Skip Navigation


Rheumatology Advance Access originally published online on January 30, 2007
Rheumatology 2007 46(5):849-855; doi:10.1093/rheumatology/kel445
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
46/5/849    most recent
kel445v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (1)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Harrison, M. J.
Right arrow Articles by Symmons, D. P. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Harrison, M. J.
Right arrow Articles by Symmons, D. P. M.
Related Collections
Right arrow Rheumatoid Arthritis
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The Author 2007. Published by Oxford University Press on behalf of the British Society for Rheumatology. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org

The overall status in rheumatoid arthritis (OSRA) measure—further evidence to support its use in clinical practice

M. J. Harrison, A. Hassell1, P. T. Dawes1, D. L. Scott2, S. M. Knight3, M. J. Davis4, D. Mulherin5 and D. P. M. Symmons

arc Epidemiology Unit, The University of Manchester, 1Department of Rheumatology, University Hospital of North Staffordshire NHS Trust, 2Academic Rheumatology Unit, King's College Hospital NHS Trust, 3Department of Rheumatology, East Cheshire NHS Trust, 4Department of Rheumatology, Royal Cornwall Hospitals NHS Trust and 5Department of Rheumatology, Mid Staffordshire General Hospitals NHS Trust.

Correspondence to: D. P. M. Symmons, Professor of Rheumatology and Musculoskeletal Epidemiology, arc Epidemiology Unit, Stopford Building, The University of Manchester, Oxford Road Manchester, M13 9PT, UK. E-mail: deborah.symmons{at}manchester.ac.uk


   Abstract
Top
 Abstract
Background
 Methods
 Results
 Discussion
 Appendix
 Acknowledgements
 References
 
Objectives. The overall status in rheumatoid arthritis (OSRA) instrument is a simple summary of health status, including disease activity (OSRA-A) and damage (OSRA-D) scores. Despite evidence of the validity of the OSRA, uptake has been low. This study aimed to assess the responsiveness and re-examine the validity of the OSRA using the measures from the British Rheumatoid Outcome Study Group (BROSG) randomized controlled trial of aggressive vs symptomatic treatment of rheumatoid arthritis (RA) patients.

Methods. 466 patients were recruited. Outcome measures included the OSRA, the OMERACT core set and the DAS28, and were collected at baseline and annually for the 3 yrs of the trial. X-rays of the hands and feet were taken at baseline and 3 yrs. Patients were assigned a Townsend score (a measure of social deprivation) according to area of residence. Construct validity was assessed by correlating the OSRA with a range of outcome measures, and testing for the known inequality in RA outcome between patients classified by social deprivation. Responsiveness to change was assessed against self-reported change over the first year of the trial.

Results. The OSRA-A and OSRA-D measures demonstrated construct validity, performing as hypothesized. The OSRA-A was the most responsive measure in the BROSG trial in detecting patient reported improvement and deterioration. The OSRA-D demonstrated similar responsiveness to alternative measures.

Conclusions. Our results demonstrate the validity and responsiveness of the OSRA, and its potential for inclusion in clinical trials. More important, as the OSRA is quick and easily calculated, uses routinely collected information, and provides useful quantitative information about a patient's status and progress it is suitable for use in the routine clinic.

KEY WORDS: Rheumatoid arthritis, Health status, Validity (epidemiology), Quality of life, Socioeconomic factors, Randomized controlled trial


   Background
Top
 Abstract
 Background
Methods
 Results
 Discussion
 Appendix
 Acknowledgements
 References
 
The overall status in rheumatoid arthritis (OSRA) instrument was developed as a simple summary measure that uses details (based on patient's history and examination) that can be routinely collected during a clinic visit [1]. The OSRA questionnaire (Appendix) has four components: demographic information, disease activity (ORSA-A) and damage (OSRA-D) scores, and a treatment category. These four components are recorded in sequence. For example, F55A.1.2.B is a female (F) aged 55 yrs (55) in the first decade of disease duration (A), with inactive disease (OSRA-A score = 1), little cumulative damage (OSRA-D score = 2) on non-steroidal anti-inflammatory drug (NSAID) therapy (B). The activity and damage scores both consist of five dimensions of disease scored using the same principle as the Apgar score, which rates dimensions of health on a three point scale (0,1,2) [2]. Both activity and damage scores range from 0 to 10 with 10 representing the most severe outcome. The activity score was developed to represent the potentially reversible aspects of rheumatoid arthritis (RA), whilst the damage score aimed to capture the irreversible impact of the disease. The introduction of anti-TNF-{alpha} therapy since the development of the OSRA necessitates the addition of an extra treatment category, G.

