Are globals for health, well-being and quality of life interchangeable? A mixed methods study in ankylosing spondylitis patients and controls

Are globals for health, well-being and quality of life interchangeable? A mixed methods study in... Abstract Objective Patients’ experience of overall health is often assessed through a single-item global question. Here, we evaluated among patients with AS and population controls whether single-item questions on the constructs health, well-being and quality of life (QoL) are interchangeable. Methods In a mixed quantitative and qualitative approach, all subjects scored the three single-item globals on a numeric rating scale (0–10, best). Next, they indicated for each of the questions which aspects they had been considering when scoring. After forced reflection, globals were scored again. Dissimilarities in scores among constructs, between patients and controls, and before or after reflection were tested using mixed linear models. Themes identified per construct in the qualitative part were linked to the International Classification of Functioning, Disability and Health. The type of themes per construct was compared between patients and controls. Results Sixty-eight AS patients and 84 controls completed the questionnaire. Patients scored significantly worse on each global than controls (mean 6.1–6.3 vs 7.2–7.6, all P < 0.01). Within groups, however, no significant differences in scores on each construct, or in scores before or after forced reflection were found. Health-related themes were relevant to each construct for patients, but were less relevant for controls when considering well-being and QoL. Emotional functions were relevant to well-being in all participants. Social roles and financial situation were more frequently related to well-being and QoL in controls. Conclusion While patients and controls identified content-related dissimilarities between the three constructs studied, this was not reflected in different scores of the globals. ankylosing spondylitis, patient-reported outcomes, health, well-being, quality of life, validity Rheumatology key messages Patients and controls did not score globals for health, well-being and quality of life differently. Nevertheless, patients and controls identified content-related dissimilarities between the three constructs studied. For patients, physical impairments and limitations are important for health, well-being and quality of life. Introduction Patient globals are often used as an outcome measure in clinical studies [1]. Patient globals are self-reported, single-item questions that are usually scored on a visual analogue scale or a numeric rating scale. They can address a wide range of domains or constructs, varying from broad (e.g. overall health) to specific (e.g. pain, fatigue). Although clinimetric limitations of patient globals, such as end-of-scale avoidance, reference shift, subjective nature and implicitness of the information, are well recognized, outcome researchers do appreciate that patient globals have proven value with regard to construct validity and discrimination, including sensitivity to change [2, 3]. Importantly, patient globals are easy to apply in studies and reduce the risk of survey fatigue when researchers are interested in a large number of outcome domains. Increasingly, patient globals are used in clinical practice as they promptly provide healthcare professionals information on specific aspects of health and may help them to monitor disease outcomes over time. In AS, research on patient globals has been performed within several domains that are relevant to understanding the impact of disease on patients’ lives, comprising for example disease activity, fatigue and work-ability/productivity [4–6]. In addition, overall functioning and health is a relevant outcome in studies and clinical care. However, limited research has been done on patient globals for overall functioning and health. The terminology used for such constructs varies from general health and well-being to quality of life (QoL) or health-related QoL. Also, terms such as happiness and life satisfaction have been introduced [7, 8]. In AS, the thermometer from the EuroQoL 5 dimensions is frequently used to assess overall health [9], and the BAS-G to address well-being [10]. No official patient global is available to assess health-related QoL in AS, but several generic and disease-specific multidimensional approaches are used, among which most frequently the SF-36 [11, 12]. Commonly, available globals to assess overall impact of disease on patients are used as if they are interchangeable, and the exact wording of the question and anchors of the rating scale is frequently not mentioned. This practice raises questions since the constructs are suggested to represent (partly) different gestalts or concepts, although with overlapping components. In its purest meaning health refers to the absence of disease, but in the broad definition of 1948 from the World Health Organization (WHO), health refers to overall physical, mental and social well-being [13]. A tentative definition of the construct well-being proposed that it is ‘an umbrella term for different valuations that people make regarding their lives, the events happening to them, their bodies and minds, and the circumstances in which they live’ [14]. For QoL, several descriptions can be found in the literature, but usually contain elements of a person’s physical, material, social and emotional status, while often emphasizing the personal appraisal or satisfaction related to these aspects [15, 16]. More recently, the WHO concluded that the term well-being in the 1948 definition was confusing, and clarified that functioning (and health) and QoL are interrelated but not interchangeable [17]. While functioning (and health) refers to the objective performance, QoL refers to the subjects’ satisfaction about one’s performance. Consensus on similarities and dissimilarities among constructs and its measurement are potentially relevant for several reasons. At the level of healthcare decision making, there is a need to decide whether we want to prioritize spending of limited resources to improve overall health, well-being or QoL [8]. Further, to enhance comparability of clinical studies, evidence-based agreement would be helpful. Along these lines, it is relevant to know how patients perceive the constructs, that is, whether scores on the globals are different or identical, and whether reflection on scores might influence appraisal of the constructs. In addition, it is relevant to understand whether patients differ in this respect from population controls. If reference shift is an important underlying mechanism explaining dissimilarities in scores or understanding of the constructs, this might influence the choice for a certain construct [18]. Therefore, in this study we used a mixed quantitative and qualitative approach to evaluate whether patients with AS score single-item globals for health, well-being and QoL differently, whether these constructs have a different meaning to patients and whether patients score the globals differently after forced reflection. Further, we compared the scores and views of patients with those of population controls, to understand whether the presence of a chronic disease might change a subject’s view on different constructs. Methods Patients and controls Dutch patients participating in the Outcome in Ankylosing Spondylitis International Study (OASIS) cohort [19] were invited during their 12th year follow-up visit in 2008–09 to complete an additional questionnaire on health, well-being and QoL. OASIS is an unselected observational cohort of patients, fulfilling the modified New York criteria for AS [20], which started in 1996. Dutch patients were recruited from the Maastricht University Medical Center and a regional hospital in the near environment. Controls were recruited from three sources: all type of employers at the municipal administrative services of a city in the neighbourhood of the Maastricht University Medical Center; all type of employers of the Maastricht University library; and members of a musical society near Maastricht, the Netherlands. For recruitment of controls, it was explicitly mentioned to the employers to approach persons from all backgrounds and all job types, to ensure that the population was representative of the society. Both OASIS and the current sub-study were approved by the ethics committee of the Maastricht University Medical Center, and patients and controls signed informed consent. Questionnaires For the purpose of this study, information on age, gender, symptom duration, HLA-B27 status, history of extra-articular manifestations (uveitis, psoriasis, IBD), disease activity (CRP level, BASDAI [21], ankylosing spondylitis disease activity score-CRP [22]), physical functioning (BASFI) [23], medication use (NSAID, DMARD, biological) and radiographic score on spinal radiographic imaging (modified Stoke Ankylosing Spondylitis Spinal Score) [24], were retrieved from the OASIS 12th year visit. Controls provided information on demographics including age and gender and completed the Self-Report Comorbidity Questionnaires to particularly check for musculoskeletal comorbidity (back pain, OA, arthritis) [25]. Controls reporting arthritis were excluded from further analysis. Next, all subjects were asked to complete a self-composed questionnaire on overall health, well-being and QoL (Questionnaire in the Supplementary data, available at Rheumatology online). First, they were invited to score their health, well-being and QoL during the last week on a horizontal numeric rating scale from 0 (worst) to 10 (best health, well-being, QoL). To avoid an influence of task description across globals, instructions were similarly formulated with a single-sentence item. No reference to a specific disease was made. As an example, the question for well-being was: please tick a box that indicates the effect your health has had on your well-being over the last week. Second, participants were asked to indicate whether they considered dissimilarities between the constructs to be present (yes/no), and if so which pair of constructs differed (health vs well-being; health vs QoL; well-being vs QoL) or if all three represented something different. Next, they were invited to think about each of the three constructs and to write down what they were taking into account when scoring themselves (five lines available per construct). Finally, all subjects were asked to score the three globals a second time after this forced reflection. Statistical analyses Descriptive statistics were used to characterize patients and controls in terms of demographics and scores for health, well-being and QoL. Paired t-tests were performed to assess differences in scores before and after forced reflection for each of the constructs, both within patients and within controls. Independent