1. Top of page
  2. Abstract


To develop and validate a disease-specific quality of life (QOL) measure for osteoarthritis (OA), the OAQoL, using the needs-based conceptual model.


In the first phase of this study, in-depth, semistructured interviews were conducted with 44 OA patients to explore the issues associated with impact of OA and to derive items for a draft OAQoL questionnaire. In phase 2, 17 OA patients were interviewed on the relevance, clarity, and ease of completion of the measure in structured interviews. In phase 3, the draft questionnaire was mailed to 635 patients to test the psychometric properties of the questionnaire using Rasch analysis. Test–retest assessment of the revised questionnaire was performed in phase 4 by mailing the questionnaire to an additional 201 participants, with a second questionnaire repeated 2 weeks later.


A 38-item draft measure was devised during phase 1 and mailed in phase 2. Rasch analysis of the draft questionnaire (n = 259) indicated initial misfit, which was rectified with the removal of 13 problematic items (χ2[75] = 83.602, P = 0.232). For the test–retest assessment (n = 60), 3 additional items were removed, leaving a 22-item OAQoL that demonstrated good fit to the Rasch model (χ2[44] = 44.559, P = 0.533) with excellent test–retest correlation (ρ = 0.925, P < 0.001; z = −0.06, P = 0.995).


The OAQoL is a simple and easy to use 22-item unidimensional questionnaire developed specifically to assess the impact of OA on QOL. The measure has been developed as a true patient-based questionnaire and demonstrates good psychometric properties, including test–retest reliability.


  1. Top of page
  2. Abstract

Osteoarthritis (OA) is one of the 10 most disabling diseases in developed countries (1) and results in a considerable impact on the individual. Patients with OA report significant pain and functional limitations (2, 3) and are more likely to perceive themselves as mentally and physically unhealthy (4). Although most research in OA has focused on pain and physical disability, there has been a recent and growing interest in measuring the real-world impact of OA on patients through formal assessment of quality of life (QOL). The World Health Organization, the International League for Rheumatology Task Force (5), and the OA Research Society strongly recommended that QOL measures be used in OA clinical research (6).

Although QOL has been identified as integral to capturing information that is important to the patient, significant flaws have been identified in the conceptual basis of existing outcomes (7). There has been considerable debate over use of the term “quality of life.” Nord et al (8) suggested that QOL is a subjective, overall feeling of well-being. However, there appears to be a necessity in medicine for clinicians to describe QOL in terms of the absence or presence of disease and its consequences. This is often referred to as health-related QOL, and instruments such as the Short Form 36 and EuroQol measure this construct. QOL is a much broader concept than one that conceives of a good QOL as life free from disability; QOL represents a holistic concept and goes beyond the activities of daily living and disease categories. QOL encompasses the social, psychological, and spiritual well-being of the person and how these aspects interact with the person's environment (9). Under this model, health is not seen as an inherent or even necessary component of QOL, but as a potential influence (10, 11).

Although several tools have been used to capture information on QOL in OA studies, 3 fundamental criticisms of currently used measures can be made. First, outcome measures commonly used in OA research such as the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) (12) and the Lequesne Index (13) and new tools such as the Cochin Index (14) are instruments that focus on QOL in terms of pain and functional impairment only. The Osteoarthritis Knee and Hip Quality of Life (OAKHQOL) (15), an instrument developed for the knee and hip, was developed based on the International Classification of Functioning, Disability and Health (16) definition of health and consequently measures health-related QOL. A further issue is that because these instruments focus on hip and knee OA, they may not be relevant to other forms of OA or multiple-site OA. Second, there is concern that outcome measures are too often derived from what clinicians, rather than patients, deem to be important. The fundamental flaw in such an approach is that clinicians are more likely to catastrophize disability (17) and ignore the socioeconomic and psychosocial issues (18), and are simply not good at predicting what patients consider to be important (19, 20). Indeed, it has been argued that developing and validating outcome tools that are devised by clinicians without the inclusion of patient needs may be inappropriate and is likely to compromise the usefulness, validity, and accuracy of the tool (21). Finally, generic, non–disease-specific instruments, including the Medical Outcomes Study 36-Item Short Form (22), the EuroQol (23), the General Well-Being Index (GWBI) (24), and the Assessment of Quality of Life (25), have been used in OA studies. Although these tools have provided an opportunity to compare diagnostic groups, generic measures lack the sensitivity to discriminate change across time (26). Disease-specific measures have been shown to be more sensitive to change (27) and have been found to predict clinical changes better than generic measures (28).

