Objective. To assess the risk of perforation or hemorrhage of peptic ulcer on treatment with nonsteroidal antiinflammatory drugs (NSAIDs), both as a class and as individual agents.

Methods. A case–control study of medication histories in 494 patients and 972 matched control subjects.

Results. The increase in risk (odds ratio) with NSAID therapy was 5.1 times the risk in controls. The odds ratio for piroxicam was 6.3 (95% confidence interval [CI] 3.3–12.0), as compared with 2.9 for diclofenac, ketoprofen, and sulindac combined (95% CI 2.0–4.2). The effect of other risk factors was also considered, and the adjusted odds ratios were 4.1 for all NSAIDs, 6.4 (95% CI 2.8–15.0) for piroxicam, and 3.3 (95% CI 2.0–5.5) for diclofenac, ketoprofen, and sulindac combined.

Conclusion. The estimate of overall risk of peptic ulcer complications with NSAIDs is similar to that found in other studies. There appear to be differences in risk between agents.