In clinical trials on the effectiveness of disease-modifying antirheumatic drugs (DMARDs) in patients with rheumatoid arthritis (RA), it is common to apply a large number of endpoint measures. This practice has several disadvantages. To determine which endpoint measures are most valuable, reports of 32 clinical trials on six DMARDs were reviewed. The frequency with which each endpoint measure was used is described and discussed, as well as the frequency with which the values of each endpoint were significantly different in statistical comparisons within or between groups, thus showing ability to discriminate between drugs not equally effective. The results of this review are discussed and compared with other reports in the literature on the choice of endpoint measures in RA clinical trials. The authors conclude that it is still common practice to evaluate multiple outcome measures. The number of measures could be reduced by using only those that are generally considered important, are sensitive to change, and are able to differentiate between drugs in clinical trials. A joint count, assessment of pain, a questionnaire on functional status, and measurement of erythrocyte sedimentation rate are sufficient.