The needs-based approach (29) was developed to devise QOL tools based on the holistic approach to outcome, rather than health-related QOL. This method is based on a validated development technique involving in-depth qualitative interviews with relevant patients. The focus of this approach is to derive an outcome measure based on issues that are important to the person with the disease and that are not defined by the medical disability model, which is the focus of health-related QOL instruments. This approach has been used to develop several disease-specific QOL tools for dermatologic, neurologic (30), and rheumatologic diseases (31–33). Using this model, the needs relevant to each condition are identified, maximizing the content validity and responsiveness of the final instruments. Interestingly, while OA is the most prevalent of the rheumatic diseases, currently there is no disease-specific QOL instrument available for this condition.

The aim of the present study was to develop and assess the psychometric properties of the OAQoL, a disease-specific, patient-based QOL measure.


  1. Top of page
  2. Abstract

The methodology used in the development and validation of the OAQoL is a standard method used in the development of needs-based QOL instruments (34, 35). It involves 4 phases: in-depth interviews, cognitive debriefing, initial psychometric testing, and test–retest assessment (Figure 1).

thumbnail image

Figure 1. Summary of the project phases. OA = osteoarthritis; OAQoL = Osteoarthritis Quality of Life questionnaire.

Download figure to PowerPoint

Phase 1: interviews.

To explore issues associated with living with OA, in-depth, semistructured interviews were conducted with 44 participants from the primary care Leeds Musculoskeletal Service and the secondary care Leeds National Health Service Trust rheumatology and orthopedics clinics. Patients with OA attending these clinics were invited to participate. To ensure a representative sample, a matrix was constructed with forced representation for sex, age (≤55 years and ≥56 years), and site of OA (hip, knee, hand, foot, and multiple sites) with approximately equal representation sought for each group. All participants with hip, hand, and knee OA fulfilled the American College of Rheumatology criteria for the diagnosis of OA (36–38). In the absence of any such criteria for patients with OA of the foot, participants were included if they had symptomatic, clinically diagnosed OA that was confirmed by radiographic evidence. Participants with significant comorbidity were excluded from this phase of the study.

The goal of the interviews was to derive items to be used in the QOL instrument based on direct quotes from persons with OA. The interviews were undertaken by 8 researchers, including psychologists and allied health professionals, who were experienced qualitative interviewers. The interviews were conducted either at the participant's home or at a location of the participant's choice and took the form of an informal, focused conversation (39). To initiate the interview, participants were asked a general question about their arthritis, such as, “Tell me how your arthritis has an impact on your day to day life,” and were encouraged to discuss any aspect of their lives. Interviewers were required to probe in depth any issues raised by those interviewed. For example, where a respondent raised an aspect of functioning as being problematic, they were then asked to state how they were affected by the functional disability. In this way the interview went beyond determining the impact of OA on symptoms and functioning (health-related QOL) by determining how this affected need fulfillment and determining participants' emotional response to the restrictions. While free conversation was encouraged, if the participants were unable to think of impact on their lives, interviewers asked prompt questions based on social activities, mood, feelings about the future, and relationships with others. Interviews were audio recorded with the permission of the interviewed participants and transcribed verbatim for data exploration and analysis.

Following the transcription of each tape, the interview was checked by the interviewer for accuracy and clarity and then cross-checked by one of the other researchers. Each transcript was coded by 2 of the 8 researchers to identify statements that related to the impact of OA on the respondents' needs. Actual quotations from the interviewed participants were used to form potential items for the measure where possible. The initial list of items was then reviewed by 6 of the 8 researchers, 4 of whom had experience in item identification using the needs-based model. Duplicated, idiosyncratic, or sex-based items were removed at this stage. The remaining items formed the basis of a draft questionnaire (OAQoL version 1).

Phase 2: cognitive debriefing.

The draft OAQoL (version 1) was field tested with relevant OA patients in order to test the applicability, relevance, comprehensibility, and completeness of the draft questionnaire. A different group of patients who were attending OA clinics at the Leeds Musculoskeletal Service clinic and who fulfilled the OA diagnostic criteria outlined in phase 1 were invited to participate. Respondents' general comments and actions during the completion were noted by the interviewer. On completion of the questionnaire, participants were then interviewed and asked general questions about the relevance, clarity, and ease of completion of the questionnaire. Next, participants were asked about any items with which they appeared to have difficulty. Finally, each participant was asked for comments on specific items that the research team had identified as potentially problematic.

