Current issues of ACP Journal Club are published in Annals of Internal Medicine


Regular use of aspirin was associated with a reduced risk for death from colon cancer

ACP J Club. 1992 Mar-April;116:59. doi:10.7326/ACPJC-1992-116-2-059

Source Citation

Thun MJ, Namboodiri MM, Heath CW Jr. Aspirin use and reduced risk of fatal colon cancer. N Engl J Med. 1991 Dec 5;325:1593-6.



To determine whether regular use of aspirin is associated with reduced risk for death from colon cancer.


Cohort study of 662 424 adults followed from 1982 to 1988. The influence of other risk factors was examined in comparisons of selected cases and matched controls.


Community-based study in the United States.


Volunteers for the American Cancer Society recruited 1 185 239 persons > 30 years of age (mean age, 57 years) for the parent study. For this analysis nonwhite persons (n = 81 614) and the 40% of the white cohort for whom there were missing or incomplete data on aspirin use were excluded. Vital status was ascertained for all 662 424 persons who were included.

Assessment of risk factors

Enrollment questionnaires included questions about medication use, smoking, diet, alcohol use, physical activity, current illnesses, family history of cancer, and demographic status. Brands of aspirin and acetaminophen were listed in separate categories. Among users, the analysis was restricted to those who reported using aspirin for at least 1 year.

Main outcome measure

Age-adjusted death rates for colon cancer based on death certificates. Rectal cancers were excluded.

Main results

Relative risk (RR) for death from colon cancer was reduced among persons taking aspirin at all frequencies (P < 0.001 for trend); the greatest reductions were for those taking aspirin ≥ 16 times/mo (RR for men 0.60 [95% CI 0.40 to 0.89]; RR for women 0.58 [CI 0.37 to 0.90]). Risk was similarly reduced when persons with colon cancer and other possibly confounding illnesses at enrollment were excluded. When 598 case patients were compared with 3058 age- and sex-matched controls (excluding those with illnesses at enrollment or missing data, and after adjustment for dietary factors, obesity, physical activity, and family history), the RR for colon cancer death diminished as the dose of aspirin use increased in both men and women. For aspirin use ≥ 16 times/mo, RR for colon cancer death for men was 0.48 (CI 0.30 to 0.76) and for women 0.53 (CI 0.32 to 0.87). No effect was seen for acetaminophen.


White men and women who used aspirin regularly had a reduced risk for death from colon cancer.

Source of funding: American Cancer Society.

Address for article reprint: Dr. M.J. Thun, Department of Epidemiology and Statistics, American Cancer Society, 1599 Clifton Road, Atlanta, GA 30329, USA.


This study suggests that colon cancer deaths may be halved by regular use of aspirin. There are a number of reasons to take the report seriously. The estimates were precise, the effect was strong, and the results were internally consistent—they persisted when the data were analyzed in several ways. Control for possible confounding factors of diet, obesity, family history, and physical activity did not change the results. There was a significant dose-response in both men and women. Moreover, possible beneficial effects are supported by previous studies. Experiments with rodents have consistently shown prevention of carcinogen-induced cancers by nonsteroidal anti-inflammatory drugs (NSAIDs). Two human case-control studies have found beneficial effects of NSAIDs. Uncontrolled studies in patients with polyposis have shown regression of colonic polyps with sulindac.

There are several limitations to the conclusions that can be drawn from this study. The study concerned mortality from colon cancer, not incidence of cancer. The results may, therefore, indicate improved survival (perhaps because of early detection) but not a decrease in cancers. Rectal cancers were excluded. Similar results for both colon and rectal cancer would strengthen the findings. Information was obtained from a single, brief, self-administered questionnaire given at the start of the study. We do not know the quantity of aspirin taken or the indications. Perhaps the underlying conditions requiring aspirin are protective, not aspirin itself. Although the authors examined several potential confounders, other unmeasured factors might account for the findings.

Should physicians prescribe aspirin to prevent colon cancer deaths? Not simply on the basis of this report, even though the results are persuasive. There are still too many unanswered questions, and even small doses of aspirin can have significant adverse effects. Patients receiving aspirin prophylaxis for other conditions can take some comfort in these results. The rest of us should probably wait for additional data.

Robert S. Sandler, MD, MPH
University of North CarolinaChapel Hill, North Carolina, USA

Robert S. Sandler, MD, MPH
University of North Carolina
Chapel Hill, North Carolina, USA