Examining data from a 1971-74 health survey and a follow-up survey in 1992, the researchers found that older men who drank moderately or heavily and had accompanying comorbidities that could be worsened by alcohol use such as gout or ulcer disease, or who took medications that could interact negatively with alcohol use, such as sedatives or pain medications, had 20 percent higher mortality rates than other drinkers.
The longitudinal study -- the first to examine in a large population the mortality risks inherent in alcohol use and comorbidity -- will be published in the May issue of the Journal of the American Geriatrics Society. It is available now online at http://www.
Previous studies have found that moderate drinking can reduce risks for vascular disease and death, said Dr. Alison Moore, associate professor of geriatrics at the David Geffen School of Medicine at UCLA, and the study's lead researcher.
"None of these studies have specifically looked at the interaction of alcohol use and conditions or medications that may be unsafe with even moderate amounts of alcohol use," she said. "This study shows that while moderate alcohol use may be fine for people who don't have other conditions that could be worsened by the use of alcohol, such alcohol use may not be fine if you take common medications for sleep, or for arthritis pain, or have depression, or have some gastrointestinal condition."
The researchers analyzed data from the 1971-74 National Health and Nutrition Examination Survey I (NHANES I) and the 1992 NHANES I Epidemiologic Follow-up Survey. Participants included 4,691 adults aged 60 and older. Some 10 percent of the people from the first survey were at-risk drinkers -- that is, they had more than two drinks a couple of times a week and they also had selected comorbidities. Eighteen percent of men and 5 percent of women were deemed at-risk drinkers.
The researchers found that 2,673 people (1,379 men and 1,294 women) from the initial survey had died by the time of the follow-up survey approximately 20 years later. Of those who abstained from drinking, 65 percent (76 percent of men and 60 percent of women) had died. Of the drinkers considered to be not-at-risk, 62 percent had died (68 percent of men, 56 percent of women), and of the drinkers considered to be at-risk, 70 percent had died (77 percent of men, 49 percent of women).
In analyses including men and women, at-risk drinkers had a 12 percent increased risk for death and abstainers had an 8 percent increased risk for death as compared to not-at-risk drinkers. In analyses done separately for men and women, men at-risk drinkers had a 20 percent increased risk of death as compared to not-at-risk drinkers, while abstainers had no increase in risk for death. Among women, neither at-risk drinkers nor abstainers had increased risks for death compared to not-at-risk drinkers.
One reason for the gender discrepancy may be that women simply don't drink as much as men, Moore said. Only 89 women in the sample were considered at-risk drinkers, versus 336 men.
"So it's possible we couldn't see any effect on death rates for women, because there were too few at-risk drinking women," she said.
The findings suggest that lower drinking thresholds should be recommended for older adults with specific but common comorbidities.
Other researchers in the study include Lisa Giuli (who conducted this project while a second-year medical student); Drs. Peifeng Hu, David Reuben, Gail Greendale and Arun Karlamangla; professor Robert Gould; and Kefei Zhou.
The John A. Hartford Foundation/American Federation for Aging Research Medical Student Geriatric Scholars Program, the Paul B. Beeson Physician Faculty Scholars in Aging Program, the Mentored Clinical Scientist Awards from the National Institute on Alcohol Abuse and Alcoholism, the National Institute on Aging and the National Institute on Aging UCLA Older Americans Independence Center, and the Resource Center for Minority Aging Research funded the study.