The validity of the OSRA has been tested in two studies [1, 3], and results have suggested that the measure is both valid and performs as well as alternative methods. The component sub-scales of the OSRA activity and damage have inter-rater reliability (tested in a subsample of 34 patients) and did not correlate too closely in a sample of 347 RA patients confirming the relative independence of these aspects of the OSRA [1]. The scales were correlated cross-sectionally with alternative measures of disease activity, damage and functional disability [1] and longitudinally with alternative activity scales, the SF-36 and the ACR coreset (n = 86) [3]. The results were generally consistent in finding stronger correlations of the OSRA activity score with related measures of activity, and the damage score with measures of function and cumulative damage. The OSRA also has a quick calculation time and low administrative burden, a sample of nine rheumatologists administering the OSRA to a total of 212 patients reported that the completion of the questionnaire added to the length of consultation marginally or not at all [1]. The OSRA activity score also correlates with the clinical patient management decisions of time to next follow-up and treatment change [3]. This provides evidence of the value of the OSRA, particularly in daily clinical use as an audit tool. However, as OSRA measures were slightly less responsive than other disease activity indices in a small 6-month longitudinal study, the OSRA was not recommended for this setting [3]. However, this study did not separate responsiveness of measures according to improvement or deterioration, and definition of change was framed over a 12-month period but only tested over 6 months follow-up period. OSRA activity and damage scales have also been found to discriminate patients with and without wrist pain in a study validating screening for extensor tendon rupture in RA, and correlate with both disease related outcome measures that were used, the HAQ and ESR [4].

Despite the positive findings of validity, the uptake of the OSRA has been limited. Only two studies of RA have used the OSRA as an outcome measure, one assessing screening for tendon rupture [4], and the other a large UK randomized trial [5, 6], which provides the data for this study. The OSRA activity score was one of two measures demonstrating a significant improvement in this trial.

This justifies a return to the assessment of the validity of the OSRA-A and OSRA-D, a more detailed examination of their responsiveness to change, and comparison with alternative measures of activity and damage, this time in the research setting of a clinical trial.


   Methods
Top
 Abstract
 Background
 Methods
Results
 Discussion
 Appendix
 Acknowledgements
 References
 
Setting
The study used the baseline, 1 yr and 3 yrs data of the British Rheumatoid Outcome Study Group (BROSG) randomized controlled trial (RCT) in patients with stable, established RA. The trial was conducted in five rheumatology centres in England (Stoke-on-Trent; Cannock; Truro; King's College Hospital, London and Macclesfield), which include teaching and district general hospitals, serving urban and rural populations.

Patients and methods
The BROSG RCT was conducted between 1998 and 2001 [5, 6]. The aim was to compare the relative clinical and cost effectiveness and utility of symptomatic and aggressive treatment for established RA. Symptomatic care patients were seen at home by a rheumatology specialist nurse every 4 months and annually by a rheumatologist for 3 yrs, with the aim of controlling symptoms. Aggressive care patients were seen at least 4 monthly in hospital and treatment altered (following predefined algorithms) with the aim of suppressing joint inflammation. They also had annual review. The primary outcome measure was change in physical function, measured using the British version of the Health Assessment Questionnaire (HAQ) [7]. The HAQ is a questionnaire measuring functional disability, and ranges from 0 (no disability) to 3 (severe disability). Patients with a baseline HAQ score > 2.5 were excluded. The BROSG trial showed no difference in the primary outcome measure (HAQ) between treatment arms at the end of the trial (adjusted for the patient's baseline HAQ, age at randomization, gender, disease duration and treatment centre) [5, 6]. The analysis of the trial data in this study was restricted to subgroups of responsive patients defined using self-reported change in health. We thus ‘converted’ the trial into a longitudinal cohort study.

