Rheumatology Advance Access originally published online on June 13, 2007
Rheumatology 2007 46(8):1355-1358; doi:10.1093/rheumatology/kem143
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Trends in medication and health-related quality of life in a population-based rheumatoid arthritis register in Malmö, Sweden
Consultant Rheumatologist, Spenshult Rheumatology Hospital, Oskarström1Consultant Rheumatologist,2Department of Rheumatology, Malmö University Hospital, Malmö, Sweden.
Correspondence to: M. K. Söderlin, Consultant Rheumatologist, Spenshult Rheumatology Hospital, 313 92, Oskarström, Sweden. E-mail: maria.soderlin{at}spenshult.se.
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionAcknowledgementsReferences |
|---|
Objectives. To study trends in treatment, health status and health-related quality of life (HRQL) in two cross-sectional surveys over a 5-yr period and in an observational follow-up sub-cohort based on a population-based rheumatoid arthritis (RA) register in Malmö, Sweden.
Material and methods. A continuously updated population-based RA register was established in Malmö city in southern Sweden in 1997. Patient-administered questionnaires in 1997 and 2002 were used to collect information on demographics, medication and health status. Cross-sectional comparisons were made between 1997 and 2002. A longitudinal analysis was also performed in the RA patients participating in both surveys.
Results. Increased proportions of patients were treated with disease-modifying anti-rheumatic drugs (DMARDs) (69 vs 52%), corticosteroids (30 vs 23%), methotrexate (52 vs 29%) and biologics (14 vs 0%) in 2002 compared with 1997. In the cross-sectional analysis, the visual analogue scores (VAS) for pain and general health and the short form 36 (SF-36) domains were slightly better in 2002 than in 1997. In the observational sub-cohort, patients treated with biologics improved significantly in several measures of health status, whereas those starting on methotrexate or undergoing other or no changes in DMARD therapy did not.
Conclusions. In this population-based RA cohort, patients were more actively treated in 2002. Small improvements were seen in health status and these improvements were exclusively attributable to treatment with biologics.
KEY WORDS: Rheumatoid arthritis, Health-related quality of life, SF-36, Biologics, Epidemiology
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAcknowledgementsReferences |
|---|
Over the past few decades, the treatment of rheumatoid arthritis (RA) has changed to more active intervention early in the disease. There is now consensus that RA should be diagnosed and treated early and the aim of the treatment should be remission [1–3].
Study settings whereby treatment, health status and health-related quality of life (HRQL) have been followed in population-based RA cohorts are relatively rare [4–9]. The Oslo RA register was established in 1994 and covers 
85% of all RA patients in Oslo, Norway. SF-36 (Short Form 36-Item Health Survey) scores from this register have been reported in several studies [4–7]. In a cross-sectional questionnaire-based study from the Oslo register, health status was found to be improved in all dimensions when cross-sectional surveys from 1994 and 2001 were compared. Symptom-modifying and disease-modifying drugs were used more extensively in 2001 [7]. The Norfolk Arthritis Register is a population-based early arthritis cohort, in which 50% of patients have RA. HRQL was assessed by SF-36 and reported 5 yrs from the date of registration 1990–91, follow-up 1995–96 [8]. The Norfolk RA patients were treated conservatively, sulfasalazine being the primary drug of choice. A recent population-based study from Rochester, MN, USA showed that clinical practice has changed according to the US guidelines for DMARD treatment in RA. Seventy per cent of the RA patients in Olmsted country had received DMARDs within 2-yrs of onset of RA (patients included in 1984–94) [9].
The aim of this study was to analyse trends in medication, health status and HRQL both cross-sectionally and in a sub-population followed longitudinally using a population-based RA register, in Malmö, Sweden, between 1997 and 2002 [10, 11]. We were especially interested in learning whether the guidelines for more aggressive treatment in RA had been translated into clinical practice and what the effects on HRQL would be.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAcknowledgementsReferences |
|---|
The Malmö RA register
In 1997, a register of all RA patients fulfilling the 1987 American College of Rheumatology (ACR) criteria [12] was established in Malmö, a city in southern Sweden with a population of 205 000 individuals over the age of 20 yrs. The register is intended to cover all the RA patients known at the Rheumatology Department of the University Hospital in Malmö and all of the RA patients attending the three to four rheumatologists practicing privately in Malmö. These office-based rheumatologists are an integrated part of the Rheumatology service and attend regular weekly meetings at the hospital department. This cooperation has resulted in equal principles of investigation, treatment and contact with the ‘rheumatological team’, i.e. rheumatology nurses, physiotherapists, occupational therapists, orthopaedic surgeons and hand surgeons [11]. The RA patients in Malmö are generally treated by rheumatologists and are usually not referred to general practitioners. In the case of early RA, the treatment principle is the same as for the rest of Sweden: a fast referral to a rheumatologist.
