We conducted a cohort study to investigate the effects of coffee and green tea consumption on all-cause mortality in a rural Japanese population.Data were obtained from 2, 855 men and women aged 40-79 years in 1989, and during the subsequent 9.9 years of follow-up.Using the Cox regression model to adjust for potential confounding factors, we calculated the multivariate hazard ratios of death from all causes separately for men and women.The multivariate hazard ratio of mortality for men who consumed two or more cups of coffee per day, compared with those who consumed less than half a cup per day, was 0.43(95% confidence interval, 0.30-0.63), and the ratio for those who consumed half to one cup of coffee per day was 0.70(95% confidence interval, 0.52-0.94).Exclusion of subjects with less than 5 years of follow-up did not substantially change the findings.No other statistically significant associations were identified between consumption of the two beverages and all-cause mortality.For men, multivariate hazard ratios of death from apoplexy showed a significant inverse association with increasing coffee consumption.The effects of habitual coffee consumption and its related factors on health in Japan need to be studied in greater detail.J Epidemiol, 2002;12:191-198