OBJECTIVE It has been hypothesized that some aspect of a traditional 'Asian' diet, that is low in animal products and high in soya, may be associated with a reduced risk of prostate cancer. This study aimed to examine the association between dietary intake and prostate cancer risk among 18,115 men in Hiroshima and Nagasaki, Japan, using prospective data from the Life Span Study. METHODS Subjects completed a food-frequency questionnaire at baseline (1963, 1965 and/or 1979) and were followed for incident prostate cancer until the end of 1996. During this time, 196 incident prostate cancer cases were identified after 252,602 person-years of observation. Poisson regression was used to calculate incidence rates for each dietary factor after adjustment for age, calendar period, city of residence, radiation dose and education level. RESULTS Fish intake was significantly associated with an increased risk of prostate cancer; men who consumed fish more than four times per week had a 54 increased risk of developing prostate cancer compared with men who ate fish less than twice per week (RR = 1.54; 95% CI, 1.03-2.31). No other food items, including soya products, were significantly associated with prostate cancer risk. CONCLUSIONS These data suggest that dietary factors may not be strong determinants of prostate cancer in these Japanese men, although the increased risk associated with a high consumption of fish warrants further study.