Phase 3: scaling properties and construct validity.

To evaluate the scaling properties and construct validity of the draft questionnaire, a postal survey was sent to 635 patients from primary or secondary care. A review of the medical records of patients who had recently attended an OA clinic was undertaken and those who fulfilled the OA diagnostic criteria outlined in phase 1 were sent an invitation to participate. Patients who did not respond were sent 2 reminder letters, after which they were deemed unwilling to participate in the study. The questionnaire pack included demographic questions, the draft OAQoL, and the following outcome measures: the WOMAC (to assess pain and functional ability for lower limb OA [12]), the Cochin Scale (to assess pain and functional ability for hand OA [14]), and the GWBI (a generic instrument that measures perceived well-being [24, 40]). Scaling properties (whether the data are interval in nature) and internal construct validity (the assessment of OAQoL data compared with a theoretical conceptual model) were assessed using Rasch analysis (41) with data entered using SPSS software, version 14 (SPSS, Chicago, IL) and analyzed using RUMM2020 (42).

Rasch analysis is a probabilistic mathematical modeling technique used to assess properties of outcome measures and is the current standard for the development of metric quality outcomes in health care (43). Data collected from ordinal questionnaires or scales that are intended to be summated into an overall score are tested against the expectations of this measurement model. Rasch analysis has been widely used in the development and validation of a number of outcome measures (31–33, 44, 45).

Using Rasch analysis, the properties of outcome measures are assessed and include the following: 1) invariance of the items across the scale (represented by a nonsignificant fit statistic [chi-square]); 2) item difficulty, or the hierarchy of items (refers to items in a questionnaire covering a range of less extreme and extreme items; represented by a spread of logit values across all items); 3) residual fit statistics, which are the differences between the observed data and what is expected by the model for each person and item (a perfect fit is represented by a mean ± SD of 0 ± 1 [46]); 4) principal components analysis of the fit residuals, which identifies possible patterns of the residuals measuring an underlying construct once the Rasch factor has been extracted, and is the principal mechanism for confirming unidimensionality (determined by evaluating the number of independent t-tests comparing possible patterns in the residuals, which should be <5% [47]); and 5) the person separation index (PSI), or the extent to which items distinguish between distinct levels of functioning (where 0.7 is considered a minimal value for group use; 0.85 for individual patient use [48]). Finally, differential item functioning (DIF) is a two-way analysis of variance of the residuals (49), which represents the stability of the instrument, irrespective of the group being evaluated. While groups may be expected to have different QOL (for instance, women may have a slightly different OAQoL score than men, or older persons may have lower scores than younger respondents), their group membership at any given level of the trait should not influence how they score.

External construct validity, or how the scale performs relative to other measures, was assessed by relating scores on the OAQoL to those on the WOMAC, Cochin Scale, and GWBI. It was predicted that there would be moderate associations between the OAQoL and these scales, indicating that they assess related but different outcome constructs. Relationships between the instruments were undertaken using Spearman's rho and data were analyzed using SPSS software, version 14.

Phase 4: test–retest reliability.

The revised OAQoL was sent to 201 patients from the primary and secondary health services. Using the same method as described for phase 3, a further cohort of recent patients with OA was sent an invitation to participate. Participants who responded were sent another questionnaire 2 weeks later and the test–retest reliability of the instrument was assessed using Rasch analysis, Spearman's rho, and Cohen's kappa.

The research was conducted in compliance with the Declaration of Helsinki with institutional review board and ethical approval granted by the Leeds West Ethics Review Board.


  1. Top of page
  2. Abstract

Participant characteristics for each phase of the study are shown in Table 1.

Table 1. Participant characteristics for each phase of the study*
 Phase 1: qualitative interviews (n = 44)Phase 2: cognitive debriefing (n = 17)Phase 3: psychometric testing (n = 259)Phase 4: test–retest (n = 60)
  • *

    Values are the number (percentage) unless otherwise indicated. IQR = interquartile range; OA = osteoarthritis; NA = not applicable; GCSE = General Certificate of Secondary Education; O level = ordinary level examinations; A level = advance level examinations.