The patients had stable established disease. The full report of the trial can be found at www.hta.nhsweb.nhs.uk/index.htm [5]. During the course of the trial only 15 (3%) of patients would have been eligible for treatment with a biologic. We showed that 30% of patients in routine rheumatology clinics would have been eligible for the trial. If the OSRA could distinguish between patients with a small degree of change (stable established disease) it would be able to distinguish between patients with more active disease starting effective therapies.

Demographic data collected included age, gender, disease duration, smoking status and comorbidity. The baseline measurements included the OMERACT core-set of outcome measures for RA clinical trials [8]: a patient global assessment, physician global assessment, 28 tender joint count, 28 swollen joint count, HAQ and laboratory assessment of disease activity (erythrocyte sedimentation rate, ESR). The patients also completed the SF-36 [9], an internationally validated generic health status measure, and the EuroQol (EQ5D) [10], a generic measure of health status and health related utility [11]. The SF-36 includes a self-reported change in health question that asks patients to rate their health compared with 1 yr ago using a 5-point Likert scale of much better, somewhat better, about the same, somewhat worse and much worse. In addition a blinded assessor completed an OSRA annually. The DAS28—a composite measure of RA disease activity which uses the 28 tender joint count, the 28 swollen joint count, the ESR and the patient global assessment, was also calculated [12]. The patient global assessment was inverted to allow calculation of the DAS28, so that 0 denoted the best health state and 100 the worst health state. Patients also completed a VAS for pain. All measures were repeated annually and at the end of the 3-yr trial. X-rays of the hands and feet were taken at baseline and end of the trial, and read by a single observer blinded to treatment allocation; a musculoskeletal radiologist, using the Larsen scoring system [13]. The number of eroded joints was also recorded.

Deprivation was assessed using the Townsend score (a measure of social deprivation based on area of residence) [14, 15] allocated to each patient on the basis of their postcode. The score uses a formula based on UK census data for unemployment (% of economically active 16 + yrs olds that are unemployed), overcrowding (% of households with ≥1 person per room), non-car ownership (% of households not owning a car), and non-home ownership (% of households not owning their own home). The Townsend Index measures the material aspects of social deprivation [16, 17], and has been shown to provide results consistent with the measure of disposable income [16], and correlate strongly with other measures of deprivation [17].

Construct validity
We tested convergent/divergent and discriminative validity. Convergent validity tests how strongly a measure correlates with other related measures. Divergent validity tests that the measure does not correlate too strongly with unrelated measures. Discriminative validity is the ability to distinguish groups defined according to a specified trait or behaviour that is known to cause differences in outcome.

Convergent/divergent validity was tested by correlating the OSRA-A and -D scales (Spearman's rank correlation) with a range of disease specific measures (ESR, DAS28, joint counts, pain, patient and physician global assessments, Larsen–Dale and eroded joint count), and two generic measures of health-related quality of life (EQ-5D and SF-36) and their utility weights (EQ-5D and SF-6D) at baseline. Our hypothesis was that the OSRA-A and OSRA-D measures would correlate with disease specific measures, and that the OSRA-A would correlate more strongly with measures of disease activity and pain and the OSRA-D would correlate more strongly with measures of function and disability.

Discriminative validity was tested using the hypothesis that patients with greater deprivation have poorer RA outcome at baseline. This hypothesis was used as we have previously shown significant relationships between increasing social deprivation as measured by the Townsend Index and higher disease activity, physical function and poorer health-related quality of life in the BROSG study [18]. RA patients with lower levels of education or from areas of social deprivation in the USA and UK have higher levels of comorbidity [19] and mortality [19, 20]. They also have a worse RA outcome when measured by laboratory markers and physical function [21–25].

The relationship between quintile of Townsend score and the outcome measures at baseline was assessed using the generalized least squares random effects model described previously [18]. The model assesses significant differences at baseline and change in outcome measures [adjusted for age, gender, disease duration, current smoking, treatment centre and treatment group (for the change analysis)] over the three years of the trial with respect to Townsend quintile. However, as our hypothesis relates to the baseline outcome, only differences in the baseline intercept are reported.

Responsiveness to change
Responsiveness is a measure of association between a change in the true value of a construct and a change in the observed score of the measure being evaluated [26]. Guyatt's effect size of difference (ESD) [27, 28] was used to compare the responsiveness of disease activity measures based on the mean difference in change between the aggressive and symptomatic groups divided by the pooled baseline S.D. of both groups.

