SUMMARYThe relationship between alcohol consumption and stroke mortality in 1975 in 46 prefectures of Japan was investigated. This was done by adjusting salt intake and several socio-economic factors, i.e., the annual per capita income, the number of persons who received public aid, the number of tatamis (a Japanese traditional floor unit) per household, the unemployment rate, and the unmarried or divorce rate, using a stepwise multiple regression analysis. As dependent variables, the sex-specific and age-adjusted mortality for the middle-aged (35-59 years) and for all ages due to stroke were used. For men, alcohol consumption was significantly related to age-adjusted stroke mortalities for the middle-aged and for all ages independent of salt intake and several socio-economic factors. Alcohol consumption was more strongly related to age-adjusted stroke mortality for the middle-aged than for all ages. For women alcohol was weakly correlated with the stroke mortality of the middle-aged. Salt intake was significantly correlated with stroke mortality for women but not for men. Furthermore, the male: female ratios of the age-adjusted stroke mortality for the middle aged and for all ages were analyzed as well, because alcohol is mostly consumed by men in Japan, and it was expected that the sex ratios would be well correlated to alcohol consumption. The results were as expected. Therefore, it was suggested that the regional difference in stroke mortality in Japan may be explained in part by that of alcohol consumption.
Stroke Vol 17, No 1, 1986STROKE has been one of the leading causes of death in Japan. 1 Hypertension is a major risk factor for stroke.2 ' 8 Mortality, the incidence of stroke, and the prevalence of hypertension have been higher in the northeastern than in the southwestern part of Japan. These areas with high stroke mortality and high prevalence of hypertension correspond to high salt intake.2 -9 "" However, alcohol intake is higher in the northeastern part of the main island (Honshu) of Japan, especially in Akita Prefecture, than in any other area. ' 2 It is well documented not only by cross-sectional epidemiological studies but also by follow-up studies that moderate or heavy level of alcohol drinking is related to hypertension.13 " 16 Furthermore, several animal and epidemiological studies suggest that alcohol is a risk factor for stroke independent of hypertension.6 -8 l7~22Therefore, whether or not the regional difference in stroke mortality across 46 prefectures is related to alcohol consumption, and independent of salt and several socio-economic factors in Japan, was analyzed by a stepwise multiple regression analysis. , excluding Okinawa, in 1975 were used to analyze the relationship between stroke mortality and alcohol consumption. Age-specific stroke mortality data by prefecture and sex were available in the Special Report of Vital Statistics in Japan, 1975.
Materials and MethodsStatistics for 46 prefecturesM The age-adjusted stroke mortalities for all ages by sex were calculated using the na...