 Male19 (43.2)3 (17.6)72 (27.8)27 (45.0)
 Female25 (56.8)14 (82.4)178 (68.7)33 (55.0)
 Missing009 (3.5)0
Age, years    
 Mean ± SD58.7 ± 15.069.2 ± 10.866.5 ± 12.566.5 ± 12.0
 Median (IQR)64 (53–69)72 (62–77.5)68 (59–76)67 (58–67)
Site of OA    
 Hip11 (25.0)2 (11.8)62 (23.9)24 (40.0)
 Knee15 (34.1)10 (58.8)104 (40.2)51 (84.0)
 Foot5 (11.4)3 (17.7)74 (28.6)14 (23.3)
 Hand5 (11.4)4 (23.6)104 (40.2)21 (35.0)
 Multiple sites8 (18.2)6 (35.4)221 (85.3)50 (83.3)
 Median joint pain, no.NANA44
Education level    
 No formal qualificationsNA5 (29.4)109 (42.2)30 (50.0)
 GCSE, O level, A level, or tradeNA6 (35.3)72 (27.7)16 (26.7)
 Diploma, degree, or higher degreeNA5 (29.4)63 (24.2)12 (20.0)
 Missing 1 (5.9)15 (5.8)2 (3.3)

Phase 1.

An overview of the themes explored is presented in Figure 2. The interviews revealed that OA had a varied and often profound impact on the QOL of affected individuals and their ability to fulfill their needs. To develop items for the questionnaire, direct quotations from the interviews of each participant were identified as possible items. Duplicate items were also identified, and through consensus, the item that appeared to best express the concept conveyed (while meeting the aforementioned criteria) was selected. The draft OAQoL (version 1) was developed in this way and comprised 38 items with a true/not true response option.

thumbnail image

Figure 2. An overview of the themes addressed in the needs-based approach to participant interviews.

Download figure to PowerPoint

Phase 2.

Field test interviews were conducted with 17 individuals (Table 1) who completed the draft OAQoL under the supervision of an interviewer. The measure took between 2 and 9 minutes to complete (mean ± SD 4.4 ± 2.2 minutes). All participants were able to respond to all of the items and, in general, the participants reported that the items were easy to understand and relevant to persons with OA, even if the items did not apply to them currently. Two items were changed on the basis of respondents' comments: “I get embarrassed using stairs” was changed to “I get embarrassed using stairs in public” because interviewed participants stated that people would be unlikely to become embarrassed using stairs when not observed, and “I can't do things spur of the moment” was considered to be confusing and was changed to “I can't do things on the spur of the moment.” The revised OAQoL (version 2) was then used in phase 3.

Phase 3.

Of 635 sent questionnaires, 397 were returned, a response rate of 62.5%. Of these, 259 patients completed the questionnaire and were classified as responders, while 138 replied that they would not like to participate and were classified as nonparticipators. There were no significant differences in age or sex between responders and nonresponders, or between responders and nonparticipators. A majority of the respondents were women (68.7%), with a mean ± SD age of 66.5 ± 12.5 years (range 21–98 years) and a mean symptom duration of 12.6 ± 9.1 years (range 0.5–45). The knee and hand were the most common sites of pain (40.2%, respectively), followed by the foot (28.6%) and hip (23.9%). Multiple joint involvement was common (median number of joints affected 4 [range 1–20]). Almost one-quarter of the sample was in paid employment and there was a wide range of educational achievement (Table 1).

Rasch analysis of the draft questionnaire (n = 259) indicated initial misfit to the model (χ2[114] = 250.036, P < 0.0000), with an item fit residual mean of −0.363 and PSI of 0.96032. Eleven items demonstrated high fit residuals, indicating that they were contributing to the poor fit. Once these items were removed, the overall fit to the Rasch model was good (χ2[81] = 89.513, P = 0.2422). A summary of the 3 worst fitting items and best fitting items is presented in Table 2. Two items demonstrated DIF by age: “I feel like a burden to others” and “pain controls my life”; the item “I feel like a burden to others” also demonstrated DIF for sex. Once these items were removed (13 in total), the data still demonstrated good fit to the model (χ2[75] = 83.602, P = 0.232), with an item fit residual mean ± SD of −0.480 ± 0.680 and a PSI of 0.96334, resulting in a 25-item version of the OAQoL (version 3). There was no significant DIF by comorbidity.

