There has been much recent interest in the hypothesis that aspirin and other non-steroidal anti-inflammatory agents (NSAIDs) protect against gastrointestinal, particularly colorectal, carcinoma. Three lines of evidence support this hypothesis: First, NSAIDs inhibit the growth of colorecal neoplasms in laboratory rodents. Epidemiological studies in humans also suggest a protective effect. Three case control and two cohort studies have examined the relation between aspirin use and large bowel neoplasia, and four of these five studies found a risk reduction for either incidence or mortality of about 50% in regular users of aspirin. Finally, two small intervention studies in patients with familial polyposis showed a short-term reduction in polyp formation in patients treated with NSAIDs. In contrast, a large randomized trial of aspirin in human subjects detected no reduction in incidence of colorectal carcinoma in those assigned to aspirin compared to placebo. Colorectal cancer is common and an effective primary prevention strategy could lead to significant public health benefits. Further research into the potential of aspirin to achieve this is keenly awaited. It is, however, premature to recommend it specifically for this purpose on the basis of current evidence, given the known adverse effects of aspirin and related compounds, especially on the gastrointestinal tract.