BACKGROUND: The effect of meat and fish consumption on cancer risk has been well studied in humans. However, studies related to biliary tract cancer (BTC) are scarce. METHODS: We examined the association between meat and fish consumption and the risk of BTC in a population-based prospective cohort study in Japan. HRs and 95% confidence intervals (CI) were estimated using the Cox proportional hazard model. RESULTS: During 1995 and 1999, 43,177 men and 49,323 women ages 45 to 74 years were enrolled and followed up for 607,757.0 person-years (men) and 728,820.3 person-years (women) until 2012, during which time 217 male and 162 female BTC cases were identified. Higher total meat consumption was significantly associated with a decreased BTC risk in men (HR for the highest vs. lowest quartiles = 0.66; 95% CI, 0.44-0.98; Ptrend = 0.011) but not in women. Similar association was observed with red meat, but no association was observed with poultry. Fish was not associated with BTC risk. We further analyzed each BTC subtype to confirm the observed association with BTC. However, significant association with each BTC subtype was not observed, although a trend of decreased extrahepatic bile duct cancer risk was observed. CONCLUSIONS: BTC risk was lower among men who consumed more meat, particularly red meat, in Japan. IMPACT: This is the first prospective study that evaluated the relationship between meat and BTC. This may provide important suggestions to elucidate the etiology of BTC.