Alcohol intake and cognitive abilities in old age: the Lothian Birth Cohort 1936 study.

Objective: Moderate alcohol consumption has been associated with better cognitive performance in late adulthood, possibly by improving vascular health. Few studies have examined the potentially confounding roles of prior cognitive ability and social class in this relationship.

Method: Participants were 922 healthy adults about 70 years old in the Lothian Birth Cohort 1936 study, for whom there are IQ data from age 11. Alcohol consumption was obtained by self-report questionnaire. Cognitive outcome measures included general cognitive ability, speed of information processing, memory, and verbal ability.

Results: Moderate to substantial drinking (>2 units/day) was associated with better performance on cognitive tests than low-level drinking (≤2 units/day) or nondrinking in men and women. After adjusting for childhood IQ and adult social class, most of these associations were removed or substantially attenuated. After full adjustment, a small, positive association remained between overall alcohol intake and memory (women and men) and verbal ability (women only). Women's overall alcohol intake was derived almost exclusively from wine. In men, effects differed according to beverage type: wine and sherry-port consumption was associated with better verbal ability, but beer was associated with a poorer verbal ability and spirits intake was associated with better memory.

Conclusions: Prior intelligence and socioeconomic status influence both amount and type of alcohol intake and may partly explain the link between alcohol intake and improved cognitive performance at age 70. Alcohol consumption was found to make a small, independent contribution to memory performance and verbal ability, but these findings' clinical significance is uncertain.