Discriminating ability of composite indices for measuring disease activity in rheumatoid arthritis: a comparison of the Chronic Arthritis Systemic Index, Disease Activity Score and Thompson's articular index.

Abstract

To compare the discriminating ability of the chronic arthritis systemic index (CASI), an index that uses the Health Assessment Questionnaire (HAQ) as the main variable, with the disease activity score (DAS) and Thompson's articular index (TAI) to detect high and low disease activity in rheumatoid arthritis (RA). Two hundred and two RA patients were examined. According to criteria proposed previously, they were divided into two subgroups: those with active disease and those with low activity. The areas under receiver operating characteristic (ROC) curves were employed to assess the diagnostic accuracy of the CASI in comparison with the DAS and TAI for the discrimination of disease activity. The difference between areas under the ROC curves of the CASI and TAI (0. 897+/-0.023 vs 0.780+/-0.032) and between the DAS and TAI (0.933+/-0. 018 vs 0.780+/-0.032) was highly significant (P=0.0001), thus reflecting the accuracy of the diagnostic assessment. No difference arose between areas under the ROC curves of the CASI and the DAS (difference between areas=0.036+/-0.022; P=0.103). The CASI discriminates just as well between high and low disease activity as does the DAS. Either index consisting of more than one variable performs better than TAI. We conclude that even including the HAQ, a severity parameter in the long term, it is possible to construct an index that, at any time point, evaluates disease activity as well.