A cohort study aimed to evaluate the effect of drinking green tea on longevity was performed. Three thousand three hundred and eighty female practitioners of chanoyu (Japanese tea-ceremony), living in Tokyo, were followed from 1980 to 1988, and 280 were dead during this period. Standardized mortality ratios were estimated 0.55 when all Japanese women was used as standard population and 0.57 when women living in Tokyo was used, indicating the possibility that green tea is a protective factor for several fatal diseases. mortality ; green tea ; cohort studyIn the year 1211 a Zen priest, named Eisai, wrote in his book "Kissa-yojoki" about the effectiveness of green tea for maintaining good health and prolonging life. There have been few investigations demonstrating his theory scientifically, but several recent epidemiological studies have shown that habitual drinking of green tea decreases the risk of stomach cancer and stroke (Kono et al. 1988 ;Oguni et al. 1989 ; Sato et al. 1989).In this cohort study mortality among chajin, who were actively practicing Chanoyu (Japanese "tea-ceremony"), was surveyed to assess the association between drinking green tea and life span. The subjects were limited to female chajin aged over 50, living in Tokyo who had been registered as teachers of the Urasenke style of Tea as of January 1, 1980. This list was updated yearly based on the notification of death or withdrawal from membership. All subjects were followed up until December 31, 1988 using the yearly updated files. Causes of deaths were identified when the death certificates were available. Assuming that all deaths and withdrawals occurred in the midst of each year, the observed age-specific person-years during the 9 years were calculated. Then the numbers of deaths expected from the person-years were estimated separately by the age-specific death rates of all Japanese women and by those of women living in Tokyo, using Vital Statistics 1985. Finally, two standardized mortality ratios (SMRs) were calculated by the actual number of deaths among the subjects and the two expected numbers of deaths. Ninety-five percent confidence intervals for estimated SMRs were calculated using Byar' s approximation postulating a Poisson distribution (Breslow and Day 1987).The total number of subjects who were enrolled in this cohort study was 3,380. Of these, 280 had died and 435 had withdrawn from membership. The total number of person-years at the end of the observation was 26,326. Table 1 shows the numbers at the beginning and the observed person-years at the end of the observation according

No abstract

No abstract

No abstract

No abstract