A glass or two of wine a day – but no more -- appears to protect older adults from developing dementia, researchers reported here at the International Conference on Alzheimer's Disease.
"Among cognitively normal older adults, one to two alcoholic drinks a day is associated with a 37 percent decreased risk of dementia over 6 years," said Dr. Kaycee Sink, a gerontologist at Wake Forest University School of Medicine.
These "moderate drinkers," who were at least 75 years old, had a lower risk of dementia than peers who abstained completely or those who had more than two drinks a day, Sink and her colleagues found.
However, she said she would not recommend that non-drinkers begin to use alcohol to try to prevent dementia.
Dr. Sink also said that her research found that patients who had been diagnosed with mild cognitive impairment did worse with any level of alcohol intake.
"Physicians need to be clear with their patients exactly what is meant by 'light,' 'moderate' and 'heavy' drinking," said Dr. Maria Carrillo, director of medical and scientific relations at the Alzheimer's Association, which sponsored the Vienna meeting.
In Sink's study, a light drinker consumed one drink or less of alcohol a day, while moderate consumption was one to two drinks. Heavy drinking involved more than two drinks a day.
Carrillo said that future studies would help the association decide whether to issue recommendations on alcohol consumption for prevention of dementia, or as part of a treatment scheme for patients with mild cognitive impairment.
She said it's possible that alcohol itself has no impact on dementia, but may reflect a lifestyle that results in more or less risk for dementia and other illnesses.
"This study does not give license to drink beyond one or two alcoholic beverages a day, since excessive alcohol consumption is associated with alcoholic dementia and other medical problems. And some people think if a little bit is good, more is better," cautioned Dr. Scott Turner, director of the Memory Disorders Program at Georgetown University.
In the study, the researchers identified 3,069 participants in the Ginkgo Evaluation of Memory Study who were age 75 or older. They had undergone a complete assessment of cognition at the start of the study, evaluated for cognitive status at six-month intervals, and then were followed for about six years.
During the study, 388 cases of dementia occurred among the participants originally classified as normal, and 188 cases were diagnosed among patients who had mild cognitive impairment at baseline.
At the outset, 1,286 volunteers were teetotalers, 55.7 percent of whom were women. Of those who drank alcohol, 39.1 percent were women. More abstainers ( 20.2 percent) were classified as having mild cognitive impairment at baseline than alcohol consumers (12.3 percent).
After five years in the study, the 2,587 normal individuals who were self-reported moderate drinkers had a 37 percent reduced risk of developing dementia when compared with the abstainers -- the group most at risk of developing dementia.
However, when the researchers scrutinized development of dementia among the 482 patients with mild cognitive impairment at baseline, they found that heavy drinkers had nearly double the risk of dementia as abstainers.
Neither Sink, Carrillo nor Turner reported any conflicts of interest.
This article was prepared in collaboration with ABC News.