Prospective Study of Alcohol Consumption and Risk of Dementia in Older Adults

Abstract & Commentary

Synopsis: Compared with abstention, consumption of 1-6 drinks weekly of alcoholic beverages was associated with a lower risk of dementia among adults older than 65 years.

Source: Mukamal KJ, et al. JAMA. 2003;289: 1405-1413.

Alcohol consumption is a common practice, but the notion that drinking alcoholic beverages has health benefits remains controversial. Like drinking, dementia is common among adults older than 65 years and thus, is an important public health concern. Given that there are more than 360,000 new cases of Alzheimer dementia diagnosed annually in the United States, there is a pressing need to identify modifiable factors that may cause or prevent dementia. The present study was undertaken to see if there was a link between alcohol consumption and the risk of dementia. Pre-existing data were conflicting, with some data indicating that alcohol promoted cortical atrophy or cerebral hemorrhage, which would promote dementia. Other data indicated a beneficial effect, possibly because of a reduced risk of cardiovascular disease or increased cerebral circulation. Prior studies also suggested sex-linked differences in the health effects of alcohol consumption.

This was a meticulously conducted, prospective, nested case-control study of 373 persons with incident dementia and 373 controls who were among 5888 adults older than 65 participating in the Cardiovascular Health Study, a prospective, population-based cohort study in 4 US communities. Participants underwent magnetic resonance imaging of the brain and cognitive testing between 1992 and 1994 and were followed until 1999. The ascertainment of dementia was thorough, involving extensive testing done in 4 stages. Alcohol consumption was estimated by questionnaire at 6-month intervals for the course of the study. A number of factors that might modify the association between alcohol consumption and dementia were assessed, including sex, race, age, educational attainment, body mass index, diabetes, income level, physical activity, smoking, depression, use of hormone replacement therapy, cardiovascular disease, stroke, lipoprotein profiles, and apolipoprotein E genotype (e4 allele).

Of the 373 cases, 258 had Alzheimer disease, 44 had vascular dementia, 54 had combined AD and vascular dementia, and 17 had other kinds of dementia. Age-adjusted rates of dementia were 56 per 1000 among black participants and 36 per 1000 among white subjects. Using the entire cohort, logistic regression models that adjusted for potentially confounding variables indicated that the lowest odds ratio for dementia occurred among those consuming 1-6 drinks per week. The highest odds ratio occurred among those consuming 14 or more alcoholic drinks weekly. Participants who consumed 1-6 drinks weekly had a 54% lower odds of experiencing dementia that those who abstained [OR 0.46; CI, 0.27-0.77]. This same reduction held when AD risk was analyzed separately from all causes of dementia. With increasing alcohol intake, women seemed to fare better than men. For women, the odds ratio fell to 0.23 [CI, 0.09-0.61] for those who drank 7-13 alcohol beverages weekly. For men, the OR increased to 1.42 for those consuming 7-13 drinks weekly. When considering the risk in those who drank more than 14 drinks weekly, for women, the OR remained low at 0.39 while for men it was 2.40. Although Mukamal and colleagues state that the type of alcoholic beverage (wine, beer, liquor) did not change the association, Table 4 in the manuscript showed that among those drinking more than 14 drinks weekly, the OR for wine was 0.62, while it was 1.96 for beer and 1.08 for liquor. Having APOE e4 further increased the risk of dementia in those drinking more than 14 drinks weekly. These associations were similar in those who met criteria for depression.

Comment by Sarah L. Berga, MD

Alcohol consumption is often considered a taboo subject of discussion, even in physician’s offices. But patients need to know which lifestyle factors make a difference. Counseling patients about alcohol consumption is not as straightforward as counseling patients about smoking cessation, because there is less room for dogmatism when it comes to alcohol. As far as I know, there are no known health benefits associated with smoking or using tobacco. In contrast, studies have indicated that light-to-moderate drinking may confer a range of health benefits when compared to abstention. The present study, which has many strengths, suggested that there is a relatively narrow dose range in which alcohol may provide neuroprotection from dementia. Because the dose range for benefit was wider for women than for men and for those who drank wine rather than beer or liquor, one wonders if the different results in men and women are explained by alcohol preferences. Did women drink more wine than men? Did men drink more beer and liquor than women? Mukamal et al did not provide these data nor discuss this aspect. They do caution that the study results should not be interpreted as suggesting an increased intake of alcohol is recommended for women. As I have pointed out in previous reviews on this subject, women are more at risk than men for alcohol-induced myopathy and cardiomyopathy. Also, some have suggested a link between alcohol consumption and breast cancer in women. Women may absorb alcohol more quickly than men and achieve higher blood levels, placing them at increased risk for motor vehicle and other accidents after drinking. Thus, it would be premature on the basis of this study to counsel women that it was safe to drink more than 6 drinks weekly. We want to screen for those who are alcoholics, while reassuring the rest of our patients that "light" drinking might provide health benefits.

Dr. Berga is Professor and Director, Division of Reproductive Endocrinology and Infertility, University of Pittsburgh.