To estimate the responsiveness to important change from the patient perspective, we used the groups of patients reporting ‘somewhat better’ and ‘somewhat worse’ health at the end of the first year of the trial. The smallest detectable change (somewhat better or somewhat worse) was chosen to classify the minimum important difference (MID). For both groups, the MID, the mean change [29], and effect size (ES) [30, 31], the mean change divided by the S.D. for the whole group at baseline, were calculated.


   Results
Top
 Abstract
 Background
 Methods
 Results
Discussion
 Appendix
 Acknowledgements
 References
 
A total of 466 patients were recruited to the trial, and their baseline characteristics are shown in Table 1. A total of 440 (94.4%) patients attended their first year follow-up and 399 (85.6%) patients completed the HAQ at the end of the BROSG trial (Fig. 1). Two patients attending the first year assessment (<0.5%) and 14 (3.5%) patients completing the trial did not complete an OSRA assessment. Seventeen (3.6%) patients died during follow-up, 19 (4.1%) were lost to follow-up and 31 withdrew (6.7%). The 31 patients withdrawing from the study comprised 21 patients who withdrew consent and 10 patients who developed a serious comorbidity.


View this table:
[in this window]
[in a new window]

  		TABLE 1. Baseline characteristics of trial participants

 

Figure 1
View larger version (11K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		FIG. 1. Flowheet of patient numbers throughout the trial. LTF, lost to follow-up.

 
Construct validity
Convergent/divergent
The correlation between OSRA-A and OSRA-D measures was 0.36 (P < 0.001). The OSRA-A correlated most strongly with DAS28 (0.59), pain VAS (0.56) and 28 tender joint count (0.54), and least strongly with the measures of radiological damage [Larsen–Dale (0.06) & erosions (0.05)] (Table 2 and Fig. 2). OSRA-D correlated strongly with the HAQ (0.64), and least strongly with the 28 swollen joint count (0.09) and laboratory measure of inflammation [ESR (0.22)]. There was a much stronger correlation with radiological damage outcome measures than was seen for the OSRA-A. However, these were poor to moderate in strength.


View this table:
[in this window]
[in a new window]

  		TABLE 2. Baseline correlation between OSRA-A and OSRA-D and other measures of disease activity and damage

 

Figure 2
View larger version (22K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		FIG. 2. Radar chart of correlation between the OSRA-A and OSRA-D measures with RA-specific outcome measures. ESR, erythrocyte sedimentation rate; HAQ, health assessment questionnaire; VAS, visual analogue scale. Sign of co-efficients removed for the patient global assessment, EQ-5D and SF-6D.

 
With regards to the generic measures, the OSRA-A was most highly correlated with the SF-36 pain scale (–0.60), the physical composite scale (–0.49), the social function scale (–0.43) and the overall utility scores [SF-6D (–0.49) and EQ-5D (–0.41)], and least correlated with the SF-36 role emotional (–0.24), mental health composite scales (–0.25) and mental health scale (Table 3). The OSRA-D correlated moderately with the physical function (–0.60) and physical composite scores (–0.54) of the SF-36 and with both utility measures [SF-6D (–0.43) and EQ-5D (–0.52)].


View this table:
[in this window]
[in a new window]

  		TABLE 3. Baseline correlation between OSRA-A and OSRA-D and subscales of the generic health-related quality of life SF-36 and HRQoL utility measures

 
Discriminative validity
Both the OSRA-A and OSRA-D were able to detect significant differences between quintiles defined by the Townsend Index of deprivation (Table 4). The co-efficients presented in Table 4 describe linear relationships of poorer baseline scores in more deprived quintiles, adjusted for age, gender, disease duration, current smoking, treatment centre and treatment group compared with the least deprived quintile. The adjusted increase in OSRA-A score of 0.273 and OSRA-D score of 0.204 for each quintile increase in Townsend score equate to mean differences of 1.1 and 0.8 across the quintiles of the Townsend score for the OSRA-A and -D, respectively. We previously reported that the OSRA-A was able to detect the greater improvement in disease activity experienced by patients from more deprived areas reported in the 3 yrs of follow-up in the BROSG trial [18].


