We prospectively investigated whether coffee consumption was associated with decreased risk of colorectal cancer and whether cigarette smoking and stage of disease modify the association in the Singapore Chinese Health Study. During the first 12 years of follow-up, 961 colorectal cancer cases occurred in the cohort of over 60,000 middle-aged or older Chinese men and women living in Singapore. Baseline dietary exposures were assessed through in-person interviews using a validated food frequency questionnaire. The relation between coffee consumption and colorectal cancer risk was assessed by proportional hazards (Cox) regression. No overall association between coffee intake and colorectal cancer was observed. However, in analysis by subsite and stage restricted to ever smokers, the coffee-colon cancer association became statistically significant for advanced disease (P for trend = 0.01). The hazard ratio was 0.56 (95% confidence interval = 0.35-0.90) for advanced colon cancer in drinkers of 2 or more cups per day compared with those who drank no coffee or less than 1 cup per day. Although there is a null association between coffee intake and risk of colorectal cancer overall, coffee may protect against smoking related advanced colon cancer.