Although moderate alcohol intake in diabetic Caucasians is associated with a reduction in coronary heart disease mortality, no study in Japanese with diabetes has examined the association between alcohol intake and mortality outcomes. We analyzed the relationship between alcohol intake and all-cause and cause-specific mortality using the database from NIPPON DATA80. At the baseline in 1980, data on history, lifestyle, and physical examinations were collected on study participants aged 30 years and older from randomly selected areas in Japan. After excluding participants with comorbidities, we followed 4,018 male participants (3,614 nondiabetics, 195 with impaired glucose tolerance and 209 diabetic) for 19 years. During the 19 years of follow-up, there were 990 deaths; 328 were from cardiovascular disease and 157 from all-heart diseases. With the never-drinking category serving as a reference, the Cox multivariate-adjusted hazard ratios for non-daily and daily drinkers for cardiovascular mortality were 0.43 (95% confidence intervals: 0.19-0.95) and 0.45 (0.25-0.80), respectively, and 0.33 (0.12-0.91) and 0.31 (0.15-0.67) for all-heart disease mortality in the combined impaired glucose tolerance and diabetic Japanese men. Alcohol drinking in men with glucose intolerance was associated with a significant reduction in cardiovascular and all-heart disease mortality as seen in the general population in Japan.

Estudio basado en un amplio número de hombres japoneses que relaciona consumición de alcohol y mortalidad entre diabéticos y no diabéticos durante un periodo de 19 años. Las categorías de toma de bebida fueron limitadas a: 1 casi nunca bebe, 2 bebe a diario 3 bebe ocasionalmente 4 paró de beber pero lo hacía anteriormente.

A large population-based study of Japanese men has evaluated the association between alcohol consumption and mortality over a 19-year follow-up period among diabetics and non-diabetics. The categories for alcohol intake were limited to (1) almost never drink; (2) drink daily; (3) drink occasionally; and (4) stopped drinking, but used to drink. Approximately 400 subjects developed impaired glucose tolerance (IGT) or overt diabetes mellitus (DM) during follow up and overall there were 990 deaths.

Among subjects without IGT or DM, all-cause mortality (adjusted HR=0.75), CVD mortality (HR=0.53), and all-heart deaths (HR = 0.41) were significantly lower for non-daily drinkers and showed a tendency to be higher for former drinkers. No significant associations were seen for non-daily or daily drinkers for stroke, but the daily drinkers had an increased risk of cancer deaths (HR=1.51). Among subjects with IGT-DM, significantly lower HRs were seen for all-cause mortality (HR=0.55) for non-daily drinkers. For both all-heart deaths and acute myocardial infarction deaths, hazard ratios were significantly lower for non-daily drinkers (HR=0.36, HR=0.19, respectively) and for daily drinkers (HR=0.35, HR=0.34, respectively). For daily drinkers, no other associations between alcohol and mortality were significant.

This study thus supports a large number of observational studies in western countries inand all-cause mortality are lower among moderate (here, “non-daily”) drinkers, both for subjects with and without IGT or DM. In many previous studies, the decrease in cardiovascular death rates associated with alcohol intake among diabetics is even more marked than it is for non-diabetics, and the same tendency was seen in this study.