Journal of Gerontology: MEDICAL SCIENCES 2005, Vol. 60A, No. 8, 1065–1070
Copyright 2005 by The Gerontological Society of America
Cognitive Function in Drinkers Compared to Abstainers in The New Mexico Elder Health Survey Robert D. Lindeman,1,2 Sharon J. Wayne,2 Richard N. Baumgartner,1,2 and Philip J. Garry2,3 1
Department of Internal Medicine, 2Aging and Genetic Epidemiology Program, and 3Department of Pathology, University of New Mexico School of Medicine, Albuquerque. Background. Recent epidemiological studies have shown that individuals who ingest alcohol regularly have a higher level of cognitive function and are less likely to develop dementia than those who abstain. The purpose of this study was to compare nine measures of cognitive function in drinkers compared to abstainers. Methods. A cross-sectional community survey was conducted of 883 randomly selected Hispanic and non-Hispanic white men and women, age 65 years of age, undergoing a paid home interview and 4-hour interview/examination in a senior health clinic (The New Mexico Elder Health Survey). The interviews included questions on frequency and quantity of alcohol ingested. Results. Participants who consumed alcohol had significantly better mean scores on 7 of 9 cognitive function tests and less frequently had scores below selected ‘‘cut points’’ compared to those who abstained from all alcohol intake. Scoring used multivariate linear and logistic regression models adjusted for sex, ethnicity, age, level of education, and evidence of depression. Conclusions. Participants in the New Mexico Elder Health Survey (nearly equal numbers of Hispanic and non-Hispanic white men and women) who consumed alcohol had better scores on their cognitive tests than did those participants who abstained.
T
HE earliest reports examining the effect of moderate alcohol consumption on cognitive function provided support for the commonly held assumption that chronic ingestion of alcoholic beverages, even at moderate levels of intake, resulted in poorer performance on cognitive tests (1,2). Subsequent research led to the conclusion that there was no significant association between social drinking and the level of cognitive function (3–5). The first reports indicating that drinking in moderation might actually be beneficial for cognitive function appeared in the mid-1990s (6,7). Subsequently, the Epidemiology of Vascular Aging (EVA) study (8) reported that moderate alcohol ingestion by women was positively associated with better performance on 7 of 10 cognitive tests; however, it did not find similar associations in men. Similarly, when results from the Framingham study were reported in 1999, women who drank moderately (2–4 drinks/day) showed superior performance in many cognitive domains compared to abstainers (9). Again, fewer significant relationships were found in men, and then only at higher levels of intake (4–8 drinks/day). One subsequent study supported these findings (10). Others have examined prospectively the relationship between moderate alcohol consumption and the risk for developing dementia. In the Rotterdam study involving participants age 55 years and older (11), light to moderate drinking (1–3 drinks/day) was associated with a lower risk of incident dementia compared to abstainers (odds ratio [OR], 0.46; 95% confidence interval [CI], 0.27–0.77). In the Cardiovascular Health Study involving individuals age 65
years and older (12), light to moderate drinking (1–3 drinks/ day) was similarly associated with a lower risk of dementia (OR, 0.58; 95% CI, 0.38–0.90). The New Mexico Elder Health Survey (NMEHS) was a study of health and health-related issues on nearly equal numbers of elderly (65 years of age) Hispanic and nonHispanic white (NHW) men and women randomly selected from the Health Care Financing Administration (Medicare) rolls of Bernalillo County (Albuquerque), New Mexico (13–19). One purpose of this study was to examine the relationship between alcohol intake and nine measures of cognitive function adjusted for differences in age, sex, ethnicity, and level of education. METHODS
Study Design/Participants Prospective participants (2200; equal numbers of Hispanic and NHW men and women) were randomly selected from the 1990 list of 50,700 Health Care Financing Administration registrants (Medicare recipients), age 65 years and older, residing in Bernalillo county (Albuquerque), New Mexico. After eliminating those who had died, moved from the county, could not be located, or were ineligible because they did not meet criteria to qualify as either Hispanic or NHW, 1666 eligible participants were contacted between 1993 and 1996. Eleven hundred and thirty (67.8%) agreed to participate in paid home interviews. An additional 29 interviewees were found not qualified because they did not meet ethnicity standards (self-identification and three of four 1065
LINDEMAN ET AL.