t-tests were performed to assess the mean differences in scores between patients and controls on each construct. Differences in scores for each pair of construct (health/well-being, health/QoL or well-being/QoL) were plotted against their mean, using Bland and Altman plots, for patients and controls separately [26]. Mixed model analyses were used to test differences in scores between constructs and the role of group membership (patient or control) and forced reflection (before and after reflection), while adjusting for age and gender. For each of the sources of variation of interest, interactions were tested. For each pair of constructs, the proportion of patients and controls indicating that the constructs differed was calculated. All statistical analyses were performed using SPSS 23.0. The answers to the open question, in which the subjects could reflect what the constructs represented to them, were categorized into themes by going back and forth between answers and the list of themes. Each time an answer was given that could not be categorized into one of the themes available so far, a new theme was added. Themes were further linked to International Classification of Functioning, Disability and Health (ICF) components (body functions, activities and participation, environmental factor or personal factor) and next to the most precise ICF categories. A.G. executed the linking process after being familiarized with the ICF and ICF classification by A.B., who had followed an official training in linking at the ICF research branch. For themes that could not be immediately linked to an ICF category, A.G. and A.B. deliberated on the ICF chapter or category that best represented the information in the theme. For personal factors, for which no ICF classification exists as yet, the wording of the theme was maintained [27–29]. For each construct, the average number of themes identified, as well as the proportion of themes representing a specific ICF category, was calculated for patients and controls separately. An ICF category was considered relevant when 10% or more of all themes within the construct related to this ICF category. Results Of the initial 135 patients in the two Dutch centres of the OASIS study that took part in this project, 68 participated in the 12th year survey and all completed the supplementary questionnaire. Forty-seven (69%) were male, mean (s.d.) age was 56.8 (10.9) years and the average symptom duration was 34.9 (11.2) years. Of the 321 questionnaires that were sent out to healthy controls, 84 were returned (27% response). Forty-four were male (52%) with a mean (s.d.) age of 50.4 (13.9) years (Table 1). Fourteen (16.6%) controls reported a comorbidity that might impact daily functioning, but none of them reported arthritis. Characteristics of the participants are shown in Table 1. Table 1 Characteristics of patients and controls Variable Patients Controls Age, mean (s.d.), years 56.8 (10.9) 50.4 (13.9) Male gender, n (%) 47 (69) 44 (52) Symptom duration, mean (s.d.), years 34.9 (11.2) NA HLA-B27 positive, n (%) 57 (85) NA History of IBD, n (%) 15 (25) NA History of psoriasis, n (%) 6 (10) NA History of uveitis, n (%) 22 (36) NA CRP, mean (s.d.), mg/l 8.4 (13.4) NA BASDAI, 0–10, mean (s.d.) 3.7 (1.8) NA ASDAS-CRP, mean (s.d.) 2.3 (0.9) NA BASFI, 0–10, mean (s.d.) 4.6 (2.5) NA mSASSS score, 0–72, mean (s.d.) 27.6 (23.0) NA NSAID use, n (%) 39 (65) NA DMARDs use, n (%) 7 (12) NA Biological use, n (%) 18 (27) NA Variable Patients Controls Age, mean (s.d.), years 56.8 (10.9) 50.4 (13.9) Male gender, n (%) 47 (69) 44 (52) Symptom duration, mean (s.d.), years 34.9 (11.2) NA HLA-B27 positive, n (%) 57 (85) NA History of IBD, n (%) 15 (25) NA History of psoriasis, n (%) 6 (10) NA History of uveitis, n (%) 22 (36) NA CRP, mean (s.d.), mg/l 8.4 (13.4) NA BASDAI, 0–10, mean (s.d.) 3.7 (1.8) NA ASDAS-CRP, mean (s.d.) 2.3 (0.9) NA BASFI, 0–10, mean (s.d.) 4.6 (2.5) NA mSASSS score, 0–72, mean (s.d.) 27.6 (23.0) NA NSAID use, n (%) 39 (65) NA DMARDs use, n (%) 7 (12) NA Biological use, n (%) 18 (27) NA ASDAS: ankylosing spondylitis disease activity score; mSASSS: modified Stoke Ankylosing Spondylitis Spinal Score; NA: not assessed. Table 1 Characteristics of patients and controls Variable Patients Controls Age, mean (s.d.), years 56.8 (10.9) 50.4 (13.9) Male gender, n (%) 47 (69) 44 (52) Symptom duration, mean (s.d.), years 34.9 (11.2) NA HLA-B27 positive, n (%) 57 (85) NA History of IBD, n (%) 15 (25) NA History of psoriasis, n (%) 6 (10) NA History of uveitis, n (%) 22 (36) NA CRP, mean (s.d.), mg/l 8.4 (13.4) NA BASDAI, 0–10, mean (s.d.) 3.7 (1.8) NA ASDAS-CRP, mean (s.d.) 2.3 (0.9) NA BASFI, 0–10, mean (s.d.) 4.6 (2.5) NA mSASSS score, 0–72, mean (s.d.) 27.6 (23.0) NA NSAID use, n (%) 39 (65) NA DMARDs use, n (%) 7 (12) NA Biological use, n (%) 18 (27) NA Variable Patients Controls Age, mean (s.d.), years 56.8 (10.9) 50.4 (13.9) Male gender, n (%) 47 (69) 44 (52) Symptom duration, mean (s.d.), years 34.9 (11.2) NA HLA-B27 positive, n (%) 57 (85) NA History of IBD, n (%) 15 (25) NA History of psoriasis, n (%) 6 (10) NA History of uveitis, n (%) 22 (36) NA CRP, mean (s.d.), mg/l 8.4 (13.4) NA BASDAI, 0–10, mean (s.d.) 3.7 (1.8) NA ASDAS-CRP, mean (s.d.) 2.3 (0.9) NA BASFI, 0–10, mean (s.d.) 4.6 (2.5) NA mSASSS score, 0–72, mean (s.d.) 27.6 (23.0) NA NSAID use, n (%) 39 (65) NA DMARDs use, n (%) 7 (12) NA Biological use, n (%) 18 (27) NA ASDAS: ankylosing spondylitis disease activity score; mSASSS: modified Stoke Ankylosing Spondylitis Spinal Score; NA: not assessed. Table 2 shows the scores on the three constructs for patients and controls. Scores before and after reflection were not significantly different. Patients scored significantly worse on all three globals compared with controls. Mixed modelling adjusted for age and gender indicated that only the scores between patients and controls differed significantly (P < 0.01), but did not reveal any effect attributable to the construct assessed, or to an effect of forced reflection (Table 3). Significant interactions were not found between construct and group, or between the effect of forced reflection and group. The Bland and Altman plots visualize the difference in scores (before reflection) between health/well-being, health/QoL and well-being/QoL against the mean, separately for patients and controls (Fig. 1). In some individuals the score between two constructs showed large dissimilarity, with differences up to five points. However, there seems no pattern in these dissimilarities. Table 2 Scores on the three globals for patients and controls, before and after forced reflection Patients, n = 68 Controls, n = 84 Patients vs controls mean difference; P-valuea Construct Before After P-valueb Before After P-valueb Before After Health 6.1 (2.5) 6.2 (2.4) 0.50 7.6 (1.6) 7.5 (1.5) 0.50 1.4; <0.01 1.3; <0.01 Well-being 6.1 (2.5) 6.1 (2.4) 0.92 7.2 (1.3) 7.4 (1.3) 0.06 1.1; <0.01 1.3; <0.01 Quality of life 6.3 (2.6) 6.4 (2.4) 0.93 7.5 (1.6) 7.5 (1.4) 0.43 1.0; <0.01 1.2; <0.01 Patients, n = 68 Controls, n = 84 Patients vs controls mean difference; P-valuea Construct Before After P-valueb Before After P-valueb Before After Health 6.1 (2.5) 6.2 (2.4) 0.50 7.6 (1.6) 7.5 (1.5) 0.50 1.4; <0.01 1.3; <0.01 Well-being 6.1 (2.5) 6.1 (2.4) 0.92 7.2 (1.3) 7.4 (1.3) 0.06 1.1; <0.01 1.3; <0.01 Quality of life 6.3 (2.6) 6.4 (2.4) 0.93 7.5 (1.6) 7.5 (1.4) 0.43 1.0; <0.01 1.2; <0.01 Mean (s.d.) values are presented. a Rounding differences may be present, independent t-test. b Paired t-test. Table 2 Scores on the three globals for patients and controls, before and after forced reflection Patients, n = 68 Controls, n = 84 Patients vs controls mean difference; P-valuea Construct Before After P-valueb Before After P-valueb Before After Health 6.1 (2.5) 6.2 (2.4) 0.50 7.6 (1.6) 7.5 (1.5) 0.50 1.4; <0.01 1.3; <0.01 Well-being 6.1 (2.5) 6.1 (2.4) 0.92 7.2 (1.3) 7.4 (1.3) 0.06 1.1; <0.01 1.3; <0.01 Quality of life 6.3 (2.6) 6.4 (2.4) 0.93 7.5 (1.6) 7.5 (1.4) 0.43 1.0; <0.01 1.2; <0.01 Patients, n = 68 Controls, n = 84 Patients vs controls mean difference; P-valuea Construct Before After P-valueb Before After P-valueb Before After Health 6.1 (2.5) 6.2 (2.4) 0.50 7.6 (1.6) 7.5 (1.5) 0.50 1.4; <0.01 1.3; <0.01 Well-being 6.1 (2.5) 6.1 (2.4) 0.92 7.2 (1.3) 7.4 (1.3) 0.06 1.1; <0.01 1.3; <0.01 Quality of life 6.3 (2.6) 6.4 (2.4) 0.93 7.5 (1.6) 7.5 (1.4) 0.43 1.0; <0.01 1.2; <0.01 Mean (s.d.) values are presented. a Rounding differences may be present, independent t-test. b Paired t-test. Table 3 Mixed modelling analysis across scores on the three constructs of interest Source of variation Estimate 95% CI P-value Increasing age −0.01 −0.02; 0.00 0.11 Female gender −0.05 −0.32; 0.23 0.74 Patients vs controls −1.28 −1.56; −1.01 <0.01 Construct 0.11 −0.04; 0.27 0.16 Time (before–after reflection) −0.02 −0.28; 0.24 0.89 Intercept 7.73 6.95; 8.51 <0.01 Source of variation Estimate 95% CI P-value Increasing age −0.01 −0.02; 0.00 0.11 Female gender −0.05 −0.32; 0.23 0.74 Patients vs controls −1.28 −1.56; −1.01 <0.01 Construct 0.11 −0.04; 0.27 0.16 Time (before–after reflection) −0.02 −0.28; 0.24 0.89 Intercept 7.73 6.95; 8.51 <0.01 Table 3 Mixed modelling analysis across scores on the three constructs of interest Source of variation Estimate 95% CI P-value Increasing age −0.01 −0.02; 0.00 0.11 Female gender −0.05 −0.32; 0.23 0.74 Patients vs controls −1.28 −1.56; −1.01 <0.01 Construct 0.11 −0.04; 0.27 0.16 Time (before–after reflection) −0.02 −0.28; 0.24 0.89 Intercept 7.73 6.95; 8.51 <0.01 Source of variation Estimate 95% CI P-value Increasing age −0.01 −0.02; 0.00 0.11 Female gender −0.05 −0.32; 0.23 0.74 Patients vs controls −1.28 −1.56; −1.01 <0.01 Construct 0.11 −0.04; 0.27 0.16 Time (before–after reflection) −0.02 −0.28; 0.24 0.89 Intercept 7.73 6.95; 8.51 <0.01 Fig. 1 View largeDownload slide Bland and Altman plots for each pair of constructs The Bland and Altman plots visualize the difference against the mean in scores for each pair of constructs, for patients (A–C) and controls (D–F) separately. Fig. 1 View largeDownload slide Bland and Altman plots for each pair of constructs The Bland and Altman plots visualize the difference against the mean in scores for each pair of constructs, for patients (A–C) and controls (D–F) separately. Forty-four (64%) patients and 31 (36%) controls indicated there was no dissimilarity between one or more pair of constructs. Patients most frequently considered health to be different from either well-being or QoL (n = 12 and n = 8, respectively). Two patients (8%) and 32 controls (60%) considered that all three globals represented other constructs. Themes when reflecting on, respectively, health, well-being and QoL were provided by 50 (74%), 39 (57%) and 38 (56%) of the patients and 73 (87%), 64 (76%) and 69 (82%) of the controls (Table 4). The average number of themes provided per construct was 2.5, 1.8 and 1.6 by the patients, and 3.6, 3.7 and 3.4 by