Late-life alcohol consumption and cognitive function in elderly men.

Moderate alcohol consumption (one to two drinks per day) has been associated with better cognitive function and lower risk of developing dementia in the elderly. In light of alcohol's well-known neurotoxic properties, more evidence from well-controlled population-based studies is required. The objective of this study was to examine whether self-reported alcohol intake at age 70 is linked to cognitive function (assessed by trail making tests (TMTs) A and B, which are measures of attention, mental speed, and flexibility) in a population-based cohort consisting of 652 cognitively healthy elderly men. Linear regression models were used to assess both cross-sectional (i.e., age 70) and prospective (i.e., age 77) associations between alcohol intake and cognitive function. The analyses were adjusted for education, body mass index, energy intake, self-reported physical activity, smoking, a history of hypertension or diabetes, apolipoprotein E ε4 status, and cholesterol levels at the age of 70. Baseline data were obtained from 1990 to 1996. Self-reported alcohol intake (mean 6.9 ± 7.1 g/day) was associated with better performance on TMT-B at ages 70 and 77 (β = -0.87, p < 0.001). In contrast, alcohol intake was not predictive of the difference in performance on these tests between ages 70 and 77. Despite cross-sectional associations with performance in a test of executive functioning, moderate intake of alcohol was not linked to differences in cognitive performance between ages 70 and 77 in the present study. Thus, our findings do not support the view that daily moderate alcohol consumption is a recommendable strategy to slow cognitive aging in elderly populations.