In 1997, the register covered an estimated 90% of the RA patient population in the area and it can thus be considered to be population-based. The prevalence of RA and the sex and age distributions in the Malmö RA register were found to be comparable with the RA prevalence in studies from Halland in Sweden and to the data from a population-based RA register in Oslo, Norway [13, 14]. Furthermore, validation against patients seeking primary healthcare for musculoskeletal symptoms in 1999 revealed that the register entails >90% of those in this group that might be suspected to have RA, based on fulfilment of at least four of the seven ACR criteria. All patients are registered using the unique national 10-digit ID number assigned to all Swedish residents who were alive in 1947 or born thereafter. When the Malmö RA register was established in 1997, it included all living RA patients known to attend the hospital or to visit any of the privately practicing rheumatologists after 1993. Due to problems encountered in continuously updating the register during the years that followed, a new register was established in 2001–02. The second time around, only patients who were seen during the previous 2 yrs were included, which probably explains the slightly lower number of cases in 2001 as compared with 1997. The study was approved by the ethical committee at Lund University.
Questionnaires were sent to the patients in the Malmö RA register in 1997 and 2002. Demography, working status, medication with DMARDs, visual analog scale (VAS) for general health and pain, use of healthcare, the Swedish version of the Health Assessment Questionnaire (HAQ) [15] and HRQL as measured by the Swedish version of SF-36 were assessed [16, 17]. Normative data from the Swedish population for SF-36 were used as a reference [18]. One reminder was sent to patients who did not answer the questionnaire the first time.
Statistical analysis
Data were analysed with version 14.0 of the Statistical Package for the Social Sciences (SPSS Inc., Chicago, IL, USA). 
2 test, t-test and paired t-test were used when appropriate. An analysis of variance (ANOVA) was used to compare changes in health status and HRQL for patients starting with different therapies in the follow-up sub-cohort of 357 patients.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAcknowledgementsReferences |
|---|
Cross-sectional analyses in 1997 and 2002
In 1997, 1016 patients were included in the Malmö RA register. A total of 29 patients had the diagnosis of juvenile rheumatoid arthritis (JRA). The JRA patients were excluded, leaving 987 patients with a diagnosis of RA. Of these, 724 patients (73%) answered the questionnaire completely, including the HAQ. SF-36 was analysed in these 724 patients. The 263 patients (27%) who did not answer the questionnaire in 1997 had a mean age of 59 years (SD 17) and 77% were women.
In 2002, the Malmö RA register had 895 patients. Of these, 594 patients (66%) answered the questionnaire completely, including the HAQ. SF-36 was analysed in these 594 patients in 2002. The 301 patients who did not answer the HAQ in 2002 had a mean age of 65 years (SD 16) and 76% were women.
The demographics and health status measurements for the 724 patients who had answered the HAQ in 1997, and the demographics for the 594 patients who had answered the HAQ in 2002 are shown in Table 1. In summary, the RA patient populations in 1997 and 2002 were comparable with regard to age, gender and disease duration. The RA patients in 2002 had received more DMARDs earlier (2.4 vs 2.0), and more patients were on DMARDs (69 vs 52%), methotrexate (44 vs 24%), and corticosteroids (30 vs 23%) as compared with 1997. In 2002, 62% of the patients underwent treatment with one DMARD, 7% were treated with two DMARDs and only two patients received three DMARDs. The corresponding figures for 1997 were 52, 2 and 0%, respectively.
|
In 2002, 14% of the patients underwent treatment with biologics (TNF-
inhibitors (n = 85) or anakinra (n = 2)). Compared with the patients not on such therapy, those on biological treatment were younger (mean age 59 vs 65 yrs on traditional DMARDs, P = 0.0001) and showed differences in the following parameters that were, however, not statistically significant: better VAS pain score (mean 39 vs 40), better VAS general health score (mean 37 vs 39) and disease duration (mean 16 yrs vs. 17 yrs). HAQ scores were significantly worse for the patients who received biologics (mean 1.3 vs 1.1, P = 0.02). The scores for VAS general health were significantly better (non-overlapping 95% confidence intervals) in 2002 than in 1997 and the scores for VAS pain and HAQ tended to be better in 2002. All of the SF-36 domains, with the exception of vitality, were better in 2002, but the differences were small: 0–4 points. All of the SF-36 scores in 1997 and 2002 in the Malmö RA register were significantly lower than in the normal Swedish population (data not shown).