the controls for health, well-being and QoL, respectively. Table 4 shows for each construct, the ICF category names representing the themes and the proportion of the total identified themes per ICF category, separately for patients and controls. Patients with AS associated health in particular with body functions, more specifically with pain, limitations in movement functions and energy level. Controls also associated health with (absence of) pain and (absence of) limitations in movement. In addition, some impairments not mentioned by patients were brought forward by controls, mainly seeing and hearing functions and other emotional functions. Well-being in patients was again associated with pain, movement functions and energy level, but also with depressed feelings and activities related to mobility (cycling, walking and running). Controls considered more frequently other emotions and social life, and (although just below our 10% threshold) also attitudes of colleagues and leisure as part of well-being. For QoL, patients identified again energy and limitations in movement function as relevant aspects, and in addition, activities related to mobility and social life. For controls, QoL was more frequently related to social life and leisure activities, but also to economic self-sufficiency (just below our 10% threshold). Interestingly, patients but not controls identified to some extent personal factors, among which coping styles, to be relevant facilitators for health (dealing with limitations, adapting to pain, adjusting to energy) and QoL (accepting). Table 4 Themes organised according to ICF categories, separately for patients and controls ICF classification ICF category name Patients (n = 68) Controls (n = 84) Health, n = 50 (74%) Well-being, n = 39 (57%) QoL, n = 38 (56%) Health, n = 73 (87%) Well-being, n = 64 (76%) QoL, n = 69 (82%) n = 124 themes n = 70 themes n = 61 themes n = 262 themes n = 235 themes n = 233 themes Body functions and mental functions Pain (b280) 14.5 15.7 8.2 11.4 2.1 5.5 Mobility (movement functions) 21.8 14.3 14.8 21.0 2.7 4.0 Energy level (b1300) 12.1 12.9 9.8 8.5 1.4 Sleep functions (b134) 9.7 8.6 3.3 3.4 3.4 0.9 Other body complaints (e.g. seeing, hearing) 6.5a 2.9 1.6 12.5 4.1 3.2 Anxiety (b152) 1.1 0.7 2.4 Depressed feelings/emotional functions (b152) 4.0 12.9 6.6 Stress (b152) 2.8 6.2 3.2 Other emotional functions (b1528) 1.6 8.9 1.6 13.1 17.1 7.2 Activities and participation Social life (d7) 4.0 5.7 11.5 0.6 22.6 23.7 Recreation and leisure (d920) 5.6 6.6 2.3 7.5 11.2 Education and work (d8) 0.8 1.6 2.8 3.4 3.2 Household tasks (d6) 4.0 3.3 0.6 1.4 0.9 Mobility (moving around) (d5) 3.2 12.9 11.5 4.0 5.5 4.0 Environmental factors Attitudes of colleagues (e325) 7.5 4.9 Financial self-sufficiency (e165) 0.8 1.4 6.6 2.7 9.0 Natural environment (e2) 2.8 5.5 4.9 Products and technologies (e1) 1.1 0.7 0.9 Food (e110) 9.7 4.1 5.5 Health services (e580) 1.7 Daily necessities (e1) 1.4 3.2 Personal factors Coping styles (pf) 6.5 2.9 8.2 Others 4.8a 1.4 1.6 2.8 1.8 ICF classification ICF category name Patients (n = 68) Controls (n = 84) Health, n = 50 (74%) Well-being, n = 39 (57%) QoL, n = 38 (56%) Health, n = 73 (87%) Well-being, n = 64 (76%) QoL, n = 69 (82%) n = 124 themes n = 70 themes n = 61 themes n = 262 themes n = 235 themes n = 233 themes Body functions and mental functions Pain (b280) 14.5 15.7 8.2 11.4 2.1 5.5 Mobility (movement functions) 21.8 14.3 14.8 21.0 2.7 4.0 Energy level (b1300) 12.1 12.9 9.8 8.5 1.4 Sleep functions (b134) 9.7 8.6 3.3 3.4 3.4 0.9 Other body complaints (e.g. seeing, hearing) 6.5a 2.9 1.6 12.5 4.1 3.2 Anxiety (b152) 1.1 0.7 2.4 Depressed feelings/emotional functions (b152) 4.0 12.9 6.6 Stress (b152) 2.8 6.2 3.2 Other emotional functions (b1528) 1.6 8.9 1.6 13.1 17.1 7.2 Activities and participation Social life (d7) 4.0 5.7 11.5 0.6 22.6 23.7 Recreation and leisure (d920) 5.6 6.6 2.3 7.5 11.2 Education and work (d8) 0.8 1.6 2.8 3.4 3.2 Household tasks (d6) 4.0 3.3 0.6 1.4 0.9 Mobility (moving around) (d5) 3.2 12.9 11.5 4.0 5.5 4.0 Environmental factors Attitudes of colleagues (e325) 7.5 4.9 Financial self-sufficiency (e165) 0.8 1.4 6.6 2.7 9.0 Natural environment (e2) 2.8 5.5 4.9 Products and technologies (e1) 1.1 0.7 0.9 Food (e110) 9.7 4.1 5.5 Health services (e580) 1.7 Daily necessities (e1) 1.4 3.2 Personal factors Coping styles (pf) 6.5 2.9 8.2 Others 4.8a 1.4 1.6 2.8 1.8 The proportion with which a theme was indicated within a construct is presented and sums up to 100% per construct (rounding differences may be present). aSome are related to side effects of treatment. QoL: quality of life; ICF: International Classification of Functioning and Health; pf: personal factor. Table 4 Themes organised according to ICF categories, separately for patients and controls ICF classification ICF category name Patients (n = 68) Controls (n = 84) Health, n = 50 (74%) Well-being, n = 39 (57%) QoL, n = 38 (56%) Health, n = 73 (87%) Well-being, n = 64 (76%) QoL, n = 69 (82%) n = 124 themes n = 70 themes n = 61 themes n = 262 themes n = 235 themes n = 233 themes Body functions and mental functions Pain (b280) 14.5 15.7 8.2 11.4 2.1 5.5 Mobility (movement functions) 21.8 14.3 14.8 21.0 2.7 4.0 Energy level (b1300) 12.1 12.9 9.8 8.5 1.4 Sleep functions (b134) 9.7 8.6 3.3 3.4 3.4 0.9 Other body complaints (e.g. seeing, hearing) 6.5a 2.9 1.6 12.5 4.1 3.2 Anxiety (b152) 1.1 0.7 2.4 Depressed feelings/emotional functions (b152) 4.0 12.9 6.6 Stress (b152) 2.8 6.2 3.2 Other emotional functions (b1528) 1.6 8.9 1.6 13.1 17.1 7.2 Activities and participation Social life (d7) 4.0 5.7 11.5 0.6 22.6 23.7 Recreation and leisure (d920) 5.6 6.6 2.3 7.5 11.2 Education and work (d8) 0.8 1.6 2.8 3.4 3.2 Household tasks (d6) 4.0 3.3 0.6 1.4 0.9 Mobility (moving around) (d5) 3.2 12.9 11.5 4.0 5.5 4.0 Environmental factors Attitudes of colleagues (e325) 7.5 4.9 Financial self-sufficiency (e165) 0.8 1.4 6.6 2.7 9.0 Natural environment (e2) 2.8 5.5 4.9 Products and technologies (e1) 1.1 0.7 0.9 Food (e110) 9.7 4.1 5.5 Health services (e580) 1.7 Daily necessities (e1) 1.4 3.2 Personal factors Coping styles (pf) 6.5 2.9 8.2 Others 4.8a 1.4 1.6 2.8 1.8 ICF classification ICF category name Patients (n = 68) Controls (n = 84) Health, n = 50 (74%) Well-being, n = 39 (57%) QoL, n = 38 (56%) Health, n = 73 (87%) Well-being, n = 64 (76%) QoL, n = 69 (82%) n = 124 themes n = 70 themes n = 61 themes n = 262 themes n = 235 themes n = 233 themes Body functions and mental functions Pain (b280) 14.5 15.7 8.2 11.4 2.1 5.5 Mobility (movement functions) 21.8 14.3 14.8 21.0 2.7 4.0 Energy level (b1300) 12.1 12.9 9.8 8.5 1.4 Sleep functions (b134) 9.7 8.6 3.3 3.4 3.4 0.9 Other body complaints (e.g. seeing, hearing) 6.5a 2.9 1.6 12.5 4.1 3.2 Anxiety (b152) 1.1 0.7 2.4 Depressed feelings/emotional functions (b152) 4.0 12.9 6.6 Stress (b152) 2.8 6.2 3.2 Other emotional functions (b1528) 1.6 8.9 1.6 13.1 17.1 7.2 Activities and participation Social life (d7) 4.0 5.7 11.5 0.6 22.6 23.7 Recreation and leisure (d920) 5.6 6.6 2.3 7.5 11.2 Education and work (d8) 0.8 1.6 2.8 3.4 3.2 Household tasks (d6) 4.0 3.3 0.6 1.4 0.9 Mobility (moving around) (d5) 3.2 12.9 11.5 4.0 5.5 4.0 Environmental factors Attitudes of colleagues (e325) 7.5 4.9 Financial self-sufficiency (e165) 0.8 1.4 6.6 2.7 9.0 Natural environment (e2) 2.8 5.5 4.9 Products and technologies (e1) 1.1 0.7 0.9 Food (e110) 9.7 4.1 5.5 Health services (e580) 1.7 Daily necessities (e1) 1.4 3.2 Personal factors Coping styles (pf) 6.5 2.9 8.2 Others 4.8a 1.4 1.6 2.8 1.8 The proportion with which a theme was indicated within a construct is presented and sums up to 100% per construct (rounding differences may be present). aSome are related to side effects of treatment. QoL: quality of life; ICF: International Classification of Functioning and Health; pf: personal factor. Discussion In the present study, 65% of the patients and 37% of controls indicated that health, well-being and QoL mean similar things to them, and this is also reflected in there being no meaningful differences in the absolute scores on the patient globals for these constructs. Moreover, the scores on the different globals did not systematically change after reflecting on the content of the constructs, suggesting that the information subjects use when scoring these constructs is implicitly accounted for. While there were some similarities in themes that patients and controls associated with the constructs, there were also clear dissimilarities in the type and frequency of aspects considered relevant when scoring themselves for each of the three constructs. On average, patients took into account fewer aspects per construct compared with controls. Across constructs however, patients identified more aspects when rating overall health than when rating the other two constructs, while controls tended to identify a similar number of different themes for each of the three constructs. Although survey fatigue cannot be entirely ruled out, as patients completed this questionnaire in the same period as the regular OASIS follow-up questionnaire, it seems more plausible that patients—due to the presence of a chronic disease—have on average a more narrow biomedical focus when evaluating health, whereas controls have a broader vision, likely also taking into account potential diseases that are not (yet) present. On the same line, patients associated disease-related aspects, in particular pain, fatigue, movement functions and mobility, to all three constructs. For controls, impairments in body and mental functions were an essential part of experienced health, but less for well-being and QoL. For both patients and controls, emotional functions were frequently identified within the construct well-being. Controls also considered social life when evaluating well-being, and economic self-sufficiency when evaluating QoL. This was less clear for patients. It is of note that in our study, coping or beliefs were identified by patients as relevant factors when evaluating their own overall health and QoL, whereas controls never considered personal factors in relation to the constructs. The current data suggest that patients implicitly consider their level of adaptation and coping when scoring globals. The qualitative findings of our study are partly in accordance with a study from Fagerlind et al. [30], who reported on the constructs health and QoL from semi-structured interviews among 22 patients with RA. They emphasized that the understanding of the constructs are both individual and heterogeneous. In line with our findings, participants recognized overlap between the constructs, and identified impairments in body functions and limitations in activities as key parts of both constructs. While in the interviews in the Fagerlind study well-being was not included as a separate construct, their patients spontaneously mentioned well-being as (an exclusive) part of the construct health. Living conditions, social network but