Because alcohol consumption is known to cause dementia [1], we studied the relation between wine consumption and cognitive deficit in an area where wine drinking represents almost all the alcohol consumption by elderly persons.

Data collected during baseline screening of the "Paquid" study, an evaluation of a cohort of 3777 community residents 65 years of age or older living in the area of Bordeaux, France, were analyzed. The general methodology of Paquid has been described [2].

Patients were classified as nondrinkers if they drank no wine, as mild drinkers if they drank no more than 0.25 L/d, as moderate drinkers if they drank between 0.25 and 0.5 L/d, and as heavy drinkers if they drank more than 0.5 L/d. Cognitive functioning was examined by the Mini-Mental State Examination (MMSE) [3], and cognitive impairment was defined as an MMSE score lower than 24. Odds ratios and 95% confidence intervals (CIs) were computed using a logistic regression model. The dependent variable was the presence or absence of a global cognitive deficit. A bivariate analysis, considering wine consumption as the only covariate, and a multivariate analysis (which included wine consumption, age, educational level, and principal lifetime occupation) were done.

Information about current wine consumption was recorded for 3766 persons. Although mild and heavy consumption was not significantly related to cognitive performance, moderate wine consumption was associated with a lower risk for cognitive deficit than was nonconsumption (odds ratio = 0.63, P = 0.0004) (see Table 1 above). After adjustment for age, educational level, and occupational category, however, no significant difference was found (P > 0.05). One potential confounding variable was occupational category, in that moderate drinking was much lower in housewives (1.9%) and those in domestic service (6.8%) than in farm workers (12.4%), blue-collar workers (14%), farm managers (14.4%), or other occupational categories (14.7%).

These results indicate that, on average, wine consumption has no deleterious effect on cognitive function in elderly community residents. Wine consumption remains one of the last pleasures of this stage in life; our findings argue against prohibiting mild or moderate wine consumption in the elderly.