Table 2. Summary of worst fitting and best fitting items*
ItemLocationSEFit resχ2DFP
  • *

    Location refers to the location of the item along the metric ruler, SE (the standard error of the measure), and fit res (the fit residuals or how well each item relates to the overall model). A significant χ2 indicates that an item does not fit the model.

I find it difficult to sit through a film or TV programme0.0410.1684.66537.1573< 0.0001
I feel the arthritis is affecting my appearance0.2090.1694.36922.85530.0004
I take it out on people close to me1.4110.1880.96815.11930.0018
I feel slowed down−3.4570.247−0.5910.98830.8040
Walking for pleasure is out of the question−1.9210.189−0.4480.83930.8401
I get embarrassed using stairs in public0.5690.173−0.2820.54230.9096

Distribution of the items indicated that range of difficulty covered by the items was comprehensive. Testing of the local dependency of the items indicated that only 3.02% of the independent t-tests (95% confidence interval [95% CI] −2%, 5%) were found to be outside the range, confirming the unidimensionality of the instrument.

The validity of the OAQoL was assessed by investigating the relationships with other measures commonly used in OA. The 25-item version of the OAQoL (version 3) demonstrated significant moderate correlation with the pain and stiffness domains of the WOMAC (ρ = 0.67 and ρ = 0.71, respectively) and good correlation with the WOMAC disability domain (ρ = 0.80). There was a good correlation between the OAQoL and the GWBI (ρ = −0.68) but only a moderate correlation between the OAQoL and the Cochin Scale (ρ = 0.49).

Phase 4.

In phase 4, 125 of the 201 questionnaires were returned (response rate 62.3%), including 49 nonparticipators. Of the individuals who agreed to participate, 62 completed the questionnaire on 2 occasions. Two respondents had to be excluded due to changes in their treatment during the test–retest period. The profile of this group was similar to that in phase 1 (Table 1) with the exception that more participants reported knee OA in the second round of mailed questionnaires. Once again, there were no differences in age or sex between responders and nonresponders.

Rasch analysis indicated that, while the questionnaire was unidimensional, there was a high fit residual for 1 item (“I worry I let people down”) and DIF associated with sex for the item “I worry I hold people back” and site of OA for the item “It takes me longer to complete household tasks.” These items were removed, leaving a 22-item OAQoL (final version) that demonstrated good fit to the Rasch model (χ2[44] = 44.559, P = 0.533), with an item fit residual mean ± SD of −0.228 ± 1.022 and a PSI of 0.94992. Two items demonstrated borderline DIF for comorbidities (“walking for pleasure is out of the question” and “I feel slowed down”); however, the DIF was not significant. One item (“I can't go places I want to go”) demonstrated borderline DIF for site of OA, but this item was retained in the final questionnaire. The person item threshold map is represented in Figure 3. Once again, testing of the local dependency indicated that only 4.67% of the independent t-tests were significant (95% CI 2%, 13%), supporting the strict unidimensionality of the instrument.

thumbnail image

Figure 3. Person-item threshold distribution for the final 22-item Osteoarthritis Quality of Life questionnaire (OAQoL). This diagram presents the distribution of items: the x-axis is the logit score and represents the interval scaling of the items according to the Rasch model, with –5 being excellent quality of life and 5 being poor quality of life. The lower histogram is where individual items are located along the scale; the top histogram represents the number of people and their total OAQoL logit score.

Download figure to PowerPoint

The test–retest for the total score was explored using Spearman's rho, and for each item, Cohen's kappa. There was an excellent significant correlation between the 2 total OAQoL scores from time 1 to time 2 (ρ = 0.93), with no systematic differences between the scores on each occasion (z = −0.06, P = 0.995), suggesting excellent test–retest reliability. Kappa values for each item ranged from moderate (κ = 0.512) to excellent (κ = 0.926), with most items demonstrating values in the range 0.65–0.85.

To reassess the construct validity of the revised OAQoL, the relationship between the 22-item OAQoL and other measures was again calculated. The results were similar to those found for the draft 25-item OAQoL (version 3), with a significant moderate correlation with the pain and stiffness domains of the WOMAC (ρ = 0.67 and ρ = 0.71, respectively), the GWBI (ρ = −0.65), and the Cochin Scale (ρ = 0.49) and good correlation with the WOMAC disability domain (ρ = 0.78). The instructions and first page of the final OAQoL are shown in Figure 4. The full questionnaire is available from Professor Tennant (

thumbnail image

Figure 4. Instructions and first 7 items of the Osteoarthritis Quality of Life questionnaire (OAQoL).