1066
Table 1. Demographic Characteristics (Sex, Ethnicity, Age, and Years of Education) of 827 Participants by Alcohol Intake Status
Alcohol Intake
No.
Never 419 1 day or fewer/wk 232 2–5 days/wk 69 Daily, ,4 drinks 96 Daily, 4 drinks 11 All participants 827 p for trend ,.001
Years of Age (y) Education % Male % Hispanic Mean 6 SD Mean 6 SD 44 56 65 81 100 53 ,.001
54 42 29 27 45 45 ,.001
74.5 6 6.6 73.1 6 5.1 70.7 6 4.0 74.2 6 5.9 71.3 6 5.0 73.7 6 6.0 ,.001
10.8 13.2 14.1 13.1 11.1 12.0
6 6 6 6 6 6
4.2 3.7 3.5 4.0 4.7 4.2
Notes: Not included in this table are 14 participants who had no data on alcohol intake, another 10 who had no cognitive function/affective test results, and 30 who had Mini-Mental Status Examination (MMSE) scores ,18. SD ¼ standard deviation.
grandparents identified as Hispanic or NHW, respectively), had died, or moved after the home interview and before an examination could be completed. Of the 1101 individuals interviewed at home and found eligible, 883 (80.2%) participated in a 4-hour, paid interview/examination by a nurse practitioner, nurse, and nutritionist. There were no other exclusionary criteria as long as informed consent could be obtained from the participant or legal guardian. A senior health center was used for 89% of the examinations with the remainder performed in the home or in a nursing home. All participants gave a written informed consent, and the research was approved by the Human Research Review Committee of the University of New Mexico Health Sciences Center. Further details on the design and survey instruments used in the study, and the demographic characteristics of the recruited participants are published elsewhere (13–15).
Interview and Examination Interviews were obtained directly from 96% of the participants, and from a spouse, relative, or caretaker in 4% of the cases. A history of alcohol intake was obtained by the nurse, who asked the question, ‘‘How often do you drink alcoholic beverages?’’ Possible answers were: (a) never (Group 1), (b) one or fewer days per week (Group 2), (c) 2– 5 days per week (Group 3), (d) daily, fewer than four drinks per day (Group 4), and (e) daily, four or more drinks per day (Group 5). Neuropsychological Assessment Cognitive test measures and the functions that they were intended to estimate were: (a) the Mini-Mental State Examination (MMSE) (orientation, recall, attention, language, and visual graphic ability), (b) Wechsler Adult Intelligence Scale-Revised (WAIS-R) Digits Forward (attention and immediate memory), (c) Fuld Object-Memory Evaluation (learning and secondary memory), (d) clock drawing (visuoconstruction), and (e) two Color Trail Making Tests (psychomotor speed and cognitive flexibility). The MMSE and other tests used in this survey were administered in English or Spanish, depending on the participant’s language of choice. How these tests were administered and scored in this population, and references to the validated cognitive tests are published elsewhere (16–19).
Three indicators of depression were obtained, specifically a self-report of a past history of depression, whether or not the participant was currently taking antidepressants, and the 15-question short version of the Yesavage Geriatric Depression Scale (GDS) administered in either Spanish or English.