Five-year follow-up of an observational cohort of 357 RA patients
An observational cohort was formed from the 357 RA patients who answered both the 1997 and the 2002 questionnaires completely (Table 1). The patients in this cohort were significantly more extensively treated with DMARDs, methotrexate and corticosteroids in 2002 than in 1997. There was a slight worsening in HAQ, unchanged values for SF-36 subscales adjusted for age and sex using Swedish reference data and improvement in the VAS pain scale (Table 1). When stratifying the cohort by changes made in treatment with DMARDs (2002 vs 1997), there were significant differences with regard to improvement for VAS pain, VAS global assessment scales and for the bodily pain domain of the SF-36. Improvement was seen for patients being treated with biologics during the intervening period, but not for subgroups starting on methotrexate or undergoing other or no changes in DMARD therapy (Table 2).
|
New patients included after 1997
There were 287 new RA patients in the register in 2002, of which 183 (64%) had complete data. Of those included in 1997, 408 were not in the 2002 register. Of these, 213 (52%) were deceased and 195 (48%) were lost to follow-up (Fig. 1). The mean age of the 287 new patients included after 1997 was 61 years (SD 16, 95% CI 59–63). The mean HAQ was 0.91 (95% CI 0.8–1.0), mean VAS pain was 40 mm (95% CI 36–43) and mean VAS general health was 38 mm (95% CI 34–41). The SF-36 scores were almost identical to the scores of the cross-sectional cohort of 594 patients in 2002 (data not shown).
|
Mortality
Patients in the 1997 register who had died before 2002 differed in several aspects compared with those who were alive in 2002. They were older (74 yrs vs 59 yrs, P = 0.0001), had a longer disease duration (19 yrs vs 14 yrs, P = 0.001), had more severe disease (VAS pain 53 mm vs. 44 mm, P = 0.004, VAS global 54 mm vs 42 mm, P = 0.0001, HAQ 1.6 vs 1.1, P = 0.0001), had been as extensively treated with DMARDs (no of previous DMARDs 1.2 vs 1.1, P = 0.4), had fewer ongoing DMARDs, 0.29 vs 0.33, P = 0.2), but more often treated with prednisolone (22% vs 12%, P = 0.0001).
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAcknowledgementsReferences |
|---|
In this population-based RA study, we found an improvement in the health status and HRQL in parallel with an increased and more intensive use of treatment with DMARDs and biologics. The improvements seen appear to be largely attributable to the increased use of treatment with biologics.
In 2002, as compared with 1997, more RA patients were treated with DMARDs, the patients had had more previous DMARDs and more patients were on corticosteroids and methotrexate. Fourteen per cent of the RA patients in Malmö were on biologics (almost exclusively TNF- inhibitors). Our figures of methotrexate treatment in 2002 (44%) and corticosteroid treatment in 2002 (30%) are similar to those from Olmsted County, Rochester, MN, USA (40 and 31%, respectively). About 40% of the patients in Olmsted County who were on hydroxychloroquine and methotrexate had combination therapy after 1995 [9]. The fact that only a few patients had combination DMARD treatment in Malmö mirrors local treatment traditions.
More extensive treatment with biologics is most probably needed, both with regard to the proportion of patients treated and treatment time, if improvement in health status is to be seen in population-based RA cohorts. From the analyses of the two cross-sectional groups, it is difficult to assess cause and effect of treatment and changes in SF-36 scores. Caution is required when comparing two cross-sectional samples that may superficially look similar, bearing in mind that the samples are not independent of each other. This, however, also implies that the two groups have a similar distribution with regards to age and sex (Table 1), which makes differences in these variables unlikely to explain the observed improvements in health status. In the cross-sectional analyses, all of the mean SF-36 domains, with the exception of vitality, were slightly better in 2002, but the changes were only 1–4 points. An improvement of three points has been claimed to represent clinical significance in the SF-36 [18, 19]. The observed differences between 2002 and 1997 are supported to some extent by the analysis of the subgroup of 357 patients who responded to both surveys, bearing in mind that these subjects had aged 5 yrs between the two surveys. Changes were seen especially for the subgroup that had been treated with biologics, where significant improvements were found for the VAS scales for pain and general health and the bodily pain subscale of the SF-36. In addition, the point estimates for other subscales such as physical role, general health and emotional role had also improved considerably.