View this table:
[in this window]
[in a new window]

  		TABLE 4. Differences in baseline score (intercept) presented as a linear function of deprivation quintile adjusted for age, gender, disease duration, smoking, treatment and centre (n = 466)

 
Responsiveness and sensitivity to change
The OSRA-A was the most responsive disease activity score used in the BROSG trial. The OSRA-A detected a 0.40 (P = 0.01) greater mean reduction (adjusted for baseline value and predictors of missing data) in activity in the aggressive treatment arm than the symptomatic arm of the BROSG trial. This change summarized, using Guyatt's effect size of difference, gave a statistic of 0.23 (Table 5). This suggests that the difference between treatment arms is just over one-fifth of the baseline S.D., which corresponds to a small change using Cohen's benchmark [31].


View this table:
[in this window]
[in a new window]

  		TABLE 5. Responsiveness of disease activity and damage measures to changes between treatment arms of the BROSG trial

 
The MID for the OSRA-A, based on the self-reported change in health, was 0.98 in deteriorating patients and 0.79 in improving patients. The corresponding MID for the OSRA-D were 0.38 and 0.34, respectively (Table 6). These values represent the smallest change on the OSRA measures that is important from the patient's perspective. Our MID estimates are larger than the significant differences detected between the treatment arms in the BROSG trial (adjusted mean difference: OSRA-A –0.40, OSRA-D –0.05). This suggests that these significant differences between treatment arms probably do not represent an important outcome, but were due to the trial being over-powered to detect differences in the OSRA.


View this table:
[in this window]
[in a new window]

  		TABLE 6. Minimum important difference and effect sizes for measures from the BROSG trial based on self-reported change

 
The effect sizes for the disease activity measures were approximately twice as large as those for function or damage (Table 6). Amongst the disease activity measures, the OSRA-A was the most responsive instrument for measuring both improvement and deterioration. Other disease activity measures demonstrated similar responsiveness for measuring improvement (physician global) and deterioration (DAS28, patient global assessment and pain VAS). However, no other measure was equal to the OSRA-A in measuring both directions of change. Thus the OSRA-A is the most flexible measure for detecting either direction of change.


   Discussion
Top
 Abstract
 Background
 Methods
 Results
 Discussion
Appendix
 Acknowledgements
 References
 
These findings add to the evidence of the construct validity of the OSRA The OSRA-A and -D scales both correlated moderately with similar measures and less strongly with unrelated scales, and were able to detect known differences in health status across groups defined by social deprivation. The OSRA-A was also able to detect the greater improvement in the patients from more deprived areas seen during the 3-yr follow-up of the BROSG trial.

The OSRA-A was the disease activity measure that was most sensitive to change in the BROSG trial. The OSRA-A was the most sensitive whether comparing small treatment effects between the arms of the trial, or the minimum detectable improvement or deterioration. The equal or next best measure of disease activity differed according to whether the change being observed was improvement or worsening, so the OSRA-A was the most flexible measure, being the most sensitive to change irrespective of the direction.

The OSRA-D was slightly more responsive than the HAQ. The measure showed almost equal responsiveness in both improving and deteriorating patients. The former is a concern as the measure is supposed to measure the irreversible impacts of RA, and consequently the best outcome should be no change. The reversibility was attributable to the inclusion of physical and social function aspects in the OSRA-D [3]. These are influenced as much by current disease status as permanent damage, it is the description of the measure as non-reversible, not the performance of the OSRA-D that is at fault. This section of the questionnaire should perhaps be referred to as ‘IMPACT’ instead of ‘DAMAGE’ when used in the future. The OSRA-D was found to correlate moderately strongly with the physical function, role physical and social function subscales of the SF-36 and the usual activity, self-care and mobility scales of the EQ-5D questionnaire.

In conclusion, the performance of the OSRA in the BROSG trial demonstrates, for the first time, its potential as a useful and responsive addition to the measures currently recommended for inclusion in clinical trials. However, in practice this is unlikely to happen. Firstly, the DAS28 has already achieved widespread use in clinical trials and its components are currently recommended as part of the ACR core set of disease activity measures [32] and OMERACT core set of outcome measures for RA RCTs. Furthermore, the component measures of the DAS28 are assessed from the patient perspective (with the exception of the ESR), which is the preferred perspective for the basis of clinical decision making according to OMERACT [33]. The OSRA is predominantly based on the physician's assessment of the patient. However, the OSRA does have advantages over the DAS28; it can be calculated quickly and easily in real time and is not dependent on the results of a blood test. The OSRA-A was also more responsive to change than the DAS28. In summary, our results reinforce the validity and usefulness of the measure as a straightforward summary measure based on information routinely collected during a clinic visit, it is quick to perform and gives a useful quantitative guide to the patient's progress.

