Although most colorectal cancers arise from adenomatous polyps, most adenomas do not progress to invasive cancer. Understanding the epidemiology of advanced adenomas, specifically those with severe dysplasia, carcinoma in situ, or intramucosal carcinoma, is crucial to uncovering why some adenomas progress and some do not. Using data from four colonoscopy-based adenoma case-control studies, we compared two case groups: subjects with advanced adenomas (those with severe dysplasia, carcinoma in situ, or intramucosal carcinoma; n = 119) and subjects with nonadvanced adenomas (those with none, mild, or moderate dysplasia; n = 441) to a polyp-free control group (n = 1866) in regard to frequently studied risk factors for colorectal neoplasia. All of the cases were newly diagnosed and had no prior history of adenomas. We used an unordered polytomous logistic model to calculate multivariate odds ratios for advanced and nonadvanced adenoma cases relative to polyp-free controls. Among women, ever use of hormone replacement therapy was more strongly associated with reduced risk of advanced adenomas relative to polyp-free controls [odds ratio (OR), 0.4; 95% confidence interval (CI), 0.2-0.9] than with reduced risk of nonadvanced adenomas (OR, 0.7; 95% CI, 0.4-1.0). Among men, increased physical activity (>or=2 h/week) was more strongly associated with reduced risk for advanced adenomas (OR, 0.4; 95% CI, 0.2-1.0) than with reduced risk for nonadvanced adenomas (OR, 0.8; 95% CI, 0.5-1.2). Apart from these differences, most other risk factors, including body size and cigarette smoking were similar in their association with advanced and nonadvanced adenomas, suggesting that many risk factors for colorectal neoplasia may be important to adenoma formation but not to dysplasia per se.