We conducted an epidemiologic case control study of NSAIDs among 489 lung cancer patients and 978 control subjects. The case patients were diagnosed and treated during 1996-1999 at the James Cancer Hospital and Research Institute, Columbus, OH. Each lung cancer diagnosis was verified by examination of the pathology report. Population controls free of disease were obtained from health screening clinics and frequency-matched to the cases at a 2:1 rate. Matching characteristics included age, gender, and pack-years of cigarette smoking. In order to assess the effects of NSAIDs on tobacco carcinogenesis, only heavy smokers were included in the control group. Information on the use of aspirin, ibuprofen, and prescription NSAIDs was obtained by personal interviews. Effects of NSAIDs on lung cancer risk were assessed by estimating odds ratios (relative risks) with 95% confidence intervals and performing trend tests. Daily intake of NSAIDs for at least 2 years prior to interview was associated with a 68% reduction in the relative risk of lung cancer (RR, 0.32; 95% CI, 0.23-0.44; p<0.01). The inverse trend of lung cancer risk with increasing NSAID use was highly significant (p<0.01). Results were similar for men (RR, 0.41) and women (RR, 0.22), and for the individual compounds, aspirin (RR, 0.25) and ibuprofen (RR, 0.39). These results combined with the current molecular evidence suggest that regular NSAID intake may prevent tobacco carcinogenesis through COX-2 blockade.