Statistical Analyses The distributions of demographic variables (age, years of education, ethnicity, and sex) were compared between each drinker group and abstainers, and differences were tested using analysis of variance for means and chi-square tests for percentages. For continuous outcome variables, e.g., scores on each of the nine cognitive function tests and GDS scores, group comparisons between participants who were drinkers and those who were abstainers were conducted using generalized linear regression models, and adjusted means were obtained using the least squares method. Means were adjusted for age and level of education (years) for each sex/ ethnicity group; for ethnicity, age, and level of education for all men and all women; and then for ethnicity, sex, age, and level of education for all participants. For the categorical outcome variables (history of depression, taking antidepressants, GDS .6) and for scores below versus above selected cut-points on the nine tests of cognitive function, logistic regression analyses were used to calculate OR values for each drinking category compared to the abstainer category, while controlling for age, years of education, sex, and ethnicity. Added to subsequent linear regression models was the GDS score as an indicator of the presence of depression. Differences were considered statistically significant if the p value was ,.05. All analyses were conducted using SAS software (20). RESULTS Of the 883 participants in the NMEHS, data collection (information on alcohol intake and history of depression, and cognitive and affective testing) was available for 857 individuals (14 had missing information on alcohol intake; another 10 had no data on cognitive or affective testing). Because individuals with dementia would be expected to score very low on these cognitive tests, thereby skewing and biasing (weighting) the means, participants with MMSE scores ,18 were eliminated from the linear regression analyses, but examined separately. There were 30 individuals (10 Hispanic men, 3 NHW men, 13 Hispanic women, 4 NHW women) with MMSE scores ,18. Of these 30, 28 participants (of 447, or 6.3%) were in the abstainer group (Group 1), and 2 (of 234, or 0.9%) were in the group that drank 1 day per week or less (Group 2). There were no participants with MMSE score ,18 among the 176 participants in the other three drinker groups (Groups 3– 5). If those participants with low MMSE scores had not been eliminated, the cognitive test differences between drinkers and abstainers would have been even more impressive. Table 1 shows the distribution of the remaining 827 participants by status of alcohol intake, with distributions by sex and ethnicity for each group. Seventy-nine men and 28 women admitted to consuming alcohol on a daily basis. Because only 11 men (no women) admitted to consuming
COGNITIVE FUNCTION IN DRINKERS VERSUS ABSTAINERS
four or more drinks daily, this group representing ‘‘heaviest drinkers’’ was eliminated from subsequent analyses. Significant differences ( p , .001) existed in the ethnicity and sex distributions by drinking status. Hispanics accounted for over half of the abstainer group, but only one quarter of the daily drinkers. A similar pattern was seen for women. This made it imperative that adjustments be made to the regression analyses for differences in these two variables. There also were significant differences in the mean ages and levels of formal education ( p , .001) with drinkers being younger and having more years of formal education. Table 2 compares the unadjusted mean scores (6 standard deviation) and mean scores adjusted for age, level of education (years), sex, and ethnicity for the nine cognitive tests and the GDS score by drinking status for Groups 1–4. Percentages of participants with a history of depression, currently taking antidepressants, or having a GDS score .6 are shown by drinking status at the bottom of the table. OR values comparing drinkers to abstainers were calculated with adjustments for differences in age, level of education, sex, and ethnicity. Using adjusted multivariate linear regression analyses for Groups 1–4, we found that drinkers had better cognitive test scores on all nine tests, and these were significantly better in seven of nine tests ( p , .05). For each of the four ethnicity/ sex groups, drinkers scored better than abstainers on all nine cognitive tests. Because only two Hispanic women admitted to drinking daily, these differences were not considered analyzable. When ‘‘all men’’ and ‘‘all women’’ were examined separately, the magnitudes of the differences in the cognitive test scores observed between drinkers and abstainers were comparable. All three measures of depression suggested that depression was more common in the abstainer group. Because depression is recognized to give results on cognitive tests suggesting impairment of cognitive function, the regression analyses were rerun with the GDS score as a covariate. These analyses failed to change the results appreciably. An alternative method for examining these data would be to determine the odds for having a low cognitive test score by drinking status. This would require selection of an appropriate cut-point for each of the nine cognitive tests below which one could say there was impaired cognition. This could be a clinically relevant value, the lowest quartile for all participants, or some other arbitrary cut-point. The relative risk for an individual in each drinking category of being below this cut-point was determined by setting the risk for abstainers at 1.0 with calculation of a p for trend by drinking status. Shown in Table 3 are the OR values adjusted for differences in ethnicity, sex, age, and level of education (unadjusted OR values are not shown because they were presented in the previous analysis and were not materially different from the adjusted values). It now was appropriate to include the 30 participants with low (MMSE , 18) scores on their cognitive tests as they would not adversely weight the OR values as they would when mean values were calculated. All but one of the 27 comparisons of participants in the three drinking categories with the abstainers showed OR
1067
Table 2. Means 6 Standard Deviations of the Nine Unadjusted (Un.) and Adjusted for Differences in Ethnicity, Sex, Age, and Years of Education (Adj.) Cognitive Test Scores and the Geriatric Depression Scale (GDS) by Alcohol Intake Alcohol Intake Test No. of participants
Never
1 Day/Wk
2–5 Days/Wk
Daily, ,4
419
232
69
96
p for Trend
MMSE (29) Un. Adj.
