Koch M, Fitzpatrick AL, Rapp SR, Nahin RL, Williamson JD, Lopez OL, et al. Alcohol Consumption and Risk of Dementia and Cognitive Decline Among Older Adults With or Without Mild Cognitive Impairment. JAMA Network Open 2019:2(9):e1910319. doi: 10.1001/jamanetworkopen.2019.10319
IMPORTANCE: Substantial heterogeneity and uncertainty exist in the observed associations between alcohol consumption and dementia.
OBJECTIVE: To assess the association between alcohol consumption and dementia and the roles of mild cognitive impairment (MCI) and apolipoproteinEε4 (APOEE4) genotype in modifying this association.
DESIGN, SETTING, AND PARTICIPANTS: This cohort study used data from the Ginkgo Evaluation of Memory Study, conducted from 2000 to 2008 among US community-dwelling participants. This study analyzed 3,021 participants aged 72 years and older who were free of dementia. Data analysis was performed from 2017 to 2018.
EXPOSURES: Self-reported alcohol consumption, drinking frequency, and quantity
MAIN OUTCOMES AND MEASURES: Using multivariable proportional hazards regression and linear mixed models, the risk of dementia and the rate of change over time in the Modified Mini-Mental State Examination were estimated.
RESULTS: Among 3,021 participants, the median (interquartile range) age was 78 (76-80) years;1,395 (46.2%) were female. During a median (interquartile range) follow-up of 6.0 (4.9-6.5) years, 512 cases of dementia occurred. For 7.1 to 14.0 drinks per week compared with less than 1.0 drink per week, the hazard ratios for dementia were 0.63 (95% CI, 0.38-1.06) among 2,548 participants without MCI and 0.93 (95% CI, 0.47-1.84) among 473 participants with MCI. Among participants with MCI, the hazard ratio for dementia was 1.72 (95% CI, 0.87-3.40) for more than14.0 drinks per week compared with less than 1.0 drink per week. The association of alcohol intake with dementia differed for participants with and without baseline MCI (P for interaction=.03). Among participants without MCI, daily low-quantity drinking was associated with lower dementia risk than infrequent higher quantity drinking (hazard ratio, 0.45; 95% CI, 0.23-0.89; P=.02). Findings were consistent when stratified by sex, age, and APOEE4 genotype. Compared with drinking less than 1.0 drink per week, complete abstention (in participants without MCI) and the consumption of more than 14.0 drinks per week (in participants with MCI) were associated with lower Modified Mini-Mental State Examination scores (mean difference at follow-up compared with baseline, −0.46 point [95%CI,−0.87 to −0.04 point] and −3.51points [95% CI, −5.75 to −1.27 points], respectively).
CONCLUSIONS AND RELEVANCE: In this study, complete abstention and consuming more than 14.0 drinks per week (compared with drinking <1.0 drink per week) were associated with lower cognitive scores among participants aged 72 years and older. Particular caution is needed among individuals with MCI who continue to drink alcohol.
While many previous studies have shown that moderate drinkers appear to have a lower risk of developing dementia, Forum members consider this to be an important paper for several reasons:
- It focuses on elderly subjects, with a median age of 78 years, an age group with limited previous research and in which the development of dementia is common.
- It provides a very complete and careful attention to the assessment of cognitive function, as subjects underwent numerous and repeated cognitive assessments. For example, if there was any suggestion of cognitive impairment, cerebral magnetic resonance imaging and referral to an expert panel of clinicians, who reviewed and validated presence of dementia, was carried out.
- There was an assessment of both the amount and frequency of alcohol intake; also, the authors used HDL and APOA levels to validate reported alcohol intake (associations had p < 0.001 for both) and did sensitivity analyses such as testing for competing causes of death (12-13% of subjects died during follow up).
- There was a very broad assessment of potential confounders, including body weight, smoking, diabetes, cardiovascular and other diseases, depression, race, ethnicity, clinic site, educational level, social activity, medication use, and genotype for APOE-E4.
