Data from 58 premarketing studies of the nonsteroidal antiinflammatory drug flurbiprofen were pooled for analyses of adverse drug reactions (ADRs). These studies included 5602 patients treated with flurbiprofen (N = 4123), aspirin (N = 1033), or placebo (N = 446) for varying durations. Diagnoses included rheumatoid arthritis, osteoarthritis, and other painful musculoskeletal conditions. In these studies serious upper gastrointestinal ADRs occurred in flurbiprofen-treated patients at less than one half the rate seen in aspirin-treated patients. The incidence of serious urinary tract ADRs was lower with flurbiprofen than with aspirin. The flurbiprofen group had no serious clinical ADRs related to the hemic/lymphatic system. The most common laboratory abnormality was a decrease in hematocrit, which occurred less often than in the aspirin group. We also evaluated serious flurbiprofen-related ADRs in 4370 patients in a variety of other studies and reviewed published reports of flurbiprofen clinical trials and case reports. These reviews showed no additional, unanticipated patterns of intolerance. These clinical safety data indicate that in the doses studied, flurbiprofen is a well tolerated agent for patients requiring nonsteroidal antiinflammatory drug therapy.