25.2 6 2.6 26.6 6 2.0 27.1 6 1.5 27.0 6 1.7 25.9 6 2.0 26.2 6 2.0 26.3 6 1.7 26.7 6 2.1 .001
Draw pentagon (10) Un. Adj.
8.7 6 1.9 9.2 6 1.3 8.9 6 1.6 9.1 6 1.5
9.5 6 1.1 9.2 6 1.5 9.0 6 1.6 9.1 6 1.6 .360
6.6 6 1.6 6.9 6 1.4 6.7 6 1.4 6.9 6 1.5
7.0 6 1.5 7.2 6 1.3 6.8 6 1.5 7.1 6 1.6 .016
7.3 6 1.7 7.6 6 1.6 7.4 6 1.4 7.5 6 1.5
7.9 6 1.4 7.6 6 1.6 7.6 6 1.6 7.6 6 1.6 .111
Fuld–No. retrieved (10) Un. Adj. Fuld–Total recalled (10) Un. Adj. Fuld–No. of names Un. Adj.
15.5 6 5.3 16.5 6 5.1 17.7 6 5.3 16.6 6 5.6 15.8 6 4.7 16.2 6 4.6 16.8 6 4.7 17.0 6 4.6 .012
Digits forward (8) Un. Adj.
5.6 6 2.1 6.5 6 1.9 5.8 6 1.6 6.2 6 1.7
7.1 6 2.0 6.9 6 1.9 6.4 6 1.7 6.4 6 1.7 .002
5.9 6 1.8 6.4 6 1.3 6.1 6 1.4 6.2 6 1.5
6.8 6 1.1 6.7 6 1.2 6.3 6 1.4 6.5 6 1.5 .016
Clock face (8) Un. Adj. Color trail 1 (25) Un. Adj.
20.4 6 5.4 21.9 6 4.4 23.1 6 3.2 22.5 6 3.8 20.9 6 4.5 21.4 6 21.5 21.5 6 4.4 22.2 6 4.4 .025
Color trail 2 (25) Un. Adj.
12.1 6 4.7 13.6 6 4.5 15.5 6 4.6 14.8 6 4.4 12.5 6 4.3 13.0 6 4.1 14.0 6 4.2 14.3 6 4.2 .001
GDS (15) Un. Adj.
2.7 6 2.9 1.6 6 1.9 2.4 6 1.5 1.8 6 2.4
0.9 6 1.3 1.5 6 2.2 1.5 6 2.4 1.7 6 2.4 .005
Percent (Odds Ratio) History of depression 31.6% (1.0) 18.5% (.60) 10.1% (.33) 11.5% (.36) .001 Taking antidepressants 6.2% (1.0) 4.3% (.62) 4.3% (.57) 2.1% (.29) .040 GDS . 6 10.3% (1.0) 3.0% (.39) 1.4% (.23) 6.3% (.73) .226 Notes: The numbers in parentheses are the highest possible scores for that test. Also shown is p for trend for the adjusted scores. At the bottom of the table, percentages of participants with a history of depression, currently on antidepressants, and with a Geriatric Depression Scale (GDS) score . 6 are shown by alcohol intake with odds ratios adjusted for sex, ethnicity, age, and years of education obtained by logistic regression analysis in parentheses. MMSE ¼ Mini-Mental State Examination.
values , 1.0 with five of the nine cognitive tests showing statistically significant trends by drinking category (Table 3). As examples, 11.2% of the participants had an MMSE score (without the intersecting pentagons added in) , 24, clinically recognized as an appropriate cut-point to indicate cognitive impairment. The OR values for Group 1
LINDEMAN ET AL.