also a positive look on life were exclusively related to QoL. In contrast, we found that half of the patients who reported dissimilarities between the globals indicated that health differed from well-being. While not being the primary focus of the study, it was noted that patients with AS scored health, well-being and QoL only slightly lower than healthy subjects (on average 1.0–1.4 lower on a 0–10 scale). Despite the fact that no formal thresholds are available to decide whether these between-group differences are clinically relevant, but keeping in mind that differences of 1.5 are often considered clinically relevant at the individual patient level [31], we had expected larger differences between patients and controls. An attractive explanation might be that reference shift plays an important role here: that patients adapt to their disease and adjust their sense of normal [18]. Our qualitative study seems to suggest that personal resources (coping, adapting) may play an even more important role than to social context when rating overall health, well-being and QoL. Our study has clear limitations. First, although face-to-face interviews or cognitive debriefings might be a preferred qualitative approach, such methods are limited in the number of subjects that can be included. The large amount of qualitative data in our study allowed semi-quantitative analyses. Second, we have chosen a limited number of constructs that might reflect the life impact of disease. We could have added health-related QoL, happiness, satisfaction and likely other constructs. However, it was not our aim to be exhaustive on that line. We wished to explore commonly used constructs. Third, our study design did not account for a possible order effect, as we did not change the order of the globals in subsets of the survey. It cannot be excluded that the score on well-being, which was the first question, influenced the scores on the other constructs. Fourth, the question on generalizability should be addressed. This study was done in patients with AS with longstanding disease. While we think that the main conclusions will hold for other rheumatic diseases, it would be interesting to understand the change over the course of the disease in scores and underlying themes that drive these scores compared with controls. Fifth, due to low response in the control population we were not able to match for age and gender. While scores have been adjusted statistically for such differences, this was not feasible for the qualitative part of the study. Furthermore, selection bias might have occurred in the control population. We cannot exclude that this might have influenced the results. In summary, while patients and controls recognized that health, well-being and QoL were different constructs, this was not reflected in differences in the mean scores on the globals. For patients, themes related to physical impairments and limitations were relevant for health, well-being and QoL, whereas controls related these themes to health only. For patients and controls, emotional functions were specifically related to well-being. For controls, social life and economic self-sufficiency were important for QoL. The final score for any global measure seems to be the result of an implicit but complex process, in which themes are weighted and appraised in a varying ways. Acknowledgement Conception and design: A.B.; data acquisition: A.G. and A.B.; data analysis and interpretation of data: all authors; drafting the work or revising it critically for important intellectual content: all authors; final approval: all authors. Funding: No specific funding was received from any bodies in the public, commercial or not-for-profit sectors to carry out the work described in this manuscript. Disclosure statement: The authors have declared no conflicts of interest. Supplementary data Supplementary data are available at Rheumatology online. References 1 Kilic L , Erden A , Bingham CO 3rd , Gossec L , Kalyoncu U. The reporting of patient-reported outcomes in studies of patients with rheumatoid arthritis: a systematic review of 250 articles . J Rheumatol 2016 ; 43 : 1300 – 5 . Google Scholar CrossRef Search ADS PubMed 2 Harrison MJ , Boonen A , Tugwell P , Symmons DP. Same question, different answers: a comparison of global health assessments using visual analogue scales . Qual Life Res 2009 ; 18 : 1285 – 92 . Google Scholar CrossRef Search ADS PubMed 3 Krabbe PF , Essink-Bot ML , Bonsel GJ. The comparability and reliability of five health-state valuation methods . Soc Sci Med 1997 ; 45 : 1641 – 52 . Google Scholar CrossRef Search ADS PubMed 4 Spoorenberg A , van Tubergen A , Landewe R et al. Measuring disease activity in ankylosing spondylitis: patient and physician have different perspectives . Rheumatology (Oxford) 2005 ; 44 : 789 – 95 . Google Scholar CrossRef Search ADS PubMed 5 van Tubergen A , Coenen J , Landewé R et al. Assessment of fatigue in patients with ankylosing spondylitis: a psychometric analysis . Arthritis Rheum 2002 ; 47 : 8 – 16 . Google Scholar CrossRef Search ADS PubMed 6 Escorpizo R , Bombardier C , Boonen A et al. Worker productivity outcome measures in arthritis . J Rheumatol 2007 ; 34 : 1372 – 80 . Google Scholar PubMed 7 Fowler JH , Christakis NA. Dynamic spread of happiness in a large social network: longitudinal analysis over 20 years in the Framingham Heart Study . BMJ 2008 ; 337 : a2338 . Google Scholar CrossRef Search ADS PubMed 8 Naci H , Ioannidis JPA. Evaluation of wellness determinants and interventions by citizen scientists . JAMA 2015 ; 314 : 121 – 2 . Google Scholar CrossRef Search ADS PubMed 9 EuroQol Group. EuroQol–a new facility for the measurement of health-related quality of life . Health Policy 1990 ; 16 : 199 – 208 . CrossRef Search ADS PubMed 10 Jones SD , Steiner A , Garrett SL , Calin A. The Bath Ankylosing Spondylitis Patient Global Score (BAS-G) . Br J Rheumatol 1996 ; 35 : 66 – 71 . Google Scholar CrossRef Search ADS PubMed 11 Aaronson NK , Muller M , Cohen PD et al. Translation, validation, and norming of the Dutch language version of the SF-36 Health Survey in community and chronic disease populations . J Clin Epidemiol 1998 ; 51 : 1055 – 68 . Google Scholar CrossRef Search ADS PubMed 12 Doward LC , Spoorenberg A , Cook SA et al. Development of the ASQoL: a quality of life instrument specific to ankylosing spondylitis . Ann Rheum Dis 2003 ; 62 : 20 – 6 . Google Scholar CrossRef Search ADS PubMed 13 WHO . Constitution of the World Health Organization: Principles 1948. http://apps.who.int/gb/bd/PDF/bd47/EN/constitution-en.pdf? ua=1 (16 March 2018, date last accessed). 14 Diener E. Guidelines for national indicators of subjective well-being and ill-being . J Happiness Stud 2006 ; 7 : 397 – 404 . Google Scholar CrossRef Search ADS 15 Felce D , Perry J. Quality of life: its definition and measurement . Res Dev Disabil 1995 ; 16 : 51 – 74 . Google Scholar CrossRef Search ADS PubMed 16 Eurostat . Quality of Life Indicators - Measuring Quality of Life 2015 [updated 05-11-2015]. http://ec.europa.eu/eurostat/statistics-explained/index.php/Quality_of_life_indicators_-_measuring_quality_of_life (16 March 2018, date last accessed). 17 World Health Organization. Division of Mental Health and Prevention of Substance Abuse . ( 1997 ). WHOQOL : measuring quality of life. Geneva : World Health Organization. http://www.who.int/iris/handle/10665/63482 (7 May 2018, date last accessed). 18 Rapkin BD , Schwartz CE. Toward a theoretical model of quality-of-life appraisal: implications of findings from studies of response shift . Health Qual Life Outcomes 2004 ; 2 : 14 . Google Scholar CrossRef Search ADS PubMed 19 Spoorenberg A , van der Heijde D , de Klerk E et al. Relative value of erythrocyte sedimentation rate and C-reactive protein in assessment of disease activity in ankylosing spondylitis . J Rheumatol 1999 ; 26 : 980 – 4 . Google Scholar PubMed 20 van der Linden S , Valkenburg HA , Cats A. Evaluation of diagnostic criteria for ankylosing spondylitis. A proposal for modification of the New York criteria . Arthritis Rheum 1984 ; 27 : 361 – 8 . Google Scholar CrossRef Search ADS PubMed 21 Garrett S , Jenkinson T , Kennedy LG et al. A new approach to defining disease status in ankylosing spondylitis: the Bath Ankylosing Spondylitis Disease Activity Index . J Rheumatol 1994 ; 21 : 2286 – 91 . Google Scholar PubMed 22 Lukas C , Landewe R , Sieper J et al. Development of an ASAS-endorsed disease activity score (ASDAS) in patients with ankylosing spondylitis . Ann Rheum Dis 2009 ; 68 : 18 – 24 . Google Scholar CrossRef Search ADS PubMed 23 Calin A , Garrett S , Whitelock H et al. A new approach to defining functional ability in ankylosing spondylitis: the development of the Bath Ankylosing Spondylitis Functional Index . J Rheumatol 1994 ; 21 : 2281 – 5 . Google Scholar PubMed 24 Creemers MC , Franssen MJ , van’t Hof MA et al. Assessment of outcome in ankylosing spondylitis: an extended radiographic scoring system . Ann Rheum Dis 2005 ; 64 : 127 – 9 . Google Scholar CrossRef Search ADS PubMed 25 Sangha O , Stucki G , Liang MH , Fossel AH , Katz JN. The Self-Administered Comorbidity Questionnaire: a new method to assess comorbidity for clinical and health services research . Arthritis Rheum 2003 ; 49 : 156 – 63 . Google Scholar CrossRef Search ADS PubMed 26 Altman DG , Bland J. Measurement in medicine: the analysis of method comparison studies . J R Stat Soc Ser D (The Statistician) 1983 ; 32 : 307 – 17 . 27 World Health Organisation . International Classification of Functioning, Disability and Health: ICF. Geneva: WHO 2001 . 28 Cieza A , Brockow T , Ewert T et al. Linking health-status measurements to the international classification of functioning, disability and health . J Rehabil Med 2002 ; 34 : 205 – 10 . Google Scholar CrossRef Search ADS PubMed 29 Cieza A , Geyh S , Chatterji S et al. ICF linking rules: an update based on lessons learned . J Rehabil Med 2005 ; 37 : 212 – 8 . Google Scholar CrossRef Search ADS PubMed 30 Fagerlind H , Ring L , Brülde B , Feltelius N , Lindblad AK. Patients’ understanding of the concepts of health and quality of life . Patient Educ Couns 2010 ; 78 : 104 – 10 . Google Scholar CrossRef Search ADS PubMed 31 Strand V , Boers M , Idzerda L et al. It’s good to feel better but it’s better to feel good and even better to feel good as soon as possible for as long as possible. Response criteria and the importance of change at OMERACT 10 . J Rheumatol 2011 ; 38 : 1720 – 7 . Google Scholar CrossRef Search ADS PubMed © The Author(s) 2018. Published by Oxford University Press on behalf of the British Society for Rheumatology. All rights reserved. For permissions, please email: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Rheumatology Oxford University Press

Are globals for health, well-being and quality of life interchangeable? A mixed methods study in ankylosing spondylitis patients and controls

Loading next page...
