Exploring a national practice-based register for the management of rheumatoid arthritis

Abstract

Rheumatoid arthritis (RA) is a chronic, inflammatory, auto-immune disease with an unpredictable and debilitating course. From diagnosis and onward, RA patients who need treatment with anti-rheumatic drugs will require on-going contact with health care professionals to ensure adequate management of their disease. A national quality register for RA, established in Sweden in 1995, aims to support efforts by both rheumatologists and patients to optimise patients future health. The large amounts of data in the register enable investigation of the utility of the register to further the development of the management of RA.

The aim of this thesis was to increase our knowledge about aspects of the clinical management of RA as determined by the rheumatologist. Data from the Swedish Rheumatology Quality Register (SRQ) were used to analyse factors influencing prescription in incident RA and to study the relationship between clinical rheumatologists assessments of disease activity and standardised outcome measures, in particular the disease activity score based on 28 joints counts (DAS28).

The decision to prescribe a disease modifying anti-rheumatic drug (DMARD) in incident RA was associated with the patient's age, disease activity as measured by DAS28, and the calendar year of the patient's first consultation (year), with some variation by hospital type. On the whole, DMARD prescription increased over the five year period with DMARDs being prescribed more frequently in university and county hospitals compared with district hospitals. Although increasing disease activity would more likely elicit a DMARD prescription in all hospital types, this influence was greater in district hospitals than in either university or county hospitals. When the association of year with increasing DMARD prescription was analysed using Statistical Process Control, the increase was accounted for by a single upward step in July 1998, which could well be related to the promulgation of treatment recommendations that same year.

The relationship between the physician's assessment of disease activity and the DAS28 showed that when the rheumatologist deems disease activity to be none in patients treated with biologics, this is not far removed from the DAS28-based definition of remission but is most closely concordant with the recently proposed definition of minimal disease activity. It is likely that DAS28 will remain useful for monitoring quality on a group level but it cannot replace the rheumatologist's global assessment to justify treatment decisions at the individual patient level. In a cross-section of unselected RA patients, both the standard DAS28 based on the erythrocyte sedimentation rate and DAS28 using high sensitive C-reactive protein (DAS28-CRP) were benchmarked against the rheumatologists global assessment of disease activity. Both DAS28 and DAS28-CRP tend to overestimate disease activity compared with the physician's global assessment of disease activity. New cut-offs for DAS28-CRP disease activity states are proposed.

In summary, the existence of a national practice-based register for RA has enabled the identification of factors driving prescription of DMARDs in the treatment of incident RA during a period of change in rheumatological practice. The study has also generated knowledge about the utility of the different versions of DAS28 in populations typical for Swedish clinical practice.