Download figure to PowerPoint


  1. Top of page
  2. Abstract

The goals of this study were to develop the OAQoL, an OA-specific QOL measure; to explore its psychometric properties; and to conduct a preliminary validation of the measure. While functional impairment and pain have been reported extensively in the literature, themes identified during the interviews indicated that individuals with OA often reported substantial restrictions of life choices and an increased dependency on others. Several issues emerged in the development of the OAQoL related to both the affective and cognitive impact of OA. It was not surprising to find that the final OAQoL included several items that related to the loss of independence, impact on others, and a sense of frustration, fear, and annoyance related to living with OA.

The 22-item OAQoL is a questionnaire that is brief, easy to use, and practical to administer in the clinic, in a clinical trial, or as a postal survey. The application of the needs-based model in OA is valuable because it provides important information on the global impact of the disease from the patient's perspective. The OAQoL items were generated directly from statements made by patients with OA. Furthermore, the measure was derived from and tested against Rasch principles. The OAQoL is a unidimensional measure that has the potential for parametric analysis using Rasch-transformed scores. As an OA-specific instrument, it is likely to be a more sensitive and specific outcome than that provided by generic measures, but this remains to be confirmed. Finally, given that the current sample included several different OA sites and included participants with OA in a number of joints, the instrument has been validated for use in upper limb, lower limb, and combination OA.

Preliminary validation of the OAQoL indicated that there was a moderate association with the 3 domains of the WOMAC, the Cochin Scale, and the GWBI, indicating that the scales assess related but distinct concepts.

The final OAQoL includes 1 item (“I can't go places I want to go”) that demonstrated borderline DIF during the test–retest phase. The decision to keep this item in the final version was based on 2 factors. First, the DIF had only borderline statistical significance and was not significant in the initial Rasch analysis of the OAQoL. Second, the item is one that is similar to items in several of the other needs-based QOL instruments (32, 33, 50) and therefore has the potential to be included in an item bank of QOL instruments in the rheumatic diseases.

This study focused on developing and validating an outcome measure for assessing the impact of OA on QOL using the needs-based model. Further research is necessary to test the clinical responsiveness and applicability of the OAQoL across different cultural contexts, including adaptation for use in other languages and cultures. This measure may provide valuable patient-centered information concerning the experience of individuals with OA and impact on their QOL.


  1. Top of page
  2. Abstract

Ms Keenan had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study design. Keenan, McKenna, Doward, Conaghan, Emery, Tennant.

Acquisition of data. Keenan, McKenna, Doward, Conaghan.

Analysis and interpretation of data. Keenan, McKenna, Doward, Conaghan, Tennant.

Manuscript preparation. Keenan, McKenna, Doward, Conaghan, Emery, Tennant.

Statistical analysis. Keenan, Tennant.


  1. Top of page
  2. Abstract

Merck had no role in the study design, data collection, analysis of results, or preparation of this manuscript.


