An inverse association between coffee consumption and the risk of colorectal cancer has been reported in several case–control studies, but results from prospective cohort studies have been inconclusive. We conducted a prospective cohort study among a Japanese population to clarify the association between coffee consumption and the risk of colorectal cancer incidence. We used data from the Miyagi Cohort Study for this analysis. Usable self-administered questionnaires about coffee consumption were returned from 22,836 men and 24,769 women, aged 40–64 years, with no previous history of cancer. We used the Cox proportional-hazard regression model to estimate hazard ratios and 95% confidence intervals. During 11.6 years of follow-up (425,303 person-years), we identified 457 cases of colorectal cancer. Coffee consumption was not associated with the incidence of colorectal, colon or rectal cancer. The multivariate-adjusted hazard ratio (95% confidence interval) of colorectal cancer incidence for 3 or more cups of coffee per day as compared with no consumption was 0.95 (0.65–1.39) for men and women (p for trend = 0.55), 0.91 (0.56–1.46) for men (p for trend = 0.53) and 1.16 (0.60–2.23) for women (p for trend = 0.996). Coffee consumption was also not associated with incidence of either proximal or distal colon cancer. We conclude that coffee consumption is not associated with the incidence risk of colorectal cancer in the general population in Japan. © 2006 Wiley-Liss, Inc.