Background : Epidemiological studies have indicated that aspirin consumption can lower the risk of large-bowel cancer. These studies are not entirely consistent, however, and their interpretation has been complicated by the possibility that cancer symptoms may have led patients to avoid aspirin or that aspirin may have influenced cancer diagnosis and treatment. Purpose : Our purpose was to determine the effect of aspirin on risk of large-bowel neoplasms in a study in which aspirin use would not be expected to affect tumor detection and tumor-related symptoms would not likely influence aspirin use. A less complicated assessment of the relationship between aspirin and large-bowel tumors should thus be possible. Methods : We studied 793 patients enrolled in a clinical trial of nutrient supplements to prevent large-bowel adenomas. Unlike invasive cancers, adenomas usually do not cause symptoms or detectable gastrointestinal bleeding; thus, adenomas are unlikely to influence aspirin use. Each patient had at least one large-bowel adenoma diagnosed and removed shortly before study entry and had been judged to be free of further tumors by colonoscopy. Use of aspirin was assessed by responses on questionnaires administered 6 and 12 months after enrollment. We performed complete colonoscopies on all patients 1 year after they entered the study and removed all polyps. Results : Patients who reported taking aspirin on both questionnaires (consistent users) had a lower risk of new adenomas at their 1-year follow-up colonoscopy (odds ratio = 0.52; 95% confidence interval = 0.31–0.89) compared with patients who did not report using aspirin on either of the questionnaires. The apparent protective effect of consistent aspirin use was present among both men and women and did not appear to be influenced by the number of prior adenomas. Conclusions : These data further support the hypothesis that aspirin has an antineoplastic effect in the large bowel. Nevertheless, the question of whether aspirin should be used to prevent large-bowel tumors would be best answered by a randomized controlled clinical trial specifically designed to address this issue. (J Natl Cancer Inst 85:912–916, 1993)