The present study examined the association between dietary factors and metabolic syndrome in a 7-year follow-up of Japanese–Brazilians. In 1993, a survey estimated the prevalence of diabetes in a Japanese–Brazilian population aged 40–79 years. From 647 subjects studied at baseline, 394 (60·9%) participated in a second survey (2000); after exclusion of subjects with self-reported diseases, 23·4% (n 151) were included in the present analysis. Metabolic syndrome diagnosis was based on the National Cholesterol Education Program criteria modified for Asians. Food intake was assessed by a validated food-frequency questionnaire. At baseline, mean ages were 56·1 (sd 9·7) and 54·7 (sd 10·6) years for men and women, respectively. Similar cumulative incidence rates of the metabolic syndrome were found in both sexes (36·9 (95% CI 26·6, 48·1)% for men and 38·8 (95% CI 27·1, 51·4) % for women). In 1993, mean values of blood pressure, waist circumference for men and 2h plasma glucose for both sexes were higher among subjects who developed metabolic syndrome when compared with those who did not. Comparisons of nutrient intakes between subsets of subjects who developed or did not develop metabolic syndrome showed an association of protein intake with the syndrome only in men (96·4 (sd 56·8) v. 74·3 (sd 26·0) /; P<0·05). Odds ratios of the metabolic syndrome across tertiles of food groups intake showed that among men the highest tertile of red meat consumption was associated with a 4·7-fold increase in risk of developing the syndrome, after adjustments for confounders. However, the statistical significance of this model disappeared when saturated fatty acids were added. The data raised the possibility of a role of red meat consumption for the occurrence of metabolic syndrome in Japanese–Brazilian men. The present findings may have implications for the prevention of metabolic syndrome in this high-risk population.