Formula


   Appendix
Top
 Abstract
 Background
 Methods
 Results
 Discussion
 Appendix
Acknowledgements
 References
 
A. Current disease activity

  1. Well-being. The patient is asked how they feel compared with their last visit. (0, same or better; 1, worse; 2, much worse).
  2. Pain. The patient is asked to describe their pain in the last week. (0, mild; 1, moderate; 2, severe).
  3. EMS. The patient is asked for how long they were stiff that morning. (0, <30 min; 1, 30–60 min; 2, >60 min).
  4. Active joints. The following joints are examined: shoulders, elbows, wrists, MCPs, PIPs, hips, knees, ankles, MTPs. An active joint is swollen + tender. MCPs, MTPs, PIPs count as one joint. (0, <3; 1, 3–10; 2, >10 joints active).
  5. Extra-articular disease. (0, none; 1, new nodules; 2, other, e.g. vasculitis, serositis).

B. Cumulative disease damage (impact)

  1. Function. The patient is asked: (a) Do you get out of the house (except in other people's cars)? (i.e. housebound) (b) Can you dress yourself? (c) Can you feed yourself? (d) Can you climb stairs? (0, mild; 1, problems with (b–d); 2, housebound)
  2. Social life: The patient is asked: is your social life is as active as that of a healthy person of your age? (0, normal; 1, moderately impaired; 2, severely impaired)
  3. Number of destroyed or replaced large joints (hips, knees, shoulders). (0, none; 1, 1–2; 2, >3)
  4. Small joint damage (hands, feet, neck). (0, mild; 1, needs collar, special shoes, hand splints; 2, surgery done or needed)
  5. Organ impairment (present for more than 6 months). (0, none or mild; 1, skin or muscle; 2, other organ)

C. Overall status measure for rheumatoid arthritis

1. Demographic
Sex (M/F): {square}
Age (in years): {square}{square}{square}
Disease duration: A ≤10, B = 10–20, C ≥ 20 {square}

2. Activity 0 1 2
Well-being Same/better Worse Much worse {square}
Joint pain None/mild Moderate Severe {square}
EMS <30 min 30–60 min >60 min {square}
Active joint count <3 3–10 >10 {square}
Extra-articular disease None New nodules Vasculitis, serositis, other EAD {square}
Total {square}{square}

3. Damage (impact)
Function Mild Problems to dress, eat, climb stairs Housebound {square}
Social life Normal Moderately impaired Severely impaired {square}
Number of destroyed or replaced large joints 0 1–2 ≥3 {square}
Small joint damage Mild Needs collar, special shoes, hand splints Surgery performed or required {square}
Organ impairment (>6–12 months) None/mild Skin or muscle Major organ {square}
Total {square}{square}

4. Drug therapy
A. None/analgesics B. NSAID alone C. Second-line
D. Second-line + steroid E. Cytotoxic ± steroids F. Steroids G. Biologics {square}

SUMMARY:{square}.{square}{square}{square}.{square}.{square}{square}.{square}{square}.{square}


   Acknowledgements
Top
 Abstract
 Background
 Methods
 Results
 Discussion
 Appendix
 Acknowledgements
References
 
This project was funded by the NHS Executive, UK (NHS HTA project number 94/45/02). The views and opinions expressed within do not necessarily reflect those of the NHS Executive. The NHS Executive commissioned this work, but played no part in the design, data collection, analysis, interpretation, report writing or decision to publish this article. We gratefully acknowledge the help of the rheumatologists and specialist rheumatology nurses in the BROSG centres. We are grateful to Dr C. Ehrhardt who referred some of the patients to the King's College recruitment centre, and Dr J. Saklatvala (Stoke-on-Trent) who kindly read the X-rays. The arc Epidemiology Unit is funded by a programme grant from the Arthritis Research Campaign, UK.

The authors have declared no conflicts of interest.