1068
Table 3. Odds Ratios for Each Participant To Be Below a Cut-Point for Each Cognitive Test and .6 on the Geriatric Depression Scale (GDS) by Drinking Category Adjusted for Sex, Ethnicity, Age, and Years of Education Alcohol Intake Test No. of participants MMSE (,24/29) Draw pentagon (,7/10) Fuld–No. retrieved (,5/10) Flud–Total recalled (,6/10) Fuld–No. of names (,12/—) Digits forward (,4/8) Clock face (,4/8) Color trail 1 (,16/25) Color trail 2 (,10/25) GDS (.6/15)
Never
1 Day/Wk
2–5 Days/Wk
Daily, ,4
p for Trend
447 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00
234 0.50 0.64 0.49 0.73 0.75 0.46 0.59 0.63 0.66 0.36
69 0.19 0.96 0.83 0.42 0.78 0.34 1.01 0.41 0.35 0.21
96 0.34 0.83 0.63 0.78 0.53 0.48 0.39 0.32 0.32 0.62
.005 .373 .198 .178 .036 .017 .115 .002 .001 .080
Notes: The nondrinkers are the referent group; each drinker category compares those drinkers to nondrinkers. The numbers in parentheses show the cutpoint selected and the maximum score for each cognitive test and the Geriatric Depression Scale (GDS). The p value is for trend. MMSE ¼ Mini-Mental State Examination.
(abstainers) through Group 4 (daily drinkers) were 1.0, 0.50, 0.19, and 0.45, respectively, with a p for trend of .005. Similarly, 23.4% of the participants had scores ,10 on the Color Trails 2 test with OR values of 1.0, 0.66, 0.35, and 0.32, respectively, with a p for trend of .001.
DISCUSSION Those participants in the NMEHS who ingested alcohol had significantly better test scores than did the abstainers on seven of nine cognitive function tests after adjusting for differences in sex, ethnicity, age, and years of education. Because individuals with incipient dementia would have low scores that would skew the means on each cognitive test, those participants with MMSE scores , 18 were examined separately. All 30 participants with low MMSE scores were in the groups of abstainers or those who drank one or fewer times per week; none of those who drank two or more times per week had low MMSE scores. Similarly, the odds for having a low cognitive test score were consistently lower in the drinkers than in the abstainers. There was a consistent trend that, as drinking frequency increased, cognitive test scores became higher, and were less likely to fall below the selected cut-point for that test. Hispanics had lower cognitive test scores and less formal education than did NHWs. They also were more likely to abstain from drinking, and when they did drink, they drank less frequently. Regression analyses were used adjusting for differences in ethnicity, sex, age, and level of education to examine the differences in cognitive test scores between drinkers and abstainers. The drinkers had less evidence of depression by history and affective testing. Because depression can impair cognitive function, the regression equations were rerun adding the GDS scores as a covariable. The drinkers continued to score higher on all cognitive function tests.
A broad battery of cognitive tests was used to insure that sufficient cognitive domains were studied. Except for the MMSE, the tests were not dependent upon language skills. This was important because many of our Hispanic participants used Spanish as their primary language. Even after providing a choice of English or Spanish versions on all tests, the Hispanics scored lower on the MMSE and on the other tests as well. Our experience in this population using the MMSE and the Fuld Object-Memory Evaluation, the latter not requiring language skills, is described elsewhere (21). We feel the common criticisms of small sample size, potential for patient selection bias, and the failure of cognitive testing to cover a diversity of cognitive domains have been adequately addressed in this study. Two prospective studies have shown a lower rate of incident dementia in drinkers compared to abstainers (11,12) consistent with our findings of a higher percentage of individuals with low cognitive test scores in the abstainer and infrequent drinker categories compared to the more frequent drinkers. In the Rotterdam study involving participants age 55 years and older (11), light to moderate drinking (1–3 drinks/day) was associated with a lower risk of incident dementia compared to abstainers (OR, 0.46; 95% CI, 0.27–0.77). In the Cardiovascular Health Study involving individuals age 65 years and older (12), light to moderate drinking (1–3 drinks/day) was similarly associated with a lower risk of new onset dementia (OR, 0.58; 95% CI, 0.38–0.90). Our cross-sectional observation that drinkers had better cognitive test scores than abstainers also is