A longitudinal study of drinking and cognitive performance in elderly Japanese American men: The Honolulu-Asia Aging Study

Objectives. This study prospectively describes the relationships between alcohol intake and subsequent cognitive performance among participants in the Honolulu Heart Pro,sram (HHP). Methods. Alcohol intake was assessed at Exam III of the HHP, and cognitive performance was measured approximately 18 years later with the Cognitive Abilities Screening Instrument (CASI). Complete information was available for 3556 participants, aged 71 to 93 years at follow-up. Results. In multivariate analyses, the relationship between drinking and later cognitive performance appeared nonlinear, as nondrinkers and heavy drinkers (more than 60 ounces of alcohol per month) had the lowest CASI scores and the highest risks of poor and intermediate CASI outcomes. Compared with nondrinkers, the risk of a poor CASI score was lowered by 22% to 40% among men who consumed 1-60 ounces of alcohol per month. Conclusions. We report a positive association between moderate alcohol intake among middle-aged men and subsequent cognitive performance in later life. However, it is possible that the health risks associated with drinking outweigh any potential benefits for many elderly persons.