The key finding for the large majority of participants, those without mild cognitive impairment at baseline (n=2,548), was that complete abstention from alcohol was associated with a higher risk of dementia than seen in any of the groups consuming alcohol. For 473 subjects with MCI at baseline, there was no effect on risk of dementia for up to about 10 drinks/week, then a slight but insignificant increase in risk with greater intake. The authors state that the group found to have the highest risk of dementia was the group with MCI at baseline that consumed > 14 drinks/week. In most of the data, results were similar for all types of alcohol. Some of the associations were not statistically significant, but closely followed patterns shown in previous research.
The frequency of drinking was found to be important, as the authors stated that “Daily low-quantity drinking was associated with a lower dementia risk (HR=0.45, 95% CI 0.23-0.89) compared with infrequent higher-quantity drinking.” The authors also state: “The association of alcohol intake and ADAS-Cog scores was not modified by MCI at baseline.” Obviously, however, MCI at baseline increased the risk of dementia being diagnosed during follow up, regardless of alcohol intake. Forum members concluded that this well-done study strongly supports previous research indicating that the moderate intake of alcohol in elderly subjects is associated with a lower risk of developing dementia.
Specific comments by Forum members: Reviewer Finkel agreed with the overall favorable views of this paper by Forum members, but added: “My ire was aroused because the authors, on the title page of the article, instructed us that the meaning of the paper should be the following: ‘These findings suggest that physicians caring for older adults need to carefully assess the full dimensions of drinking behavior and cognition when providing guidance to patients about their alcohol consumption.’ While that may be correct, the meaning of the paper is what the reader takes from it, not what the author or editor instructs him/her to derive — the meaning of this paper is clearly not what the authors say it is! The meaning is that this paper supports a J-shaped curve for the relation of alcohol to cognition among older individuals.
“Is this not another example of authors/editors being afraid to admit that data supported drinking’s hormetic role, that alcohol in any form or dose might be beneficial, especially to mature individuals?” Other Forum members agreed that the key conclusion of this paper is that moderate drinking is not only not harmful to cognitive function, but appears to be associated with a reduced risk of dementia.
Reviewer Skovenborg agreed: “I share Finkel’s ire and resent the routine alcohol-benefit disclaimer you will typically find in papers published in American journals. I have been told that the disclaimers are necessary for the authors in order to protect future grants to finance their research. And that everybody, with knowledge of that being the case, routinely disregard the ‘disclaimer-speak’ and just read what the data support. This may not be a mature, evidence-based system but rather a system infected with political considerations; however, it seems to be a fact of today’s life in research.”
Forum member Goldfinger stated: “I agree that the findings in this paper are consistent with what has already been established. It speaks to the anti-atherosclerotic, anti-inflammatory, and antioxidant mechanisms of alcohol. Again, in the absence/limited significance of anti-oxidant activity of non-wine alcohol in comparison with red wine, and its bioactive non-alcohol components, we can presume that some of the negative effects particularly in the >14 drinks per week exclusive red wine consumers (at least not the most severe drinkers), may have continued benefits. This is, of course, an unproven thesis as beverage type was not evaluated very extensively.
“I also want to point out that under-reporting, certainly in place in any subjective survey of alcohol consumption, likely attenuates the negative effects in the > 14 per week crowd. I must suspect that those categorized in the < 14 drinks per week are really more robust consumers.”
Forum member Mcevoy agreed that the methodology in this study is sound, adding: “The findings among the non-MCI participants are as expected, with a protective association with moderate drinking. There are so few MCI participants, particularly in the higher drinking ranges, that those results must be viewed with caution until replicated in a larger sample.”
