Diagnostic Criteria in Rheumatoid Arthritis

Abstract

For rheumatoid arthritis (RA), criteria have been developed for classification of groups of patients in population surveys, selection for therapeutic trials, inter-institutional comparison of patients, evaluation of diagnostic trials, diagnosis of RA in individual patients, estimation of disease frequency, and assistance in determining prognosis.

The American Rheumatism Association (ARA) “Criteria for the Classification of Rheumatoid Arthritis–Diagnostic Criteria for Rheumatoid Arthritis” were published in 1958 and have become a standard. These criteria are empiric and attempt to incorporate reasonable combinations of sensitivity and specificity. Despite the recognition that these criteria should be expected to change with improved knowledge, no further modification has been published.

Other criteria such as the ARA “Proposed Criteria for Rheumatoid Arthritis” represent an important effort to encourage uniformity in definitions and use of the term remission. Yet it is well to remember that they have never been tested in prospective studies and are considered preliminary.

The Co-operative Systematic Studies in Rheumatic Diseases, when studying slow-acting antirheumatic drugs, have chosen to use clinical endpoints such as 50% improvement in joint tenderness and swelling counts (or scores) rather than combined variables which include morning stiffness and changes in laboratory variables. This approach permits the true assessment of drug effects.

In summary, current criteria have served us well. Nonetheless, we should not become complacent and should always question their sensitivity, specificity and application.