Physical activity has repeatedly been shown to protect against colorectal cancer. Since adenomas are precursors to most colorectal cancers, we examined the relation of physical activity to the risk for colorectal adenomas in a colonoscopy-based study of 200 adenoma cases and 384 adenoma-free controls. Physical activity was assessed by telephone using a modification of a validated questionnaire comprising 17 items concerning three activity dimensions: leisure, work, and sport. Dietary information was obtained with a validated semiquantitative food frequency questionnaire. We created indices for each dimension of physical activity, which we categorized by quartile. Using logistic regression analysis, we found that leisure activity protected women against colorectal adenomas. Women in the second through fourth quartiles were considerably less likely than those in the first quartile to develop adenomas [crude odds ratio = 0.48; 95% confidence interval (CI) = 0.28-0.81]. There was no protective effect of work activity among either men or women. Men who participated in no sport were at increased risk for adenomas (odds ratio = 1.68; 95% CI = 0.93-3.02). Physical activity appears to protect against both colorectal cancer and colorectal adenomas.