The relationship between vitamin supplement use and colon cancer was assessed in a population-based case-control study among men and women aged 30-62 years. Cases were 251 men and 193 women diagnosed with colon cancer in 1985-1989 in three counties in the Seattle metropolitan area who were identified from the Surveillance, Epidemiology, and End Results cancer registry. Controls were 233 men and 194 women identified by random digit dialing. Supplement use was assessed by questions on frequency, duration, and dose per day (for individual supplements) or type (for multivitamins) during the 10-year period ending 2 years before diagnosis. All results were adjusted for age and sex and were not confounded by other measured behaviors. The average daily intake of supplemental vitamins A, C, E, folic acid, calcium, and multivitamins during the reference period were each associated with reduced risk of colon cancer (all P for trend < 0.03). The strongest associations were for use of vitamin E (odds ratio, 0.43; 95% confidence interval, 0.26-0.71 for > or = 200 IU/day versus none) and multivitamins (odds ratio, 0.49; 95% confidence interval, 0.35-0.69 for daily use versus no use; both P for trend < 0.001). These two associations were also significant using a stricter test of trend limited to supplement users, which reduces the effect of colinearity among these exposures. Because almost all vitamin D supplementation comes from multivitamin pills, the association of vitamin D use with colon cancer could not be distinguished from that of multivitamin use. Clinical trials or cohort studies with long-term assessment would be needed before public health recommendations could be made about supplement use.