Light to Moderate Alcohol Consumption Is Associated With Better Cognitive Function Among Older Male Veterans Receiving Primary Care M. C. Reid, PhD, MD, Peter H. Van Ness, PhD, MPH, Keith A. Hawkins, PsyD, Virginia Towle, MPhil, John Concato, MD, MPH, and Zhenchao Guo, MD, PhD

ABSTRACT Among older persons, the effects of light to moderate alcohol consumption on cognitive function remain inadequately defined. The authors sought to determine whether light to moderate drinking is associated with better cognitive function among older men. Participants included men aged 65 years or older enrolled in a Veteran’s Administration (VA) primary care clinic. Current (past 1 year) and lifetime use, cognitive functioning (as determined by the Trail Making Part B, Symbol Digit, FAS, and Hopkins Verbal Learning tests), and demographic, psychosocial, and medical status were obtained using standardized methods. Participants (N = 760) had a mean age of 74 (range, 65-89) years. Current drinkers (n = 509) as compared with never (n = 31) and former (n = 220) drinkers demonstrated significantly better cognitive performance on 3 (Trails B, Symbol Digit, and Hopkins Verbal Learning) of the 4 tests (P < .01 for all comparisons). In multiple linear regression models, current light to moderate drinking (ie, 7 or fewer drinks per week), as compared to a reference group of never and former drinkers, was associated with better performance on the Trails B, Symbol Digit, and Hopkins Verbal Learning tests (P < .01 for all comparisons). The number of years drinking 7 or fewer drinks per week also was independently associated with better cognitive performance. Current consumption of 7 or fewer drinks per week and the number of years drinking at this level are both associated with better cognitive performance in older male veterans receiving primary care. These findings are consistent with the hypothesis that light to moderate drinking confers cognitive benefits to older persons. (J Geriatr Psychiatry Neurol 2006;19: 98-105) Keywords: alcohol drinking; male veterans; cognitive function

Received September 15, 2005. Received revised December 23, 2005. Accepted for publication January 3, 2006. From the Division of Geriatrics and Gerontology, Weill Medical College of Cornell University, New York (Dr Reid); the Department of Internal Medicine (Ms Towle and Drs Van Ness, Concato, and Guo) and Psychiatry (Dr Hawkins), Yale University School of Medicine, New Haven, Connecticut. This study was supported by a Merit Review grant from the Department of Veterans Affairs. The work was also supported by a VA Career Development Award (Dr Reid) and a Robert Wood Johnson Foundation Generalist Physician Faculty Scholar Award (Dr Reid). Address correspondence to: Dr M. C. Reid, Division of Geriatrics and Gerontology, Weill Medical College, Cornell University, 525 East 68th Street, Box 39, New York, NY 10021. E-mail: [email protected]. DOI: 10.1177/0891988706286513

98

Among older persons, light to moderate alcohol consumption has been associated with a number of beneficial health outcomes, including a reduced risk of renal dysfunction,1 ischemic stroke,2 heart failure,3 other cardiovascular disease,4 and all-cause mortality.4 Although several cross-sectional5-7 and longitudinal8-11 studies found a beneficial relationship between light to moderate alcohol consumption and cognitive performance in doses ranging from 4 or fewer to 14 drinks per week, other investigations did not.12-17 These inconsistent results may be explained by differences in the measurement of both exposure and outcome variables, as well as differences in the populations studied. Importantly, © 2006 Sage Publications

