Relative efficiencies of the 7 rheumatoid arthritis Core Data Set measures to distinguish active from control treatments in 9 comparisons from clinical trials of 5 agents.

The 7 Core Data Set measures to assess rheumatoid arthritis (RA) were analysed for their relative efficiencies to distinguish active from control treatments in 9 comparisons of 5 agents, methotrexate, leflunomide, infliximab, adalimumab, and abatacept, in 8 clinical trials. Among the 7 measures, levels of relative efficiencies were in a similar range, highest for the physician global estimate, followed by, in order, patient global estimate, physical function on a health assessment questionnaire (HAQ), pain, swollen joint count (SJC), an acute phase reactant laboratory test - erythrocyte sedimentation (ESR) or C-reactive protein (CRP), and tender joint count (TJC). Comparisons of only 3 measures, SJC and ESR/CRP (regarded as optimal indicators of inflammation) and HAQ function (regarded as most likely to be affected by joint damage and therefore least reversible) indicated relative efficiencies for HAQ function at least as great as for SJC or ESR/CRP, although 8 of the nine comparisons involved patients with disease duration > 6.9 years. The findings indicate a strong rationale for a Core Data Set of 7 measures, as no single measure was clearly superior in relative efficiency in all clinical trials. At the same time, 'objective' laboratory ESR/CRP, TJC and SJC were not superior to 'subjective' global estimates of the physician or patient or patient self-report measures of physical function or pain, to differentiate active from control treatments. The findings challenge a traditional view that laboratory and clinical examination findings are more robust than patient self-report scores and physician global estimates to assess and monitor RA patients.