The SF-36 is a generic instrument of HRQL and it may be hypothesised that it could be less sensitive for detection of changes in health status for RA compared with disease-specific instruments. In our study, pain and global assessment VAS scales showed more pronounced improvements than SF-36 subscales over time. In analyses from the Oslo register, on the other hand, it has been shown that the SF-36 physical function scores correlate with the Modified Health Assessment Questionnaire (MHAQ) and the Arthritis Impact Measurement Scales 2 (AIMS2) physical scales and that SF-36 discriminates between different levels of disease activity and may actually be more sensitive than disease-specific measures in detecting low levels of physical disability [4].
In the cross-sectional analysis, the mean values for HAQ tended to be slightly better in 2002 than in 1997, although a comparison is not completely valid since the samples are not independent. In the longitudinally followed sub-cohort of 357 patients, the mean HAQ deteriorated, even in the subgroup exposed to biologics, probably reflecting the increasing age of the group, since HAQ has been predicted to deteriorate by 0.03 points per year in RA [20].
A major strength of the Malmö register is the fact that it is a population-based RA register, which is a unique feature [8, 9, 14]. The bias of patient selection in hospital-based RA registers can thus be avoided. Furthermore, the patients in the Malmö register are identified by their unique personal identification number.
There were 287 new RA patients included in the register in 2002, but not in 1997. Using the recent estimate of yearly incidence of RA in adults in Sweden, 24/100 000 [21], the yearly incidence of RA in the Malmö register area should be 48, i.e. 240 new RA patients in 5 years.
One possible concern may be that of left censorship, which is very difficult to estimate before 1997. By nature, cross-sectional studies do not include patients who have died, migrated or recovered prior to inclusion. In this case, one might assume that patients with more severe disease had died before the establishing of the register in 1997. A second problem, which may hamper the cross-sectional comparisons, is that a substantial number of patients had died or were lost to follow-up between 1997 and 2002. A total of 213 patients died between 1997 and 2002. The deceased patients were older, had a more severe disease and were less extensively treated with DMARDs as compared with those survived. On the other hand new patients included in the 2002 register tended to have a less severe disease, resulting in no major differences in disease duration, HAQ or age between the two surveys. A third problem is that of non-responders. However, there was no significant difference between responders and non-responders regarding gender, mean disease duration or mean age. Furthermore, one concern might be that the recruiting processes differed somewhat between the two occasions. In 2002, only patients who were seen during the previous 2 yrs were included in the register, which may present a bias of recruiting patients with more severe disease in the 2002 cohort.
In summary, in this population-based RA cohort in southern Sweden, health status improved slightly in line with more extensive DMARD therapy between 1997 and 2002. The small positive effect in health status was attributable to the increased use of biologics.
|
Acknowledgements |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAcknowledgementsReferences |
|---|
We wish to thank Jan-Åke Nilsson for statistical assistance and Drs Eva Juran and Lida Marsal for contributing to the Malmö RA register.
The authors have declared no conflicts of interest.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAcknowledgementsReferences |
|---|
- O’Dell JR. Treating rheumatoid arthritis early: a window of opportunity? Arthritis Rheum (2002) 46:283–5.[CrossRef][Web of Science][Medline]
- Guidelines for the management of rheumatoid arthritis: 2002 Update. Arthritis Rheum 2002. 46:328–46.
- Kavanaugh A, Cohen S, Cush JJ. The evolving use of tumor necrosis factor inhibitors in rheumatoid arthritis. J Rheumatol (2004) 31:1881–4.
[Free Full Text] - Kvien TK, Kaasa S, Smedstad LM. Performance of the Norwegian SF-36 Health Survey in patients with rheumatoid arthritis. II. A comparison of the SF-36 with disease-specific measures. J Clin Epidemiol (1998) 51:1077–86.[CrossRef][Web of Science][Medline]
- Brekke M, Hjortdahl P, Kvien TK. Self-efficacy and health status in rheumatoid arthritis: a two-year longitudinal observational study. Rheumatology (2001) 40:387–92.
