Aspirin use, cancer, and polyps of the large bowel

BACKGROUND: The effect of aspirin use on 490 patients with cancer of the colon, 340 with cancer of the rectum as the first primary site, and 212 patients with polyps having no coexisting malignancies was compared with that of two groups of control subjects that consisted of 524 hospital patients having no cancers and no diseases of the digestive organs and 1138 healthy visitors to the screening clinic. All subjects entered Roswell Park Cancer Institute between 1982 and 1991. METHODS: After adjustment for adulthood lifetime duration of aspirin use, sex, age, residence, and education, the risk of having cancers and polyps of the colon or the rectum among people who had been using aspirin at least for 1 year before the illness relative to that of nonusers was estimated using multiple logistic regression procedure. RESULTS: The odds ratio estimates showed that the risk of colorectal cancers declined progressively as the frequency of aspirin use increased compared with control groups. Among patients reporting use of aspirin two or more times a day, the odds ratio estimates for colorectal cancers were 0.33 (95% confidence interval CI, 0.72-0.15) and 0.44 (95% CI, 1.10-0.18) compared with those of screening clinic visitors and hospital control subjects, respectively. The odds ratio for patients with polyps who had used aspirin less than once a day relative to that of nonusers was 0.28 (95% CI, 0.62-0.13) and 0.43 (95% CI, 1.09-0.17) compared with screening clinic visitors and hospital control subjects, respectively. CONCLUSIONS: There is a risk reduction effect of aspirin use on the incidence of colorectal cancers and polyps, and this effect is dose related.