From the prospective, population-based cohort of the Rotterdam study, Haag et al report on the association between nonsteroidal anti-inflammatory drug (NSAID) use and the risk of incident stroke and ascertain whether any observed association is restricted to cyclooxygenase (COX)-2–selective NSAIDs. A total of 7636 persons 55 years and older and free of stroke at baseline (1991-1993) were followed for incident stroke until September 2004. During 70 063 person-years of follow-up, 807 persons developed a stroke. Current users of nonselective (hazard ratio [HR], 1.72; 95% confidence interval [CI], 1.22-2.44) and COX-2–selective NSAIDs (HR, 2.75; 95% CI, 1.28-5.95) had a greater risk of stroke, but this was not observed with COX-1–selective NSAID use (HR, 1.10; 95% CI, 0.41-2.97). The greater risk of stroke appears not to be limited to the use of COX-2–selective NSAIDs.