[Abstract/Free Full Text] - Brekke M, Hjortdahl P, Kvien TK. Changes in self-efficacy and health status over 5 years: a longitudinal observational study of 306 patients with rheumatoid arthritis. Arthritis Rheum (2003) 49:342–8.[CrossRef][Web of Science][Medline]
- Heiberg T, Finset A, Uhlig T, Kvien TK. Seven year changes in health status and priorities for improvement of health in patients with rheumatoid arthritis. Ann Rheum Dis (2005) 64:191–5.
[Abstract/Free Full Text] - Wiles NJ, Scott DG, Barrett EM, et al. Benchmarking: the five year outcome of rheumatoid arthritis assessed using a pain score, the Health Assessment Questionnaire, and the Short Form-36 (SF-36) in a community and a clinic based sample. Ann Rheum Dis (2001) 60:956–61.
[Abstract/Free Full Text] - Kremers HM, Nicola P, Crowson CS, O'Fallon WM, Gabriel SE. Therapeutic strategies in rheumatoid arthritis over a 40-year period. J Rheumatol (2004) 31:2366–73.
[Abstract/Free Full Text] - Kobelt G, Lindgren P, Lindroth Y, Jacobson L, Eberhardt K. Modelling the effect of function and disease activity on costs and quality of life in rheumatoid arthritis. Rheumatology (2005) 44:1169–75.
[Abstract/Free Full Text] - Jacobsson L, et al. Reumatologi i privat respektive offentlig regi i Malmö-en modell för samarbete. Läkartidningen (2001) 98:4710–16.[Medline]
- Arnett FC, Edworthy SM, Bloch DA, et al. The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis Rheum (1988) 31:315–24.[Web of Science][Medline]
- Simonsson M, Bergman S, Jacobsson LT, Petersson IF, Svensson B. The prevalence of rheumatoid arthritis in Sweden. Scand J Rheumatol (1999) 28:340–3.[CrossRef][Web of Science][Medline]
- Kvien TK, Glennas A, Knudsrod OG, Smedstad LM, Mowinckel P, Forre O. The prevalence and severity of rheumatoid arthritis in Oslo. Results from a county register and a population survey. Scand J Rheumatol (1997) 26:412–8.[Web of Science][Medline]
- Ekdahl C, Eberhardt K, Andersson SI, Svensson B. Assessing disability in patients with rheumatoid arthritis. Use of a Swedish version of the Stanford Health Assessment Questionnaire. Scand J Rheumatol (1988) 17:263–71.[Web of Science][Medline]
- Ware JE Jr, Sherbourne CD. The MOS 36-item short-form health survey (SF-36). I. Conceptual framework and item selection. Med Care (1992) 30:473–83.[Web of Science][Medline]
- Sullivan M, Karlsson J, Ware JE, eds. Swedish manual and interpretation guide (1994) Gothenburg: Sahlgrenska University Hospital.
- Kosinski M, Zhao SZ, Dedhiya S, Osterhaus JT, Ware JE Jr. Determining minimally important changes in generic and disease-specific health-related quality of life questionnaires in clinical trials of rheumatoid arthritis. Arthritis Rheum (2000) 43:1478–87.[CrossRef][Web of Science][Medline]
- Samsa G, Edelman D, Rothman ML, Williams GR, Lipscomb J, Matchar D. Determining clinically important differences in health status measures: a general approach with illustration to the Health Utilities Index Mark II. Pharmacoeconomics (1999) 15:141–55.[CrossRef][Web of Science][Medline]
- Wolfe F. A reappraisal of HAQ disability in rheumatoid arthritis. Arthritis Rheum (2000) 43:2751–61.[CrossRef][Web of Science][Medline]
- Soderlin MK, Borjesson O, Kautiainen H, Skogh T, Leirisalo-Repo M. Annual incidence of inflammatory joint diseases in a population based study in southern Sweden. Ann Rheum Dis (2002) 61:911–5.
[Abstract/Free Full Text]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  U. Bergstrom, L. T. H. Jacobsson, and C. Turesson Cardiovascular morbidity and mortality remain similar in two cohorts of patients with long-standing rheumatoid arthritis seen in 1978 and 1995 in Malmo, Sweden Rheumatology, October 25, 2009; (2009) kep301v1. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. G. Grijalva, C. P. Chung, C. M. Stein, E. F. Mitchel Jr, and M. R. Griffin Changing patterns of medication use in patients with rheumatoid arthritis in a Medicaid population Rheumatology, July 1, 2008; 47(7): 1061 - 1064. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||