 
/lp/ou_press/are-globals-for-health-well-being-and-quality-of-life-interchangeable-GDZYlxnwCQ
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press on behalf of the British Society for Rheumatology. All rights reserved. For permissions, please email: journals.permissions@oup.com
ISSN
1462-0324
eISSN
1462-0332
D.O.I.
10.1093/rheumatology/key129
Publisher site
See Article on Publisher Site

Abstract

Abstract Objective Patients’ experience of overall health is often assessed through a single-item global question. Here, we evaluated among patients with AS and population controls whether single-item questions on the constructs health, well-being and quality of life (QoL) are interchangeable. Methods In a mixed quantitative and qualitative approach, all subjects scored the three single-item globals on a numeric rating scale (0–10, best). Next, they indicated for each of the questions which aspects they had been considering when scoring. After forced reflection, globals were scored again. Dissimilarities in scores among constructs, between patients and controls, and before or after reflection were tested using mixed linear models. Themes identified per construct in the qualitative part were linked to the International Classification of Functioning, Disability and Health. The type of themes per construct was compared between patients and controls. Results Sixty-eight AS patients and 84 controls completed the questionnaire. Patients scored significantly worse on each global than controls (mean 6.1–6.3 vs 7.2–7.6, all P < 0.01). Within groups, however, no significant differences in scores on each construct, or in scores before or after forced reflection were found. Health-related themes were relevant to each construct for patients, but were less relevant for controls when considering well-being and QoL. Emotional functions were relevant to well-being in all participants. Social roles and financial situation were more frequently related to well-being and QoL in controls. Conclusion While patients and controls identified content-related dissimilarities between the three constructs studied, this was not reflected in different scores of the globals. ankylosing spondylitis, patient-reported outcomes, health, well-being, quality of life, validity Rheumatology key messages Patients and controls did not score globals for health, well-being and quality of life differently. Nevertheless, patients and controls identified content-related dissimilarities between the three constructs studied. For patients, physical impairments and limitations are important for health, well-being and quality of life. Introduction Patient globals are often used as an outcome measure in clinical studies [1]. Patient globals are self-reported, single-item questions that are usually scored on a visual analogue scale or a numeric rating scale. They can address a wide range of domains or constructs, varying from broad (e.g. overall health) to specific (e.g. pain, fatigue). Although clinimetric limitations of patient globals, such as end-of-scale avoidance, reference shift, subjective nature and implicitness of the information, are well recognized, outcome researchers do appreciate that patient globals have proven value with regard to construct validity and discrimination, including sensitivity to change [2, 3]. Importantly, patient globals are easy to apply in studies and reduce the risk of survey fatigue when researchers are interested in a large number of outcome domains. Increasingly, patient globals are used in clinical practice as they promptly provide healthcare professionals information on specific aspects of health and may help them to monitor disease outcomes over time. In AS, research on patient globals has been performed within several domains that are relevant to understanding the impact of disease on patients’ lives, comprising for example disease activity, fatigue and work-ability/productivity [4–6]. In addition, overall functioning and health is a relevant outcome in studies and clinical care. However, limited research has been done on patient globals for overall functioning and health. The terminology used for such constructs varies from general health and well-being to quality of life (QoL) or health-related QoL. Also, terms such as happiness and life satisfaction have been introduced [7, 8]. In AS, the thermometer from the EuroQoL 5 dimensions is frequently used to assess overall health [9], and the BAS-G to address well-being [10]. No official patient global is available to assess health-related QoL in AS, but several generic and disease-specific multidimensional approaches are used, among which most frequently the SF-36 [11, 12]. Commonly, available globals to assess overall impact of disease on patients are used as if they are interchangeable, and the exact wording of the question and anchors of the rating scale is frequently not mentioned. This practice raises questions since the constructs are suggested to represent (partly) different gestalts or concepts, although with overlapping components. In its purest meaning health refers to the absence of disease, but in the broad definition of 1948 from the World Health Organization (WHO), health refers to overall physical, mental and social well-being [13]. A tentative definition of the construct well-being proposed that it is ‘an umbrella term for different valuations that people make regarding their lives, the events happening to them, their bodies and minds, and the circumstances in which they live’ [14]. For QoL, several descriptions can be found in the literature, but usually contain elements of a person’s physical, material, social and emotional status, while often emphasizing the personal appraisal or satisfaction related to these aspects [15, 16]. More recently, the WHO concluded that the term well-being in the 1948 definition was confusing, and clarified that functioning (and health) and QoL are interrelated but not interchangeable [17]. While functioning (and health) refers to the objective performance, QoL refers to the subjects’ satisfaction about one’s performance. Consensus on similarities and dissimilarities among constructs and its measurement are potentially relevant for several reasons. At the level of healthcare decision making, there is a need to decide whether we want to prioritize spending of limited resources to improve overall health, well-being or QoL [8]. Further, to enhance comparability of clinical studies, evidence-based agreement would be helpful. Along these lines, it is relevant to know how patients perceive the constructs, that is, whether scores on the globals are different or identical, and whether reflection on scores might influence appraisal of the constructs. In addition, it is relevant to understand whether patients differ in this respect from population controls. If reference shift is an important underlying mechanism explaining dissimilarities in scores or understanding of the constructs, this might influence the choice for a certain construct [18]. Therefore, in this study we used a mixed quantitative and qualitative approach to evaluate whether patients with AS score single-item globals for health, well-being and QoL differently, whether these constructs have a different meaning to patients and whether patients score the globals differently after forced reflection. Further, we compared the scores and views of patients with those of population controls, to understand whether the presence of a chronic disease might change a subject’s view on different constructs. Methods Patients and controls Dutch patients participating in the Outcome in Ankylosing Spondylitis International Study (OASIS) cohort [19] were invited during their 12th year follow-up visit in 2008–09 to complete an additional questionnaire on health, well-being and QoL. OASIS is an unselected observational cohort of patients, fulfilling the modified New York criteria for AS [20], which started in 1996. Dutch patients were recruited from the Maastricht University Medical Center and a regional hospital in the near environment. Controls were recruited from three sources: all type of employers at the municipal administrative services of a city in the neighbourhood of the Maastricht University Medical Center; all type of employers of the Maastricht University library; and members of a musical society near Maastricht, the Netherlands. For recruitment of controls, it was explicitly mentioned to the employers to approach persons from all backgrounds and all job types, to ensure that the population was representative of the society. Both OASIS and the current sub-study were approved by the ethics committee of the Maastricht University Medical Center, and patients and controls signed informed consent. Questionnaires For the purpose of this study, information on age, gender, symptom duration, HLA-B27 status, history of extra-articular manifestations (uveitis, psoriasis, IBD), disease activity (CRP level, BASDAI [21], ankylosing spondylitis disease activity score-CRP [22]), physical functioning (BASFI) [23], medication use (NSAID, DMARD, biological) and radiographic score on spinal radiographic imaging (modified Stoke Ankylosing Spondylitis Spinal Score) [24], were retrieved from the OASIS 12th year visit. Controls provided information on demographics including age and gender and completed the Self-Report Comorbidity Questionnaires to particularly check for musculoskeletal comorbidity (back pain, OA, arthritis) [25]. Controls reporting arthritis were excluded from further analysis. Next, all subjects were asked to complete a self-composed questionnaire on overall health, well-being and QoL (Questionnaire in the Supplementary data, available at Rheumatology online). First, they were invited to score their health, well-being and QoL during the last week on a horizontal numeric rating scale from 0 (worst) to 10 (best health, well-being, QoL). To avoid an influence of task description across globals, instructions were similarly formulated with a single-sentence item. No reference to a specific disease was made. As an example, the question for well-being was: please tick a box that indicates the effect your health has had on your well-being over the last week. Second, participants were asked to indicate whether they considered dissimilarities between the constructs to be present (yes/no), and if so which pair of constructs differed (health vs well-being; health vs QoL; well-being vs QoL) or if all three represented something different. Next, they were invited to think about each of the three constructs and to write down what they were taking into account when scoring themselves (five lines available per construct). Finally, all subjects were asked to score the three globals a second time after this forced reflection. Statistical analyses Descriptive statistics were used to characterize patients and controls in terms of demographics and scores for health, well-being and QoL. Paired t-tests were performed to