   References
Top
 Abstract
 Background
 Methods
 Results
 Discussion
 Appendix
 Acknowledgements
 References
 

  1. Symmons DP, Hassell AB, Gunatillaka KA, Jones PJ, Schollum J, Dawes PT. (1995) Development and preliminary assessment of a simple measure of overall status in rheumatoid arthritis (OSRA) for routine clinical use. Q J Med 88:429–37.
  2. Apgar V. (1953) Proposal for a new method of evaluation of newborn infants. Anesth Analg 32:260–7.[Medline]
  3. Birrell FN, Hassell AB, Jones PW, Dawes PT. (1998) Why not use OSRA? A comparison of overall status in rheumatoid arthritis (RA) with ACR core set and other indices of disease activity in RA. J Rheumatol 25:1709–15.[Web of Science][Medline]
  4. Williamson L, Mowat A, Burge P. (2001) Screening for extensor tendon rupture in rheumatoid arthritis. Rheumatology 40:420–3.[Abstract/Free Full Text]
  5. Symmons DPM, Tricker K, Roberts C, Davies L, Dawes P, Scott DL. (2004) The British Rheumatoid Outcome Study Group (BROSG) randomised controlled trial to compare the effectiveness and cost-effectiveness of aggressive versus symptomatic therapy in established rheumatoid arthritis. Health Technology Assessment.
  6. Symmons D, Tricker K, Harrison M, et al. (2006) Patients with stable long-standing rheumatoid arthritis continue to deteriorate despite intensified treatment with traditional disease modifying anti-rheumatic drugs – results of the British Rheumatoid Outcome Study Group randomized controlled clinical trial. Rheumatology 45:558–65.[Abstract/Free Full Text]
  7. Kirwan JR and Reeback JS. (1986) Stanford Health Assessment Questionnaire modified to assess disability in British patients with rheumatoid arthritis. Br J Rheumatol 25:206–9.[Abstract/Free Full Text]
  8. OMERACT. (1993) Conference on Outcome Measures in Rheumatoid Arthritis Clinical Trials. ProceedingsApril 29–May 3, 1992Maastricht, The NetherlandsJ Rheumatol 20: pp. 527–91.
  9. Ware JE Jr and Sherbourne CD. (1992) The MOS 36-item short-form health survey (SF-36). I. Conceptual framework and item selection. Med Care 30:473–83.[Web of Science][Medline]
  10. The EuroQol Group. (1990) EuroQol – a new facility for the measurement of health-related quality of life. The EuroQol Group. Health Policy 16:199–208.[CrossRef][Web of Science][Medline]
  11. Ware JE Jr, Kosinski MA, Keller SD. (1994) SF-36 physical and mental health summary scales: a user's manual(Health Assessment Lab, New England Medical Center, Boston, MA).
  12. Tuttleman M, Pillemer SR, Tilley BC, et al. (1997) A cross sectional assessment of health status instruments in patients with rheumatoid arthritis participating in a clinical trial. J Rheumatol 24:1910–5.[Web of Science][Medline]
  13. Larsen A, Dale K, Eek M. (1977) Radiographic evaluation of rheumatoid arthritis and related conditions by standard reference films. Acta Radiol Diagn 18:481–91.[Medline]
  14. Townsend P, Phillimore P, Beattie A. (1988) Health and deprivation: inequality and the North(Croom Helm Ltd, Kent).
  15. Census Dissemination Unit. (2004) Townsend scores for England & Wales. http://census.ac.uk/cdu/Datasets/1991_Census_datasets/Area_Stats/Derived_data/Deprivation_scores/Pre_calculated_deprivation_scores.htm#4.
  16. Dolan SA, Jarman B, Bajekal M, Davies PM, Hart D. (1995) Measuring disadvantage: changes in the underprivileged area, Townsend, and Carstairs scores 1981–91. J Epidemiol Community Health 49:Suppl 2, S30–3.[Abstract/Free Full Text]
  17. Morris R and Carstairs V. (1991) Which deprivation – a comparison of selected deprivation indexes. J Public Health Med 13:318–26.[Abstract/Free Full Text]
  18. Harrison MJ, Tricker KJ, Davies L, et al. (2005) The relationship between social deprivation, disease outcome measures, and response to treatment in patients with stable, long-standing rheumatoid arthritis. J Rheumatol 32:2330–6.[Abstract/Free Full Text]
  19. Pincus T and Callahan LF. (1985) Formal education as a marker for increased mortality and morbidity in rheumatoid arthritis. J Chronic Dis 38:973–84.[CrossRef][Web of Science][Medline]
