The Leisure World Cohort data on aspirin and colorectal cancer was reanalyzed in an attempt to explain why this study is the lone observational study in which aspirin appears not to be protective. Aspirin users differed from nonusers on a number of baseline variables, some of which are related to risk of colorectal cancer. After adjustment for these variables, aspirin use still did not appear to be protective. The Leisure World cohort members are considerably older than subjects in the other studies; analyses limited to subjects younger than 75 years showed a slightly protective effect of aspirin in men but not in women. A resurvey 5 years after the start of the study indicated that one-third of the subjects had changed the frequency of their aspirin intake. About 25% of daily analgesic users reported their primary reason for using aspirin to be for preventing heart disease and stroke. Regular aspirin use at a dose adequate for preventing cardiovascular disease may be insufficient for preventing colorectal cancer.