Interest in risk factors for the recurrence of adenomatous polyps derives from the use of recurrent adenomas as surrogate end points in longitudinal studies of invasive colorectal cancer. In this case-control study, the effect of increased body mass index (BMI) on the risk of recurrent adenomas was investigated. Subjects consisted of patients seen at three colonoscopy practices in New York City, all of whom had a previous history of adenomas. On index colonoscopy, recurrent cases had an adenoma, whereas controls were normal. Men and women were analyzed separately, with different logistic models developed using backward elimination from a full model containing the covariates age at diagnosis, age-at-highest-weight, pack-years of smoking, activity level, energy intake, and fat and fiber intake. Men in the upper quartiles of BMI were found to be at greater risk of recurrent adenomas. In a model which controlled for age at diagnosis, age-at-highest-weight, activity level, pack-years of smoking and kilocalories, the estimated odds ratios were 2.2, 1.9 and 1.9 respectively for the second, third and fourth quartiles compared to the first quartile. Only the estimate for the second quartile was found to be statistically significant. No effect was observed for women, even in a model which controlled for age at diagnosis, age-at-highest-weight, pack-years and total fat. Obesity may play a role in adenoma recurrence. Confirmation of this finding would have important implications for possible prevention strategies in the future.