Alcohol Consumption and Cognitive Function in Veterans / Reid et al

prior studies of older persons were not designed to address the relationship between alcohol and cognitive function (ie, were secondary analyses of data collected for other epidemiologic purposes?). To address this limitation, we conducted a study specifically designed to ascertain the health-related effects of alcohol among older persons using previously validated measures for both independent and dependent variables, and performed analyses that adjusted for a large number of important potential confounders. We hypothesized that light to moderate consumption of alcohol, defined as 7 or fewer drinks per week, would be associated with better cognitive function. We also sought to determine whether years of light to moderate drinking were associated with better cognitive performance. Finally, we examined whether current heavy drinking and years of past heavy drinking were associated with poorer cognitive performance. METHODS Study Sample and Method of Enrollment Participants were members of a 2-year longitudinal study that sought to ascertain the effects of current and lifetime alcohol exposure on cognitive and physical function among older veterans receiving primary care. The current report presents analyses of data from the baseline assessment. Patients were eligible to participate if they were (1) enrolled in either of 2 primary care clinics at the West Haven campus of the VA Connecticut Healthcare System, (2) aged 65 years or older, (3) communitydwelling, (4) English speaking, and (5) ambulatory, as well as (6) had a telephone in their place of residence. The study was approved by the local Institutional Review Board. To minimize the potential for selection effects, research assistants screened consecutive patients following checkin at the primary care clinics. When multiple patients checked in simultaneously, the research assistants approached every other patient. A total of 1241 patients was screened for eligibility between July 1, 2000, and August 15, 2001; 935 met the eligibility criteria; 767 (82%) provided written consent and enrolled. We excluded the small number of women (n = 7) in the current analysis, leaving a final sample size of 760. Eligible nonparticipants (n = 168) did not differ from participants with respect to age (P = .12), race (P = .26), or marital status (P = .15).

Data Collection Trained research assistants conducted comprehensive face-to-face assessments of all participants.

Alcohol Measures Participants were categorized as current (consumed 1 or more drinks in the past 12 months), former (drank

99

prior to but not in the past year), or never (reported no current or former use of alcohol) drinkers. Of the 509 current drinkers, 381 (74.9%) had at least 1 drink in the past 30 days and were administered the Timeline Followback Interview.18 This validated instrument uses a calendar prompt to assist recall and provides a detailed assessment of an individual’s current drinking status. A quantity-frequency estimate (drinks per week) was generated for this group of current drinkers. To determine whether the amount of alcohol consumed over the preceding month represented their customary drinking pattern, this group of drinkers was asked whether their level of drinking during the past month was typical of the way they usually drank. Of the 381 current drinkers who had consumed at least 1 drink in the preceding month, 338 (89%) indicated that the amount of alcohol consumed during the previous month was indicative of their customary drinking pattern. Those answering no (n = 43) were then asked the same questions as those who drank during the past year, but not in the past 30 days: “How many days a week do you usually drink?” and “On those occasions, how many drinks do you have?” A quantity-frequency estimate was calculated for these participants. Individuals who had 1 or more drinks in the past 12 months but not in the preceding month (n = 128) were classified as current drinkers because these individuals reported drinking patterns that demonstrated continued (although infrequent) use of alcohol, such as drinking during the summer months only, having a few drinks during the holidays, drinking during periods when family or friends came to town, etc. To estimate lifetime consumption of alcohol, we administered a slightly modified version of the Lifetime Drinking History questionnaire.19 Participants were first asked at what age they started drinking. Participants were asked to recall all drinking patterns from when they began drinking to the present (for current drinkers) or to the time when they stopped drinking (former drinkers). Lifetime drinking patterns (up to a maximum of 15) were recorded for current and former drinkers, including the number of drinks per week and the number of years spent drinking at that level. To identify participants with a history of alcohol problems, we administered the 4-item CAGE20 questionnaire to both current and former drinkers. Participants with scores of 2 or greater were classified as having screened positive (CAGE+) for a lifetime history of abuse/dependent drinking.21,22 To identify former drinkers with a history of drinking problems, we also asked the following open-ended question: “Why did you stop drinking?” All responses were coded by 2 investigators.23 Former drinkers (n = 53) who reported no longer drinking because of 1 or more of the following were classified as having a history of an alcohol problem: developed a health

100

Journal of Geriatric Psychiatry and Neurology / Vol. 19, No. 2, June 2006

problem due to alcohol; experienced alcohol-related social or occupational impairment; sustained an alcohol-related injury; used alcohol in physically hazardous situations (eg, drinking and driving); or began an alcohol-treatment program. Test-retest reliability appraisals of the alcohol measures were conducted (n = 23, mean time between appraisals was 7 days). Kappa estimates were: for drinking status (current vs former), 0.86; for category of drinks per week (7 or fewer but greater than 0; 8-14; 15 or more drinks/ week), perfect agreement; and CAGE positivity (≥2 vs