consistent with what others have reported (6–10). Although several of these studies have reported that the differences between women drinkers and abstainers were more impressive than those observed in men (8,9), no difference between men and women was observed in the NMEHS. A previous report (15) from the NMEHS found that drinkers also had less evidence of coronary artery disease (CAD) than abstainers, a finding reported by others (22–24). One risk factor contributing to this lower incidence of CAD in drinkers is a higher mean high-density lipoprotein cholesterol concentration (23,24), a finding confirmed by our study (15). Associations between lower levels of highdensity lipoprotein cholesterol and the presence of both Alzheimer’s dementia (25) and cognitive impairment (26) have been described to be at least partly due to an increase in atherosclerotic disease. Drinkers in our study had a lower prevalence of diabetes compared to abstainers after adjusting for differences in ethnicity, sex, and age (15), similarly reported by others (27–29). Finally, the drinkers in our study were found to have significantly lower serum insulin concentrations than abstainers after adjustments for differences in ethnicity, sex, age, diabetic status, and evidence of obesity (body mass index) (30). This too has been reported by others (31–33) suggesting that drinkers are more insulin sensitive than are abstainers. The drinkers in our study weighed significantly less than the abstainers did, making the last adjustment important. Hyperinsulinemia has been described as an independent risk factor for CAD (34). A recent review published in this journal (35) constructs and supports a hypothesis that
COGNITIVE FUNCTION IN DRINKERS VERSUS ABSTAINERS
insulin resistance with hyperinsulinemia and subsequent impairment of glucose metabolism may promote neurodegeneration and facilitate the onset of Alzheimer’s dementia. These findings all provide evidence that drinking in moderation may be protective against the development of CAD, diabetes, and presumably other atherosclerotic diseases. Dementia and cognitive decline have been positively correlated with the atherosclerotic diseases (36). Mukamal and colleagues (37) found on magnetic resonance imaging (MRI) of the brain that moderate alcohol consumption was associated with a lower prevalence of white matter abnormalities and infarcts, thought to be of vascular origin, than their prevalence in abstainers. Our study population is unique in that all participants were Medicare recipients, therefore 65 years of age or older, with a mean age of 73.7 years. Most previous studies have examined younger cohorts. Furthermore, nearly equal numbers of Hispanics and NHWs were randomly selected and recruited from the community. Only one previous study involved minorities (blacks) (9). Concerns are that information is available on alcohol intake at only one point in time, and that no information is available on who might have stopped drinking because of health concerns (diabetes, heart disease, hypertension). Another concern is the disproportionately small number of daily drinkers in the Hispanic and adult female populations compared to NHW and adult male populations. We feel that the use of linear and logistic regression analyses adjusting for differences in ethnicity, sex, age, and years of education have responded adequately to this concern. Because of the cross-sectional design of the present study, no temporal relationship between alcohol intake and cognitive decline, which would be needed to show this negative association to be causal, could be demonstrated. The results of our study, however, are consistent with the results of two previously described prospective studies (11,12) showing that the incidence of new onset dementia is lower in drinkers than in abstainers. Nevertheless, these cross-sectional findings should be followed up with further large-scale prospective studies looking at cognitive function decline rather than just onset of new dementia. One may not be able to support a recommendation for older persons to initiate alcohol consumption in moderation as a way to prevent cognitive decline, diabetes, and the atherosclerotic diseases, but it does appear appropriate to discuss the potential benefits with those comfortable with a lifestyle that includes moderate alcohol consumption along with other preventative interventions, e.g., smoking cessation, increased exercise, weight and stress reduction, decreased dietary saturated fat and cholesterol intake, and control of hypertension and dyslipidemia with medications. ACKNOWLEDGMENTS This work was supported by grants from the National Institute on Aging (R01AG10941) and University of New Mexico General Clinical Research Center (GCRC) (National Center for Research Resources–GCRC Grant M01RR0997). Address correspondence to Robert D. Lindeman, MD, Aging and Genetic Epidemiology Program, University of New Mexico, Surge Bldg., Room 215, 2701 Frontier Place, N.E., Albuquerque, NM 87131-5666. E-mail:
[email protected]
1069
REFERENCES 1. Parker DA, Parker ES, Brody JA, Schoenberg R. Alcohol use and cognitive loss among employed men and women. Am J Public Health. 1983;73:521–526. 2. Parker ES. Alcohol and cognition. Psychopharmacol Bull. 1984;20: 494–496. 3. Carey KB, Maisto SA. Effect of a change in drinking pattern on the cognitive function of female social drinkers. J Stud Alcohol. 1987; 48:236–242. 4. Dent OF, Sulway OR, Broe GA, et al. Alcohol consumption and cognitive performance in a random sample of Australian soldiers who served in the Second World War. BMJ. 1997;314:1655–1657. 5. Goodwin JS, Sanchez CJ, Thomas CJ, Garry PJ. Alcohol intake in a healthy elderly population. Am J Public Health. 1987;77:173–177. 6. Hebert LE, Scherr PA, Beckett LA, et al. Relation of smoking and low to moderate alcohol consumption to change in cognitive function: a longitudinal study in a defined community of older persons. Am J Epidemiol. 1993;137:881–891. 7. Hendrie HC, Gau S, Hall KS, Hui SL, Unverzagt FW. The relationship between alcohol consumption, cognitive performance, and daily functioning in an urban sample of older Black Americans. J Am Geriatr Soc. 1996;44:1158–1165. 8. Dufouil C, Ducimetiere P, Alperovitch A. Sex differences in the association between alcohol consumption and cognitive performance. Am J Epidemiol. 1997;146:405–412. 9. Elias PK, Elias MF, D’Agostino RB, Silbershatz H, Wolf PA. Alcohol consumption and cognitive performance in the Framingham Heart Study. Am J Epidemiol. 1999;50:580–589. 10. Galanis DJ, Joseph C, Masaki KH, Petrovitch H, Ross GW, White L. A longitudinal study of drinking and cognitive performance in elderly Japanese American men: the Honolulu-Asia Aging Study. Am J Public Health. 2000;90:1254–1259. 11. Ruitenberg A, van Swieten JC, Witteman JC, et al. Alcohol consumption and risk of dementia: the Rotterdam Study. Lancet. 2002; 359:281–286. 12. Mukamal KJ, Kuller LH, Fitzpatrick AL, Longstreth WT Jr, Mittleman MA, Siscovick DS. Prospective study of alcohol consumption and risk of dementia in older adults. JAMA. 2003;289:1405–1413. 13. Lindeman RD, Romero LJ, Hundley R, et al. Prevalences of type 2 diabetes, the insulin resistance syndrome, and coronary heart disease in an elderly, biethnic population. Diabetes Care. 1998;21:959–966. 14. Romero LJ, Lindeman RD, Hundley R, et al. Outcome of recruitment and report on participation rate in the New Mexico Elder Health Survey. Ethn Dis. 1998;8:350–359. 15. Lindeman RD, Romero LJ, Allen AS, et al. Alcohol consumption is negatively associated with the prevalence of coronary heart disease in the New Mexico Elder Health Survey. J Am Geriatr Soc. 1999;47:396–401. 16. Lindeman RD, Romero LJ, La Rue A, et al. A biethnic community survey of cognition in participants with type 2 diabetes, impaired glucose tolerance, and normal glucose tolerance. The New Mexico Elder Health Survey. Diabetes Care. 2001;24:1567–1572. 17. Lindeman RD, Schade DS, La Rue A, et al. Subclinical hypothyroidism in a biethnic, urban community. J Am Geriatr Soc. 1999;47:703–709. 18. Lindeman RD, Romero LJ, Koehler KM, et al. Serum vitamin B12, C and folate concentrations in the New Mexico Elder Health Survey: correlations with cognitive and affective functions. J Am Coll Nutr. 2000;19:68–76. 19. La Rue A, Romero LJ, Ortiz IE, Liang HC, Lindeman RD. Neuropsychological performance of Hispanic and non-Hispanic older adults: an epidemiologic survey. Clin Neuropsychol. 1999;13:474–486. 20. SAS Institute: SAS User’s Guide: Basics. Version 6, 10th ed. Cary, NC: SAS Institute; 1995. 21. Ortiz IE, La Rue A, Romero LJ, Lindeman RD. Comparison of cultural bias in two cognitive screening instruments in elderly Hispanic patients in New Mexico. Am J Geriatr Psychiatry. 1997;5:333–338. 22. Gaziano JM, Buring JE, Breslow JI, et al. Moderate alcohol intake, increased levels of high-density lipoprotein and its subfractions, and decreased risk of myocardial infarction. N Engl J Med. 1993;329: 1829–1834. 23. Rimm EB, Klatsky A, Grobbee D, Stampfer MJ. Review of moderate alcohol consumption and reduced risk of coronary artery disease: is the effect due to beer, wine, or spirits? BMJ. 1996;312:731–736.