assess differences in scores before and after forced reflection for each of the constructs, both within patients and within controls. Independent t-tests were performed to assess the mean differences in scores between patients and controls on each construct. Differences in scores for each pair of construct (health/well-being, health/QoL or well-being/QoL) were plotted against their mean, using Bland and Altman plots, for patients and controls separately [26]. Mixed model analyses were used to test differences in scores between constructs and the role of group membership (patient or control) and forced reflection (before and after reflection), while adjusting for age and gender. For each of the sources of variation of interest, interactions were tested. For each pair of constructs, the proportion of patients and controls indicating that the constructs differed was calculated. All statistical analyses were performed using SPSS 23.0. The answers to the open question, in which the subjects could reflect what the constructs represented to them, were categorized into themes by going back and forth between answers and the list of themes. Each time an answer was given that could not be categorized into one of the themes available so far, a new theme was added. Themes were further linked to International Classification of Functioning, Disability and Health (ICF) components (body functions, activities and participation, environmental factor or personal factor) and next to the most precise ICF categories. A.G. executed the linking process after being familiarized with the ICF and ICF classification by A.B., who had followed an official training in linking at the ICF research branch. For themes that could not be immediately linked to an ICF category, A.G. and A.B. deliberated on the ICF chapter or category that best represented the information in the theme. For personal factors, for which no ICF classification exists as yet, the wording of the theme was maintained [27–29]. For each construct, the average number of themes identified, as well as the proportion of themes representing a specific ICF category, was calculated for patients and controls separately. An ICF category was considered relevant when 10% or more of all themes within the construct related to this ICF category. Results Of the initial 135 patients in the two Dutch centres of the OASIS study that took part in this project, 68 participated in the 12th year survey and all completed the supplementary questionnaire. Forty-seven (69%) were male, mean (s.d.) age was 56.8 (10.9) years and the average symptom duration was 34.9 (11.2) years. Of the 321 questionnaires that were sent out to healthy controls, 84 were returned (27% response). Forty-four were male (52%) with a mean (s.d.) age of 50.4 (13.9) years (Table 1). Fourteen (16.6%) controls reported a comorbidity that might impact daily functioning, but none of them reported arthritis. Characteristics of the participants are shown in Table 1. Table 1 Characteristics of patients and controls Variable Patients Controls Age, mean (s.d.), years 56.8 (10.9) 50.4 (13.9) Male gender, n (%) 47 (69) 44 (52) Symptom duration, mean (s.d.), years 34.9 (11.2) NA HLA-B27 positive, n (%) 57 (85) NA History of IBD, n (%) 15 (25) NA History of psoriasis, n (%) 6 (10) NA History of uveitis, n (%) 22 (36) NA CRP, mean (s.d.), mg/l 8.4 (13.4) NA BASDAI, 0–10, mean (s.d.) 3.7 (1.8) NA ASDAS-CRP, mean (s.d.) 2.3 (0.9) NA BASFI, 0–10, mean (s.d.) 4.6 (2.5) NA mSASSS score, 0–72, mean (s.d.) 27.6 (23.0) NA NSAID use, n (%) 39 (65) NA DMARDs use, n (%) 7 (12) NA Biological use, n (%) 18 (27) NA Variable Patients Controls Age, mean (s.d.), years 56.8 (10.9) 50.4 (13.9) Male gender, n (%) 47 (69) 44 (52) Symptom duration, mean (s.d.), years 34.9 (11.2) NA HLA-B27 positive, n (%) 57 (85) NA History of IBD, n (%) 15 (25) NA History of psoriasis, n (%) 6 (10) NA History of uveitis, n (%) 22 (36) NA CRP, mean (s.d.), mg/l 8.4 (13.4) NA BASDAI, 0–10, mean (s.d.) 3.7 (1.8) NA ASDAS-CRP, mean (s.d.) 2.3 (0.9) NA BASFI, 0–10, mean (s.d.) 4.6 (2.5) NA mSASSS score, 0–72, mean (s.d.) 27.6 (23.0) NA NSAID use, n (%) 39 (65) NA DMARDs use, n (%) 7 (12) NA Biological use, n (%) 18 (27) NA ASDAS: ankylosing spondylitis disease activity score; mSASSS: modified Stoke Ankylosing Spondylitis Spinal Score; NA: not assessed. Table 1 Characteristics of patients and controls Variable Patients Controls Age, mean (s.d.), years 56.8 (10.9) 50.4 (13.9) Male gender, n (%) 47 (69) 44 (52) Symptom duration, mean (s.d.), years 34.9 (11.2) NA HLA-B27 positive, n (%) 57 (85) NA History of IBD, n (%) 15 (25) NA History of psoriasis, n (%) 6 (10) NA History of uveitis, n (%) 22 (36) NA CRP, mean (s.d.), mg/l 8.4 (13.4) NA BASDAI, 0–10, mean (s.d.) 3.7 (1.8) NA ASDAS-CRP, mean (s.d.) 2.3 (0.9) NA BASFI, 0–10, mean (s.d.) 4.6 (2.5) NA mSASSS score, 0–72, mean (s.d.) 27.6 (23.0) NA NSAID use, n (%) 39 (65) NA DMARDs use, n (%) 7 (12) NA Biological use, n (%) 18 (27) NA Variable Patients Controls Age, mean (s.d.), years 56.8 (10.9) 50.4 (13.9) Male gender, n (%) 47 (69) 44 (52) Symptom duration, mean (s.d.), years 34.9 (11.2) NA HLA-B27 positive, n (%) 57 (85) NA History of IBD, n (%) 15 (25) NA History of psoriasis, n (%) 6 (10) NA History of uveitis, n (%) 22 (36) NA CRP, mean (s.d.), mg/l 8.4 (13.4) NA BASDAI, 0–10, mean (s.d.) 3.7 (1.8) NA ASDAS-CRP, mean (s.d.) 2.3 (0.9) NA BASFI, 0–10, mean (s.d.) 4.6 (2.5) NA mSASSS score, 0–72, mean (s.d.) 27.6 (23.0) NA NSAID use, n (%) 39 (65) NA DMARDs use, n (%) 7 (12) NA Biological use, n (%) 18 (27) NA ASDAS: ankylosing spondylitis disease activity score; mSASSS: modified Stoke Ankylosing Spondylitis Spinal Score; NA: not assessed. Table 2 shows the scores on the three constructs for patients and controls. Scores before and after reflection were not significantly different. Patients scored significantly worse on all three globals compared with controls. Mixed modelling adjusted for age and gender indicated that only the scores between patients and controls differed significantly (P < 0.01), but did not reveal any effect attributable to the construct assessed, or to an effect of forced reflection (Table 3). Significant interactions were not found between construct and group, or between the effect of forced reflection and group. The Bland and Altman plots visualize the difference in scores (before reflection) between health/well-being, health/QoL and well-being/QoL against the mean, separately for patients and controls (Fig. 1). In some individuals the score between two constructs showed large dissimilarity, with differences up to five points. However, there seems no pattern in these dissimilarities. Table 2 Scores on the three globals for patients and controls, before and after forced reflection Patients, n = 68 Controls, n = 84 Patients vs controls mean difference; P-valuea Construct Before After P-valueb Before After P-valueb Before After Health 6.1 (2.5) 6.2 (2.4) 0.50 7.6 (1.6) 7.5 (1.5) 0.50 1.4; <0.01 1.3; <0.01 Well-being 6.1 (2.5) 6.1 (2.4) 0.92 7.2 (1.3) 7.4 (1.3) 0.06 1.1; <0.01 1.3; <0.01 Quality of life 6.3 (2.6) 6.4 (2.4) 0.93 7.5 (1.6) 7.5 (1.4) 0.43 1.0; <0.01 1.2; <0.01 Patients, n = 68 Controls, n = 84 Patients vs controls mean difference; P-valuea Construct Before After P-valueb Before After P-valueb Before After Health 6.1 (2.5) 6.2 (2.4) 0.50 7.6 (1.6) 7.5 (1.5) 0.50 1.4; <0.01 1.3; <0.01 Well-being 6.1 (2.5) 6.1 (2.4) 0.92 7.2 (1.3) 7.4 (1.3) 0.06 1.1; <0.01 1.3; <0.01 Quality of life 6.3 (2.6) 6.4 (2.4) 0.93 7.5 (1.6) 7.5 (1.4) 0.43 1.0; <0.01 1.2; <0.01 Mean (s.d.) values are presented. a Rounding differences may be present, independent t-test. b Paired t-test. Table 2 Scores on the three globals for patients and controls, before and after forced reflection Patients, n = 68 Controls, n = 84 Patients vs controls mean difference; P-valuea Construct Before After P-valueb Before After P-valueb Before After Health 6.1 (2.5) 6.2 (2.4) 0.50 7.6 (1.6) 7.5 (1.5) 0.50 1.4; <0.01 1.3; <0.01 Well-being 6.1 (2.5) 6.1 (2.4) 0.92 7.2 (1.3) 7.4 (1.3) 0.06 1.1; <0.01 1.3; <0.01 Quality of life 6.3 (2.6) 6.4 (2.4) 0.93 7.5 (1.6) 7.5 (1.4) 0.43 1.0; <0.01 1.2; <0.01 Patients, n = 68 Controls, n = 84 Patients vs controls mean difference; P-valuea Construct Before After P-valueb Before After P-valueb Before After Health 6.1 (2.5) 6.2 (2.4) 0.50 7.6 (1.6) 7.5 (1.5) 0.50 1.4; <0.01 1.3; <0.01 Well-being 6.1 (2.5) 6.1 (2.4) 0.92 7.2 (1.3) 7.4 (1.3) 0.06 1.1; <0.01 1.3; <0.01 Quality of life 6.3 (2.6) 6.4 (2.4) 0.93 7.5 (1.6) 7.5 (1.4) 0.43 1.0; <0.01 1.2; <0.01 Mean (s.d.) values are presented. a Rounding differences may be present, independent t-test. b Paired t-test. Table 3 Mixed modelling analysis across scores on the three constructs of interest Source of variation Estimate 95% CI P-value Increasing age −0.01 −0.02; 0.00 0.11 Female gender −0.05 −0.32; 0.23 0.74 Patients vs controls −1.28 −1.56; −1.01 <0.01 Construct 0.11 −0.04; 0.27 0.16 Time (before–after reflection) −0.02 −0.28; 0.24 0.89 Intercept 7.73 6.95; 8.51 <0.01 Source of variation Estimate 95% CI P-value Increasing age −0.01 −0.02; 0.00 0.11 Female gender −0.05 −0.32; 0.23 0.74 Patients vs controls −1.28 −1.56; −1.01 <0.01 Construct 0.11 −0.04; 0.27 0.16 Time (before–after reflection) −0.02 −0.28; 0.24 0.89 Intercept 7.73 6.95; 8.51 <0.01 Table 3 Mixed modelling analysis across scores on the three constructs of interest Source of variation Estimate 95% CI P-value Increasing age −0.01 −0.02; 0.00 0.11 Female gender −0.05 −0.32; 0.23 0.74 Patients vs controls −1.28 −1.56; −1.01 <0.01 Construct 0.11 −0.04; 0.27 0.16 Time (before–after reflection) −0.02 −0.28; 0.24 0.89 Intercept 7.73 6.95; 8.51 <0.01 Source of variation Estimate 95% CI P-value Increasing age −0.01 −0.02; 0.00 0.11 Female gender −0.05 −0.32; 0.23 0.74 Patients vs controls −1.28 −1.56; −1.01 <0.01 Construct 0.11 −0.04; 0.27 0.16 Time (before–after reflection) −0.02 −0.28; 0.24 0.89 Intercept 7.73 6.95; 8.51 <0.01 Fig. 1 View largeDownload slide Bland and Altman plots for each pair of constructs The Bland and Altman plots visualize the difference against the mean in scores for each pair of constructs, for patients (A–C) and controls (D–F) separately. Fig. 1 View largeDownload slide Bland and Altman plots for each pair of constructs The Bland and Altman plots visualize the difference against the mean in scores for each pair of constructs, for patients (A–C) and controls (D–F) separately. Forty-four (64%) patients and 31 (36%) controls indicated there was no dissimilarity between one or more pair of constructs. Patients most frequently considered health to be different from either well-being or QoL (n = 12 and n = 8, respectively). Two patients (8%) and 32 controls (60%) considered that all three globals represented other constructs. Themes when reflecting on, respectively, health, well-being and QoL were provided by 50 (74%), 39 (57%) and 38 (56%) of the patients and 73 (87%), 64 (76%) and 69 (82%) of the controls (Table 4). The average number of themes provided per construct was 2.5, 1.8 and 1.6 by the patients, and 3.6, 3.7 and 3.4 by the controls for health, well-being and QoL, respectively. Table 4 shows for each construct, the ICF category names representing the themes and the proportion of the total identified themes per ICF category, separately for patients and controls. Patients with AS associated health in particular with body functions, more specifically with pain, limitations in movement functions and energy level. Controls also associated health with (absence of) pain and (absence of) limitations in movement. In addition, some impairments not mentioned by patients were brought forward by controls, mainly seeing and hearing functions and other emotional functions. Well-being in patients was again associated with pain, movement functions and energy level, but also with depressed feelings and activities related to mobility (cycling, walking and running). Controls considered more frequently other emotions and social life, and (although just below our 10% threshold) also attitudes of colleagues and leisure as part of well-being. For QoL, patients identified again energy and limitations in movement function as relevant aspects, and in addition, activities related to mobility and social life. For controls, QoL was more frequently related to social life and leisure activities, but also to economic self-sufficiency (just below our 10% threshold). Interestingly, patients but not controls identified to some extent personal factors, among which coping styles, to be relevant facilitators for health (dealing with limitations, adapting to pain, adjusting to energy) and QoL (accepting). Table 4 Themes organised according to ICF categories, separately for patients and controls ICF classification ICF category name Patients (n = 68) Controls (n = 84) Health, n = 50 (74%) Well-being, n = 39 (57%) QoL, n = 38 (56%) Health, n = 73 (87%) Well-being, n = 64 (76%) QoL, n = 69 (82%) n = 124 themes n = 70 themes n = 61 themes n = 262 themes n = 235 themes n = 233 themes Body functions and mental functions Pain (b280) 14.5 15.7 8.2 11.4 2.1 5.5 Mobility (movement functions) 21.8 14.3 14.8 21.0 2.7 4.0 Energy level (b1300) 12.1 12.9 9.8 8.5 1.4 Sleep functions (b134) 9.7 8.6 3.3 3.4 3.4 0.9 Other body complaints (e.g. seeing, hearing) 6.5a 2.9 1.6 12.5 4.1 3.2 Anxiety (b152) 1.1 0.7 2.4 Depressed feelings/emotional functions (b152) 4.0 12.9 6.6 Stress (b152) 2.8 6.2 3.2 Other emotional functions (b1528) 1.6 8.9 1.6 13.1 17.1 7.2 Activities and participation Social life (d7) 4.0 5.7 11.5 0.6 22.6 23.7 Recreation and leisure (d920) 5.6 6.6 2.3 7.5 11.2 Education and work (d8) 0.8 1.6 2.8 3.4 3.2 Household tasks (d6) 4.0 3.3 0.6 1.4 0.9 Mobility (moving around) (d5) 3.2 12.9 11.5 4.0 5.5 4.0 Environmental factors Attitudes of colleagues (e325) 7.5 4.9 Financial self-sufficiency (e165) 0.8 1.4 6.6 2.7 9.0 Natural environment (e2) 2.8 5.5 4.9 Products and technologies (e1) 1.1 0.7 0.9 Food (e110) 9.7 4.1 5.5 Health services (e580) 1.7 Daily necessities (e1) 1.4 3.2 Personal factors Coping styles (pf) 6.5 2.9 8.2 Others 4.8a 1.4 1.6 2.8 1.8 ICF classification ICF category name Patients (n = 68) Controls (n = 84) Health, n = 50 (74%) Well-being, n = 39 (57%) QoL, n = 38 (56%) Health, n = 73 (87%) Well-being, n = 64 (76%) QoL, n = 69 (82%) n = 124 themes n = 70 themes n = 61 themes n = 262 themes n = 235 themes n = 233 themes Body functions and mental functions Pain (b280) 14.5 15.7 8.2 11.4 2.1 5.5 Mobility (movement functions) 21.8 14.3 14.8 21.0 2.7 4.0 Energy level (b1300) 12.1 12.9 9.8 8.5 1.4 Sleep functions (b134) 9.7 8.6 3.3 3.4 3.4 0.9 Other body complaints (e.g. seeing, hearing) 6.5a 2.9 1.6 12.5 4.1 3.2 Anxiety (b152) 1.1 0.7 2.4 Depressed feelings/emotional functions (b152) 4.0 12.9 6.6 Stress (b152) 2.8 6.2 3.2 Other emotional functions (b1528) 1.6 8.9 1.6 13.1 17.1 7.2 Activities and participation Social life (d7) 4.0 5.7 11.5 0.6 22.6 23.7 Recreation and leisure (d920) 5.6 6.6 2.3 7.5 11.2 Education and work (d8) 0.8 1.6 2.8 3.4 3.2 Household tasks (d6) 4.0 3.3 0.6 1.4 0.9 Mobility (moving around) (d5) 3.2 12.9 11.5 4.0 5.5 4.0 Environmental factors Attitudes of colleagues (e325) 7.5 4.9 Financial self-sufficiency (e165) 0.8 1.4 6.6 2.7 9.0 Natural environment (e2) 2.8 5.5 4.9 Products and technologies (e1) 1.1 0.7 0.9 Food (e110) 9.7 4.1 5.5 Health services (e580) 1.7 Daily necessities (e1) 1.4 3.2 Personal factors Coping styles (pf) 6.5 2.9 8.2 Others 4.8a 1.4 1.6 2.8 1.8 The proportion with which a theme was indicated within a construct is presented and sums up to 100% per construct (rounding differences may be present). aSome are related to side effects of treatment. QoL: quality of life; ICF: International Classification of Functioning and Health; pf: personal factor. Table 4 Themes organised according to ICF categories, separately for patients and controls ICF classification ICF category name Patients (n = 68) Controls (n = 84) Health, n = 50 (74%) Well-being, n = 39 (57%) QoL, n = 38 (56%) Health, n = 73 (87%) Well-being, n = 64 (76%) QoL, n = 69 (82%) n = 124 themes n = 70 themes n = 61 themes n = 262 themes n = 235 themes n = 233 themes Body functions and mental functions Pain (b280) 14.5 15.7 8.2 11.4 2.1 5.5 Mobility (movement functions) 21.8 14.3 14.8 21.0 2.7 4.0 Energy level (b1300) 12.1 12.9 9.8 8.5 1.4 Sleep functions (b134) 9.7 8.6 3.3 3.4 3.4 0.9 Other body complaints (e.g. seeing, hearing) 6.5a 2.9 1.6 12.5 4.1 3.2 Anxiety (b152) 1.1 0.7 2.4 Depressed feelings/emotional functions (b152) 4.0 12.9 6.6 Stress (b152) 2.8 6.2 3.2 Other emotional functions (b1528) 1.6 8.9 1.6 13.1 17.1 7.2 Activities and participation Social life (d7) 4.0 5.7 11.5 0.6 22.6 23.7 Recreation and leisure (d920) 5.6 6.6 2.3 7.5 11.2 Education and work (d8) 0.8 1.6 2.8 3.4 3.2 Household tasks (d6) 4.0 3.3 0.6 1.4 0.9 Mobility (moving around) (d5) 3.2 12.9 11.5 4.0 5.5 4.0 Environmental factors Attitudes of colleagues (e325) 7.5 4.9 Financial self-sufficiency (e165) 0.8 1.4 6.6 2.7 9.0 Natural environment (e2) 2.8 5.5 4.9 Products and technologies (e1) 1.1 0.7 0.9 Food (e110) 9.7 4.1 5.5 Health services (e580) 1.7 Daily necessities (e1) 1.4 3.2 Personal factors Coping styles (pf) 6.5 2.9 8.2 Others 4.8a 1.4 1.6 2.8 1.8 ICF classification ICF category name Patients (n = 68) Controls (n = 84) Health, n = 50 (74%) Well-being, n = 39 (57%) QoL, n = 38 (56%) Health, n = 73 (87%) Well-being, n = 64 (76%) QoL, n = 69 (82%) n = 124 themes n = 70 themes n = 61 themes n = 262 themes n = 235 themes n = 233 themes Body functions and mental functions Pain (b280) 14.5 15.7 8.2 11.4 2.1 5.5 Mobility (movement functions) 21.8 14.3 14.8 21.0 2.7 4.0 Energy level (b1300) 12.1 12.9 9.8 8.5 1.4 Sleep functions (b134) 9.7 8.6 3.3 3.4 3.4 0.9 Other body complaints (e.g. seeing, hearing) 6.5a 2.9 1.6 12.5 4.1 3.2 Anxiety (b152) 1.1 0.7 2.4 Depressed feelings/emotional functions (b152) 4.0 12.9 6.6 Stress (b152) 2.8 6.2 3.2 Other emotional functions (b1528) 1.6 8.9 1.6 13.1 17.1 7.2 Activities and participation Social life (d7) 4.0 5.7 11.5 0.6 22.6 23.7 Recreation and leisure (d920) 5.6 6.6 2.3 7.5 11.2 Education and work (d8) 0.8 1.6 2.8 3.4 3.2 Household tasks (d6) 4.0 3.3 0.6 1.4 0.9 Mobility (moving around) (d5) 3.2 12.9 11.5 4.0 5.5 4.0 Environmental factors Attitudes of colleagues (e325) 7.5 4.9 Financial self-sufficiency (e165) 0.8 1.4 6.6 2.7 9.0 Natural environment (e2) 2.8 5.5 4.9 Products and technologies (e1) 1.1 0.7 0.9 Food (e110) 9.7 4.1 5.5 Health services (e580) 1.7 Daily necessities (e1) 1.4 3.2 Personal factors Coping styles (pf) 6.5 2.9 8.2 Others 4.8a 1.4 1.6 2.8 1.8 The proportion with which a theme was indicated within a construct is presented and sums up to 100% per construct (rounding differences may be present). aSome are related to side effects of treatment. QoL: quality of life; ICF: International Classification of Functioning and Health; pf: personal factor. Discussion In the present study, 65% of the patients and 37% of controls indicated that health, well-being and QoL mean similar things to them, and this is also reflected in there being no meaningful differences in the absolute scores on the patient globals for these constructs. Moreover, the scores on the different globals did not systematically change after reflecting on the content of the constructs, suggesting that the information subjects use when scoring these constructs is implicitly accounted for. While there were some similarities in themes that patients and controls associated with the constructs, there were also clear dissimilarities in the type and frequency of aspects considered relevant when scoring themselves for each of the three constructs. On average, patients took into account fewer aspects per construct compared with controls. Across constructs however, patients identified more aspects when rating overall health than when rating the other two constructs, while controls tended to identify a similar number of different themes for each of the three constructs. Although survey fatigue cannot be entirely ruled out, as patients completed this questionnaire in the same period as the regular OASIS follow-up questionnaire, it seems more plausible that patients—due to the presence of a chronic disease—have on average a more narrow biomedical focus when evaluating health, whereas controls have a broader vision, likely also taking into account potential diseases that are not (yet) present. On the same line, patients associated disease-related aspects, in particular pain, fatigue, movement functions and mobility, to all three constructs. For controls, impairments in body and mental functions were an essential part of experienced health, but less for well-being and QoL. For both patients and controls, emotional functions were frequently identified within the construct well-being. Controls also considered social life when evaluating well-being, and economic self-sufficiency when evaluating QoL. This was less clear for patients. It is of note that in our study, coping or beliefs were identified by patients as relevant factors when evaluating their own overall health and QoL, whereas controls never considered personal factors in relation to the constructs. The current data suggest that patients implicitly consider their level of adaptation and coping when scoring globals. The qualitative findings of our study are partly in accordance with a study from Fagerlind et al. [30], who reported on the constructs health and QoL from semi-structured interviews among 22 patients with RA. They emphasized that the understanding of the constructs are both individual and heterogeneous. In line with our findings, participants recognized overlap between the constructs, and identified impairments in body functions and limitations in activities as key parts of both constructs. While in the interviews in the Fagerlind study well-being was not included as a separate construct, their patients spontaneously