No study has shown a positive relationship between hypercholesterolemia and all-cause mortality in the Japanese population. Therefore, a cohort study of 17.3 years' duration was conducted on 9216 participants aged 30 years or older, selected randomly from throughout Japan. In both the lowest (<4.14mmol/L, 160mg/dl) and highest (>or=6.71mmol/L, 260mg/dl) total cholesterol (TC) groups, there was a positive association between TC and risk of all-cause mortality (hazard ratio (HR) 1.19; 95% confidence interval (CI), 1.03-1.37 and 1.36 (95% CI, 1.05-1.77), respectively). The lowest TC group had an increased risk of liver disease (HR 3.03; 95% CI, 1.70-5.43), whereas the highest TC group had an increased risk of coronary heart disease (HR 3.81; 95% CI, 1.70-5.43). After exclusion of deaths due to liver disease during the entire follow-up period and all-cause deaths within the first 5 years of follow-up, the increased HR in the lowest TC group disappeared (HR 1.05; 95% CI, 0.89-1.24). Although the cut-off point seemed to be higher than that for Western populations, hypercholesterolemia was shown to be positively associated with all-cause mortality in Japan.