
    17. ↵
      1. Levey AS,
      2. Bosch JP,
      3. Lewis JB,
      4. et al
      . A more accurate method to estimate glomerular filtration rate from serum creatinine: a new prediction equation. Modification of Diet in Renal Disease Study Group. Ann Intern Med 1999;130:461–70.
      OpenUrlCrossRefPubMedWeb of Science
    18. ↵
      1. Kirkwood BR,
      2. Sterne JAC,
      3. Kirkwood BR
      . Eoms Essential medical statistics. Malden, Mass. Oxford: Blackwell Science, 2003.
    19. ↵
      1. Walsh M,
      2. Mukhtyar C,
      3. Mahr A,
      4. et al
      . Health-related quality of life in patients with newly diagnosed antineutrophil cytoplasmic antibody-associated vasculitis. Arthritis Care Res 2011;63:1055–61.
      OpenUrlCrossRefWeb of Science
    20. ↵
      1. Suppiah R,
      2. Hadden RD,
      3. Batra R,
      4. et al
      . Peripheral neuropathy in ANCA-associated vasculitis: outcomes from the European Vasculitis Study Group trials. Rheumatology (Oxford) 2011;50:2214–22.
      OpenUrlAbstract/FREE Full Text
    21. ↵
      1. Tomasson G,
      2. Boers M,
      3. Walsh M,
      4. et al
      . Assessment of health-related quality of life as an outcome measure in granulomatosis with polyangiitis (Wegener's). Arthritis Care Res 2012;64:273–9.
      OpenUrlCrossRefWeb of Science
    22. ↵
      1. Koutantji M,
      2. Pearce S,
      3. Harrold E
      . Psychological aspects of vasculitis. Rheumatology 2000;39:1173–9.
      OpenUrlFREE Full Text
    23. ↵
      NICE. Guidance on the use of zaleplon, zolpidem and zopiclone for the short term management of insominia. 2004. http://www.nice.org.uk/nicemedia/live/11530/32845/32845.pdf (accessed Dec 2012).
    24. ↵
      NICE. Anxiety (amended). Management of anxiety (panic disorder, with or without agoraphobia, and generalised anxiety disorder) in adults in primary, secondary and community care. 2007. http://www.nice.org.uk/nicemedia/live/13314/52601/52601.pdf (accessed Dec 2012).
    25. ↵
      NICE. Depression: management of depression in primary and secondary care—NICE guidance. 2004.
    26. ↵
      1. Rinaldi S,
      2. Ghisi M,
      3. Iaccarino L,
      4. et al
      . Influence of coping skills on health-related quality of life in patients with systemic lupus erythematosus. Arthritis Rheum 2006;55:427–33.
      OpenUrlCrossRefPubMedWeb of Science
    27. ↵
      1. Neill J,
      2. Belan I,
      3. Ried K
      . Effectiveness of non-pharmacological interventions for fatigue in adults with multiple sclerosis, rheumatoid arthritis, or systemic lupus erythematosus: a systematic review. J Adv Nurs 2006;56:617–35.
      OpenUrlCrossRefPubMedWeb of Science
    28. ↵
      1. Sandstrom MJ,
      2. Keefe FJ
      . Self-management of fibromyalgia: the role of formal coping skills training and physical exercise training programs. Arthritis Care Res 1998;11:432–47.
      OpenUrlCrossRefPubMedWeb of Science
    29. ↵
      1. Herlyn K,
      2. Hellmich B,
      3. Seo P,
      4. et al
      . Patient-reported outcome assessment in vasculitis provides important data and a unique perspective. Arthritis Care Res (Hoboken) 2010;62:163945.
      OpenUrlCrossRefPubMedWeb of Science
    30. ↵
      1. Fayers PM D
      . Quality of life: the assessment, analysis and interpretation of patient-reported outcomes, 2nd edn. Chichester, UK: Wiley, 2007:566.
    31. ↵
      1. Galea S,
      2. Tracy M
      . Participation rates in epidemiologic studies. Ann Epidemiol 2007;17:643–53.
      OpenUrlCrossRefPubMed
    32. ↵
      1. Grotzinger KM,
      2. Stuart BC,
      3. Ahern F
      . Assessment and control of nonresponse bias in a survey of medicine use by the elderly. Med Care 1994;32:989–1003.
      OpenUrlCrossRefPubMedWeb of Science
    33. ↵
      1. Norton MC,
      2. Breitner JC,
      3. Welsh KA,
      4. et al
      . Characteristics of nonresponders in a community survey of the elderly. J Am Geriatr Soc 1994;42:1252–6.
      OpenUrlPubMedWeb of Science
    View Abstract

    Supplementary materials

    • Supplementary Data

      This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

      Files in this Data Supplement:

    Footnotes

    • Handling editor Tore K Kvien

    • Contributors The study was conceived by NB, GJM, GTJ and DMR and all authors contributed to design and data collection. Data analysis, interpretation and manuscript preparation was performed by NB, GJM, GTJ and DMR. All authors have critically reviewed the manuscript and approved its publication.

    • Funding NB and the study was funded by the Chief Scientist's Office (ref:CAF/08/08).

    • Competing interests None.

    • Patient consent Obtained.

    • Ethics approval North of Scotland Research Ethics Committee and the University of Aberdeen.

    • Provenance and peer review Not commissioned; externally peer reviewed.

    • Data sharing statement Access to additional unpublished data may be discussed with the corrresponding author.