mentioned well-being as (an exclusive) part of the construct health. Living conditions, social network but also a positive look on life were exclusively related to QoL. In contrast, we found that half of the patients who reported dissimilarities between the globals indicated that health differed from well-being. While not being the primary focus of the study, it was noted that patients with AS scored health, well-being and QoL only slightly lower than healthy subjects (on average 1.0–1.4 lower on a 0–10 scale). Despite the fact that no formal thresholds are available to decide whether these between-group differences are clinically relevant, but keeping in mind that differences of 1.5 are often considered clinically relevant at the individual patient level [31], we had expected larger differences between patients and controls. An attractive explanation might be that reference shift plays an important role here: that patients adapt to their disease and adjust their sense of normal [18]. Our qualitative study seems to suggest that personal resources (coping, adapting) may play an even more important role than to social context when rating overall health, well-being and QoL. Our study has clear limitations. First, although face-to-face interviews or cognitive debriefings might be a preferred qualitative approach, such methods are limited in the number of subjects that can be included. The large amount of qualitative data in our study allowed semi-quantitative analyses. Second, we have chosen a limited number of constructs that might reflect the life impact of disease. We could have added health-related QoL, happiness, satisfaction and likely other constructs. However, it was not our aim to be exhaustive on that line. We wished to explore commonly used constructs. Third, our study design did not account for a possible order effect, as we did not change the order of the globals in subsets of the survey. It cannot be excluded that the score on well-being, which was the first question, influenced the scores on the other constructs. Fourth, the question on generalizability should be addressed. This study was done in patients with AS with longstanding disease. While we think that the main conclusions will hold for other rheumatic diseases, it would be interesting to understand the change over the course of the disease in scores and underlying themes that drive these scores compared with controls. Fifth, due to low response in the control population we were not able to match for age and gender. While scores have been adjusted statistically for such differences, this was not feasible for the qualitative part of the study. Furthermore, selection bias might have occurred in the control population. We cannot exclude that this might have influenced the results. In summary, while patients and controls recognized that health, well-being and QoL were different constructs, this was not reflected in differences in the mean scores on the globals. For patients, themes related to physical impairments and limitations were relevant for health, well-being and QoL, whereas controls related these themes to health only. For patients and controls, emotional functions were specifically related to well-being. For controls, social life and economic self-sufficiency were important for QoL. The final score for any global measure seems to be the result of an implicit but complex process, in which themes are weighted and appraised in a varying ways. Acknowledgement Conception and design: A.B.; data acquisition: A.G. and A.B.; data analysis and interpretation of data: all authors; drafting the work or revising it critically for important intellectual content: all authors; final approval: all authors. Funding: No specific funding was received from any bodies in the public, commercial or not-for-profit sectors to carry out the work described in this manuscript. Disclosure statement: The authors have declared no conflicts of interest. Supplementary data Supplementary data are available at Rheumatology online. References 1 Kilic L , Erden A , Bingham CO 3rd , Gossec L , Kalyoncu U. The reporting of patient-reported outcomes in studies of patients with rheumatoid arthritis: a systematic review of 250 articles . J Rheumatol 2016 ; 43 : 1300 – 5 . Google Scholar CrossRef Search ADS PubMed 2 Harrison MJ , Boonen A , Tugwell P , Symmons DP. Same question, different answers: a comparison of global health assessments using visual analogue scales . Qual Life Res 2009 ; 18 : 1285 – 92 . Google Scholar CrossRef Search ADS PubMed 3 Krabbe PF , Essink-Bot ML , Bonsel GJ. The comparability and reliability of five health-state valuation methods . Soc Sci Med 1997 ; 45 : 1641 – 52 . Google Scholar CrossRef Search ADS PubMed 4 Spoorenberg A , van Tubergen A , Landewe R et al. Measuring disease activity in ankylosing spondylitis: patient and physician have different perspectives . Rheumatology (Oxford) 2005 ; 44 : 789 – 95 . Google Scholar CrossRef Search ADS PubMed 5 van Tubergen A , Coenen J , Landewé R et al. Assessment of fatigue in patients with ankylosing spondylitis: a psychometric analysis . Arthritis Rheum 2002 ; 47 : 8 – 16 . Google Scholar CrossRef Search ADS PubMed 6 Escorpizo R , Bombardier C , Boonen A et al. Worker productivity outcome measures in arthritis . J Rheumatol 2007 ; 34 : 1372 – 80 . Google Scholar PubMed 7 Fowler JH , Christakis NA. Dynamic spread of happiness in a large social network: longitudinal analysis over 20 years in the Framingham Heart Study . BMJ 2008 ; 337 : a2338 . Google Scholar CrossRef Search ADS PubMed 8 Naci H , Ioannidis JPA. Evaluation of wellness determinants and interventions by citizen scientists . JAMA 2015 ; 314 : 121 – 2 . Google Scholar CrossRef Search ADS PubMed 9 EuroQol Group. EuroQol–a new facility for the measurement of health-related quality of life . Health Policy 1990 ; 16 : 199 – 208 . CrossRef Search ADS PubMed 10 Jones SD , Steiner A , Garrett SL , Calin A. The Bath Ankylosing Spondylitis Patient Global Score (BAS-G) . Br J Rheumatol 1996 ; 35 : 66 – 71 . Google Scholar CrossRef Search ADS PubMed 11 Aaronson NK , Muller M , Cohen PD et al. Translation, validation, and norming of the Dutch language version of the SF-36 Health Survey in community and chronic disease populations . J Clin Epidemiol 1998 ; 51 : 1055 – 68 . Google Scholar CrossRef Search ADS PubMed 12 Doward LC , Spoorenberg A , Cook SA et al. Development of the ASQoL: a quality of life instrument specific to ankylosing spondylitis . Ann Rheum Dis 2003 ; 62 : 20 – 6 . Google Scholar CrossRef Search ADS PubMed 13 WHO . Constitution of the World Health Organization: Principles 1948. http://apps.who.int/gb/bd/PDF/bd47/EN/constitution-en.pdf? ua=1 (16 March 2018, date last accessed). 14 Diener E. Guidelines for national indicators of subjective well-being and ill-being . J Happiness Stud 2006 ; 7 : 397 – 404 . Google Scholar CrossRef Search ADS 15 Felce D , Perry J. Quality of life: its definition and measurement . Res Dev Disabil 1995 ; 16 : 51 – 74 . Google Scholar CrossRef Search ADS PubMed 16 Eurostat . Quality of Life Indicators - Measuring Quality of Life 2015 [updated 05-11-2015]. http://ec.europa.eu/eurostat/statistics-explained/index.php/Quality_of_life_indicators_-_measuring_quality_of_life (16 March 2018, date last accessed). 17 World Health Organization. Division of Mental Health and Prevention of Substance Abuse . ( 1997 ). WHOQOL : measuring quality of life. Geneva : World Health Organization. http://www.who.int/iris/handle/10665/63482 (7 May 2018, date last accessed). 18 Rapkin BD , Schwartz CE. Toward a theoretical model of quality-of-life appraisal: implications of findings from studies of response shift . Health Qual Life Outcomes 2004 ; 2 : 14 . Google Scholar CrossRef Search ADS PubMed 19 Spoorenberg A , van der Heijde D , de Klerk E et al. Relative value of erythrocyte sedimentation rate and C-reactive protein in assessment of disease activity in ankylosing spondylitis . J Rheumatol 1999 ; 26 : 980 – 4 . Google Scholar PubMed 20 van der Linden S , Valkenburg HA , Cats A. Evaluation of diagnostic criteria for ankylosing spondylitis. A proposal for modification of the New York criteria . Arthritis Rheum 1984 ; 27 : 361 – 8 . Google Scholar CrossRef Search ADS PubMed 21 Garrett S , Jenkinson T , Kennedy LG et al. A new approach to defining disease status in ankylosing spondylitis: the Bath Ankylosing Spondylitis Disease Activity Index . J Rheumatol 1994 ; 21 : 2286 – 91 . Google Scholar PubMed 22 Lukas C , Landewe R , Sieper J et al. Development of an ASAS-endorsed disease activity score (ASDAS) in patients with ankylosing spondylitis . Ann Rheum Dis 2009 ; 68 : 18 – 24 . Google Scholar CrossRef Search ADS PubMed 23 Calin A , Garrett S , Whitelock H et al. A new approach to defining functional ability in ankylosing spondylitis: the development of the Bath Ankylosing Spondylitis Functional Index . J Rheumatol 1994 ; 21 : 2281 – 5 . Google Scholar PubMed 24 Creemers MC , Franssen MJ , van’t Hof MA et al. Assessment of outcome in ankylosing spondylitis: an extended radiographic scoring system . Ann Rheum Dis 2005 ; 64 : 127 – 9 . Google Scholar CrossRef Search ADS PubMed 25 Sangha O , Stucki G , Liang MH , Fossel AH , Katz JN. The Self-Administered Comorbidity Questionnaire: a new method to assess comorbidity for clinical and health services research . Arthritis Rheum 2003 ; 49 : 156 – 63 . Google Scholar CrossRef Search ADS PubMed 26 Altman DG , Bland J. Measurement in medicine: the analysis of method comparison studies . J R Stat Soc Ser D (The Statistician) 1983 ; 32 : 307 – 17 . 27 World Health Organisation . International Classification of Functioning, Disability and Health: ICF. Geneva: WHO 2001 . 28 Cieza A , Brockow T , Ewert T et al. Linking health-status measurements to the international classification of functioning, disability and health . J Rehabil Med 2002 ; 34 : 205 – 10 . Google Scholar CrossRef Search ADS PubMed 29 Cieza A , Geyh S , Chatterji S et al. ICF linking rules: an update based on lessons learned . J Rehabil Med 2005 ; 37 : 212 – 8 . Google Scholar CrossRef Search ADS PubMed 30 Fagerlind H , Ring L , Brülde B , Feltelius N , Lindblad AK. Patients’ understanding of the concepts of health and quality of life . Patient Educ Couns 2010 ; 78 : 104 – 10 . Google Scholar CrossRef Search ADS PubMed 31 Strand V , Boers M , Idzerda L et al. It’s good to feel better but it’s better to feel good and even better to feel good as soon as possible for as long as possible. Response criteria and the importance of change at OMERACT 10 . J Rheumatol 2011 ; 38 : 1720 – 7 . Google Scholar CrossRef Search ADS PubMed © The Author(s) 2018. Published by Oxford University Press on behalf of the British Society for Rheumatology. All rights reserved. For permissions, please email: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

RheumatologyOxford University Press

Published: May 19, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off