  1. Top of page
  2. Abstract
  • 1
    World Health Organization. The burden of musculoskeletal conditions at the start of the new millennium. Geneva: WHO; 2003.
  • 2
    American College of Rheumatology Subcommittee on Osteoarthritis Guidelines. Recommendations for the medical management of osteoarthritis of the hip and knee. Arthritis Rheum 2000; 43: 190515.
  • 3
    Arokoski MH, Haara M, Helminen HJ, Arokoski JP. Physical function in men with and without hip osteoarthritis. Arch Phys Med Rehabil 2004; 85: 57481.
  • 4
    Mili F, Helmick CG, Moriarty DG. Health related quality of life among adults reporting arthritis: analysis of data from the Behavioural Risk Factor Surveillance System, US, 1996–99. J Rheumatol 2003; 30: 1606.
  • 5
    Bellamy N, Kirwan J, Boers M, Brooks P, Strand V, Tugwell P, et al. Recommendations for a core set of outcome measures for future phase III clinical trials in knee, hip, and hand osteoarthritis: consensus development at OMERACT III. J Rheumatol 1997; 24: 799802.
  • 6
    Altman R, Brandt K, Hochberg M, Moskowitz R. Design and conduct of clinical trials in patients with osteoarthritis. Osteoarthritis Cartilage 1996; 4: 21743.
  • 7
    Carr AJ, Higginson IJ. Are quality of life measures patient centered? BMJ 2001; 322: 135760.
  • 8
    Nord E, Arnesen T, Menzel P, Pinto JL. Towards a more restrictive use of the term “quality of life.” Qual Life News 2001; 26: 34.
  • 9
    Diener E. Subjective well-being. Psychol Bull 1984; 95: 54275.
  • 10
    Sarvimaki A, Stenbock-Hult B. Quality of life in old age described as a sense of well-being, meaning and value. J Adv Nurs 2000; 32: 102533.
  • 11
    Doward LC, McKenna SP. Defining patient-reported outcomes. Value Health 2004; 7 Suppl 1: S48.
  • 12
    Bellamy N, Buchanan WW, Goldsmith CH, Campbell J, Stitt L. Validation study of WOMAC: a health status instrument for measuring clinically-important patient-relevant outcomes following total hip or knee arthroplasty in osteoarthritis. J Orthop Rheumatol 1988; 1: 95108.
  • 13
    Leeb B, Schweitzer H, Montag K, Smolen J. A metaanalysis of chondroitin sulfate in the treatment of osteoarthritis. J Rheumatol 2000; 27: 20511.
  • 14
    Poiraudeau S, Chevalier X, Conrozier T, Flippo RM, Liote F, Noel E, et al. Reliability, validity, and sensitivity to change of the Cochin hand functional disability scale in hand osteoarthritis. Osteoarthritis Cartilage 2001; 9: 5707.
  • 15
    Rat AC, Coste J, Pouchot J, Baumann M, Spitz E, Retel-Rude N, et al. OAKHQOL: a new instrument to measure quality of life in knee and hip osteoarthritis. J Clin Epidemiol 2005; 58: 4755.
  • 16
    Stucki G, Ewert T, Cieza A. Value and application of the ICF in rehabilitation medicine. Disabil Rehabil 2002; 24: 9328.
  • 17
    Menzel P, Dolan P, Richardson J, Olsen JA. The role of adaptation to disability and disease in health state valuation: a preliminary normative analysis. Soc Sci Med 2002; 55: 214958.
  • 18
    Rosemann TT, Joos SS, Koerner TT, Szecsenyi JJ, Laux GG. Comparison of AIMS2-SF, WOMAC, x-ray and a global physician assessment in order to approach quality of life in patients suffering from osteoarthritis. BMC Musculoskelet Disord 2006; 7: 6.
  • 19
    Janse AJ, Gemke RJ, Uiterwaal CS, van der Tweel I, Kimpen JL, Sinnema G. Quality of life: patients and doctors don't always agree: a meta-analysis. J Clin Epidemiol 2004; 57: 65361.
  • 20
    Janse AJ, Uiterwaal CS, Gemke RJ, Kimpen JL, Sinnema G. A difference in the perception of quality of life in chronically ill children was found between parents and pediatricians. J Clin Epidemiol 2005; 58: 495502.
  • 21
    Carr AJ, Gibson B, Robinson PG. Measuring quality of life: is quality of life determined by expectations or experience? BMJ 2001; 322: 12403.
  • 22
    Brazier JE, Harper R, Jones NM, O'Cathain A, Thomas KJ, Usherwood T, et al. Validating the SF-36 health survey questionnaire: new outcome measure for primary care. BMJ 1992; 305: 1604.
  • 23
    Hurst NP, Jobanputra P, Hunter M, Lambert M, Lochhead A, Brown H, and the Economic and Health Outcomes Research Group. Validity of EuroQOL-a generic health status instrument-in patients with rheumatoid arthritis. Br J Rheumatol 1994; 33: 65562.
  • 24
    Hunt SM, McKenna S. A British adaptation of well-being, the General Well-Being Index: a new tool for clinical research. Br J Med Econ 1992; 2: 4960.
