American Rheumatism Association 1958 and 1987 revised criteria for rheumatoid arthritis--how useful to the clinician.

ABSTRACT

The American Rheumatism Association (ARA) criteria for rheumatoid arthritis (RA) were evaluated in 52 newly referred consecutive, clinically diagnosed cases of RA. The 1958 and 1987 criteria were highly correlated (r = 0.795). By the 1958 criteria, 48 cases (92.3%) had classical/definite RA while 40 cases (76.9%) satisfied the 1987 criteria traditional format. By the 1987 classification tree 50 cases (96.2%) belonged to the first three RA subsets while 2 cases (3.8%) belonged to the non-RA subsets. Comparing the clinicians opinion of the outcome with the statistical modelling of variables based on ARA criteria, we conclude that the clinicians opinion of the outcome is reflected by a combination of morning stiffness duration, number of swollen joints, erosion score on hand radiographs and erythrocyte sedimentation rate. We conclude that evaluation of ARA 1958 and 1987 criteria in individual cases has both diagnostic and prognostic significance. We propose a short questionnaire based on ARA criteria, for use in hospital out patient clinic for diagnosis of RA.