14 December 2006
Background: Moderate alcohol drinking is suggested to be beneficial for cognitive functions, but the results of previous studies have varied greatly. Little is known about the effects of midlife alcohol drinking on the cognitive functions later in life. Methods: Participants were derived from random, population-based samples studied in Eastern Finland in 1972, 1977, 1982, or 1987. A total of 1,341 participants were reexamined in 1998, after an average follow-up period of 21 years, at ages 65–79 years. Results: The participants who did not drink alcohol at midlife had a poorer performance in episodic memory, psychomotor speed, and executive function in late life as compared with infrequent and frequent drinkers, adjusted for sociodemographic and vascular factors. Also late-life nondrinkers had poorer psychomotor speed and executive function. These findings were evident especially among nonsmokers. Further, no interactions between apolipoprotein E4 and alcohol or sex and alcohol were found. Conclusions: Alcohol drinking both at midlife and later is favorably related to the function in several cognitive domains, including episodic memory, psychomotor speed, and executive function, in late life. However, it is not clear whether the association is causal, what is the possible mechanism, and what would be a safe limit of drinking for the best cognitive function.