  • 25
    Whitfield K, Buchbinder R, Segal L, Osborne RH. Parsimonious and efficient assessment of health-related quality of life in osteoarthritis research: validation of the Assessment of Quality of Life (AQOL) instrument. Health Qual Life Outcomes 2006; 4: 19.
  • 26
    Carr A. Beyond disability: measuring the social and personal consequences of osteoarthritis. Osteoarthritis Cartilage 1999; 7: 2308.
  • 27
    Bradley C. Importance of differentiating health status from quality of life. Lancet 2001; 357: 78.
  • 28
    Ritva K, Pekka R, Harri S. Agreement between a generic and disease-specific quality-of-life instrument: the 15D and the SGRQ in asthmatic patients. Qual Life Res 2000; 9: 9971003.
  • 29
    Hunt S, McKenna S. The QLDS: a scale for the measurement of quality of life in depression. Health Policy 1992; 22: 30719.
  • 30
    McKenna SP, Doward LC, Meads D, Patrick D, Tennant A. Summary of needs-based quality of life instruments [abstract]. Value Health 2004; 7: S3940.
  • 31
    Whalley D, McKenna S, de Jong Z, van der Heijde D. Quality of life in rheumatoid arthritis. Br J Rheumatol 1997; 36: 8848.
  • 32
    McKenna SP, Doward LC, Whalley D, Tennant A, Emery P, Veale DJ. Development of the PsAQOL: a quality of life instrument specific to psoriatic arthritis. Ann Rheum Dis 2004; 63: 1629.
  • 33
    Doward LC, Spoorenberg A, Cook SA, Whalley D, Helliwell PS, Kay LJ, et al. Development of the ASQOL: a quality of life instrument specific to ankylosing spondylitis. Ann Rheum Dis 2003; 62: 206.
  • 34
    McKenna SP, Doward LC, Niero M, Erdman R. Development of needs-based quality of life instruments. Value Health 2004; 7 Suppl 1: S1721.
  • 35
    McKenna SP, Doward LC. The needs-based approach to quality of life assessment. Value Health 2004; 7 Suppl 1: S13.
  • 36
    Altman R, Alarcon G, Appelrouth D, Bloch D, Borenstein D, Brandt K, et al. The American College of Rheumatology criteria for the classification and reporting of osteoarthritis of the hip. Arthritis Rheum 1991; 34: 50514.
  • 37
    Altman R, Asch E, Bloch D, Bole G, Borenstein D, Brandt K, et al. Development of criteria for the classification and reporting of osteoarthritis: classification of osteoarthritis of the knee. Arthritis Rheum 1986; 29: 103949.
  • 38
    Altman R, Alarcon G, Appelrouth D, Bloch D, Borenstein D, Brandt K, et al. The American College of Rheumatology criteria for the classification and reporting of osteoarthritis of the hand. Arthritis Rheum 1990; 33: 160110.
  • 39
    Kvale S. Interviews: an introduction to qualitative research interviewing. Thousand Oaks (CA): SAGE; 1996.
  • 40
    Gaston JE, Vogl L. Psychometric properties of the general well-being index. Qual Life Res 2005; 14: 715.
  • 41
    Rasch G. Probabilistic models for some intelligence and attainment tests. Chicago: University of Chicago; 1960.
  • 42
    Andrich D, Lyne A, Sheridon B. RUMM 2020. Perth: RUMM Laboratory; 2003.
  • 43
    Tennant A, McKenna SP, Hagell P. Application of Rasch analysis in the development and application of quality of life instruments. Value Health 2004; 7 Suppl 1: S226.
  • 44
    Duncan PW, Bode RK, Min Lai S, Perera S. Rasch analysis of a new stroke-specific outcome scale: the Stroke Impact Scale. Arch Phys Med Rehabil 2003; 84: 95063.
  • 45
    Gerber BS, Pagcatipunan M, Smith EV Jr, Basu SS, Lawless KA, Smolin LI, et al. The assessment of diabetes knowledge and self-efficacy in a diverse population using Rasch measurement. J Appl Meas 2006; 7: 5573.
  • 46
    Pallant JF, Tennant A. An introduction to the Rasch measurement model: an example using the Hospital Anxiety and Depression Scale (HADS). Br J Clin Psychol 2007; 46: 118.
  • 47
    Smith EV Jr. Metric development and score reporting in Rasch measurement. J Appl Meas 2000; 1: 30326.
  • 48
    Streiner DL, Norman GR. Health measurement scales. 2nd ed. Oxford: Oxford University; 1995.
  • 49
    Hagquist C, Andrich D. Is the sense of coherence-instrument applicable on adolescents? A latent trait analysis using Rasch-modelling. Pers Indiv Differ 2004; 36: 9558.
  • 50
    Doward LC, Whalley D, Dewar AL, McKenna SP, Tennant A, Griffiths B, et al. The development of the SLE-QOL: a quality of life instrument specific to systemic lupus erythematosus. Qual Life Res 1999; 8: 609.