1070
LINDEMAN ET AL.
24. Thun MJ, Peto R, Lopez AD, et al. Alcohol consumption and mortality among middle-aged and elderly U.S. adults. N Engl J Med. 1997; 337:1705–1714. 25. Adunsky A, Chesnin V, Davidson M, Gerber Y, Alexander K, Haratz D. A cross-sectional study of lipids and ApoC in Alzheimer’s patients with and without cardiovascular disease. J Gerontol A Biol Sci Med Sci. 2002;57:M757–M761. 26. Van Exel E, de Craen AJ, Gussekloo J, et al. Association between highdensity lipoprotein and cognitive impairment in the oldest old. Ann Neurol. 2002;51:716–721. 27. Ajani UA, Hennekens CH, Spelsberg A, Manson JE. Alcohol consumption and risk of type 2 diabetes mellitus among US male physicians. Arch Intern Med. 2000;160:1025–1030. 28. Hu FB, Manson JE, Stampfer MJ, et al. Diet, lifestyle, and the risk of type 2 diabetes mellitus in women. N Engl J Med. 2001;345: 790–797. 29. Conigrave KM, Hu BF, Camargo CA Jr, Stampfer MJ, Willett WC, Rimm EB. A prospective study of drinking patterns in relation to risk of type 2 diabetes among men. Diabetes. 2001;50:2390–2395. 30. Kenkre PV, Lindeman RD, Yau CL, Baumgartner RN, Garry PJ. Serum insulin concentrations in daily drinkers compared with abstainers in the New Mexico Elder Health Survey. J Gerontol Med Sci. 2003;58A:960–963. 31. Mayer EJ, Newman B, Quesenberry CP Jr, Friedman GD, Selby JV. Alcohol consumption and insulin concentrations: role of insulin
32. 33.
34. 35. 36. 37.
associations of alcohol intake with high-density lipoprotein cholesterol and triglycerides. Circulation. 1993;88:2190–2197. Kiechl S, Willeit J, Poewe W, et al. Insulin sensitivity and regular alcohol consumption: large, prospective, cross sectional population study (Bruneck study). BMJ. 1996;313:1040–1044. Bell RA, Mayer-Davis EJ, Martin MA, D’Agostino RB Jr, Haffner SM. Associations between alcohol consumption and insulin sensitivity and cardiovascular risk factors: the Insulin Resistance and Atherosclerosis Study. Diabetes Care. 2000;23:1630–1636. Despres JP, Lamarche B, Mauriege P, et al. Hyperinsulinemia as an independent risk factor for ischemic heart disease. N Engl J Med. 1996;334:952–957. Rasgon N, Jarvik L. Insulin resistance, affective disorders, and Alzheimer’s disease: review and hypothesis. J Gerontol Med Sci. 2004; 59A:178–183. Prince MJ. Vascular risk factors and atherosclerosis as risk factors for cognitive decline and dementia. J Psychosom Res. 1995;39:525–530. Mukamal KJ, Longstreth WT Jr, Mittleman MA, Crum RM, Siscovick DS. Alcohol consumption and subclinical findings on magnetic resonance imaging of the brain in older adults: the cardiovascular health study. Stroke. 2001;32:1939–1946.
Received February 23, 2004 Accepted July 6, 2004 Decision Editor: John E. Morley, MB, BCh
