Aim
To evaluate the association between the amount and frequency of alcohol consumption and incident dementia in older Japanese adults using large sample size data over a long follow‐up period.
Methods
This was a retrospective cohort study carried out in Japan. A total of 53 311 older adults were followed from 2008 to 2014. A health checkup questionnaire was used to assess the amount and frequency of alcohol consumption. The Dementia Scale of long‐term care insurance was used as a measure of incident dementia. Cox proportional hazards models were used to calculate adjusted hazard ratios, with their 95% confidence intervals, for the incidence of dementia across the categories of alcohol consumption by sex.
Results
During a 7‐year follow‐up period, 14 479 participants were regarded as having incident dementia. Compared with non‐drinkers, the multivariate adjusted hazard ratios for participants with alcohol consumption ≤2 units per day, occasionally (0.88, 95% CI 0.81–0.96 in men and 0.84, 95% 0.79–0.90 in women) and daily (0.79, 95% 0.73–0.85 in men and 0.87, 95% 0.78–0.97 in women) were statistically significant, and the difference between occasional and daily consumption was only statistically significant in men; however, for participants with alcohol consumption >2 units per day, occasionally (0.91, 95% 0.71–1.16 in men and 1.09, 95% 0.72–1.67 in women) and daily (0.89, 95% 0.81–1.00 in men and 1.16, 95% 0.84–1.81 in women) were not significant.
Conclusions
Alcohol consumption of ≤2 units per day, occasionally or daily, could reduce the risk of incident dementia, with greater benefit for men with such daily consumption. Geriatr Gerontol Int 2019; 19: 740–746.