  20. Maiden N, Capell HA, Madhok R, Hampson R, Thomson EA. (1999) Does social disadvantage contribute to the excess mortality in rheumatoid arthritis patients? Ann Rheum Dis 58:525–9.[Abstract/Free Full Text]
  21. Callahan LF and Pincus T. (1988) Formal education level as a significant marker of clinical status in rheumatoid arthritis. Arthritis Rheum 31:1346–57.[Web of Science][Medline]
  22. ERAS Study Group. (2000) Socioeconomic deprivation and rheumatoid disease: what lessons for the health service? ERAS Study Group. Early Rheumatoid Arthritis Study. Ann Rheum Dis 59:794–9.[Abstract/Free Full Text]
  23. Jacobi CE, Mol GD, Boshuizen HC, Rupp I, Dinant HJ, Van Den Bos GA. (2003) Impact of socioeconomic status on the course of rheumatoid arthritis and on related use of health care services. Arthritis Rheum 49:567–73.[CrossRef][Web of Science][Medline]
  24. McEntegart A, Morrison E, Capell HA, et al. (1997) Effect of social deprivation on disease severity and outcome in patients with rheumatoid arthritis. Ann Rheum Dis 56:410–3.[Abstract/Free Full Text]
  25. Marra C, Lynd LD, Esdaile JM, Kopec JA, Anis AH. (2004) The impact of low income on self-reported health outcomes in patients with rheumatoid arthritis within a pubicly funded health-care environment. Rheumatology 43:1390–7.[Abstract/Free Full Text]
  26. Bowling A. (2002) Research methods in health: investigating health and health services(Open University Press, Buckingham).
  27. Guyatt GH, King DR, Feeny DH, Stubbing D, Goldstein RS. (1999) Generic and specific measurement of health-related quality of life in a clinical trial of respiratory rehabilitation. J Clin Epidemiol 52:187–92.[CrossRef][Web of Science][Medline]
  28. Terwee CB, Dekker FW, Wiersinga WM, Prummel MF, Bossuyt PMM. (2003) On assessing responsiveness of health-related quality of life instruments: guidelines for instrument evaluation. Qual Life Res 12:349–62.[CrossRef][Web of Science][Medline]
  29. Jaeschke R, Singer J, Guyatt GH. (1989) Measurement of health status. Ascertaining the minimal clinically important difference. Control Clin Trials 10:407–15.[CrossRef][Web of Science][Medline]
  30. Kazis LE, Anderson JJ, Meenan RF. (1989) Effect sizes for interpreting changes in health status. Med Care 27:S178–89.[Web of Science][Medline]
  31. Cohen J. (1977) Statistical power analysis for the behavioural sciences(Lawrence Earlbaum, New Jersey).
  32. Felson DT, Anderson JJ, Boers M, et al. (1993) The American-college-of-rheumatology preliminary core set of disease-activity measures for rheumatoid-arthritis clinical-trials. Arthritis Rheum 36:729–40.[Web of Science][Medline]
  33. Tugwell P and Boers M. (1993) Omeract conference on outcome measures in rheumatoid-arthritis clinical-trials – introduction. J Rheumatol 20:528–30.[Web of Science][Medline]
Submitted 17 August 2006; revised version accepted 12 December 2006.
Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 Clin RehabilHome page
J. Verhoef, P. Toussaint, H. Putter, J. Zwetsloot-Schonk, and T. Vliet Vlieland
Validity and responsiveness of the Rehabilitation Activities Profile (RAP) in patients with rheumatoid arthritis
Clinical Rehabilitation, September 1, 2008; 22(9): 788 - 800.
[Abstract] [PDF]

Home page
 Rheumatology (Oxford)Home page
D. L. Scott and D. P. M. Symmons
The role of specialists in managing established rheumatoid arthritis
Rheumatology, March 1, 2008; 47(3): 237 - 238.
[Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
46/5/849    most recent
kel445v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (1)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Harrison, M. J.
Right arrow Articles by Symmons, D. P. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Harrison, M. J.
Right arrow Articles by Symmons, D. P. M.
Related Collections
Right arrow Rheumatoid Arthritis
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?