Reviewer Skovenborg commented: “I agree with the points of other Forum members, especially the very complete and careful attention to assessment of cognitive function (not seen in all studies), the looking into the importance of drinking pattern (not done in most other studies), the very thorough assessment of potential confounders (not done in many other studies), the choice of reference group (precludes the usual comments about non-drinkers), and the clear evidence of the association between alcohol intake and cognitive decline not being modified by APOE genotype (an ongoing discussion).
“What I am missing is an explanation why a moderate intake of alcohol is protective against dementia in older persons with intact cognition at baseline and not in participants with MCI at baseline. I also agree with Goldfinger that some underreporting is likely present in the group with 7.1 – 14 drinks per week. I find some inconsistency in the assessment of alcohol consumption: WINE: 6 oz. glasses of wine = 180 ml of wine = 20 g alcohol (with the 14.0 vol % alcohol that is common these days); BEER: 12 – oz cans of beer = 360 ml beer = 14.4 g alcohol with the normal alcohol strength of 5 vol %; SPIRITS: 1 shot = 1.5 oz = 45 ml = 14.4 g alcohol with the normal 40 vol % alcohol strength. It seems to me that the wine drinkers have a risk of having their alcohol consumption over-estimated in comparison with beer and spirits drinkers. For all practical purposes you have to be aware that the limit of 14 drinks per week (for moderate consumption) may be translated to 280 g alcohol as wine and 201.6 g alcohol (as beer and wine). The limit per week could be translated to 28 units of alcohol (as wine) and 20 units of alcohol (as beer and spirits) in countries with an alcohol unit = 10 g alcohol. In the UK the limit would be 35 units per week for wine drinkers and 25 units per week for beer drinkers.” Other Forum members emphasized how consumption of a beverage containing alcohol is very different when consumed with food, rather than on an empty stomach (which was not evaluated in this study).
Reviewer Mattivi wrote: “I agree with comments of others, with the additional comment that it would be necessary to consolidate the self-reported consumption with more robust biochemistry. The use of alcohol biomarkers together with the self-reported intake may improve the classification, providing a more detailed drinking history. A good example of proper application, validated in patients from psychiatric treatment units, can be found in the recent paper of the researchers at Faculty of Medicine, University of Tampere (Archer et al), who used combinations of GT, CDT, GT-CDT-combination, MCV, ALT, AST, hs-CRP, and IL-6 to assist in judging the quantity of alcohol intake among a group of depressed patients.”
Potential mechanisms for effects on dementia: Reviewer Stockley noted: “There are plausible biological mechanisms for both the positive effects on the brain’s structures and functions as well as for the negative effects of heavy consumption: for example, areas of the brain related to cognitive function where alcohol is observed to act include the hippocampus, superior frontal cortex and cerebellum. In particular, depending on the amount of alcohol in the brain, certain brain cell messaging is affected (neurotransmission). Brain grey and white matter volumes also associated with cognitive function have been seen in brain imaging studies in older adults to change with alcohol consumption in a j- or u-shaped relationship. There are also common neuro- and cardio-protective relationships for alcohol, where biological mechanisms that improve blood flow to reduce the risk of atherosclerosis and coronary heart diseases also reduce the risk of cognitive impairment, which may also partly explain the similarity in relationships between cognitive impairment/dementias and alcohol and cardiovascular diseases and alcohol.
“Although there is variation in methodology between studies assessing aspects of cognitive function and alcohol consumption, both current and over a lifetime, published reviews consistently suggest that, on balance, there is a j or u-shaped relationship between alcohol consumption and the risk of cognitive impairment or dysfunction and the development of dementias such as Alzheimer’s disease. For example, this j or u-shaped relationship has been observed despite beverage type, drinking patterns and follow-up periods, as well as demographics, genetics, and lifestyle factors such as smoking. Neafsey et al in 2011, for example, included 143 papers published between 1997 and 2011 in a meta-analysis. The meta-analysis showed a 23% reduction in risk of cognitive impairment and dementia for light to moderate consumers compared to non-drinkers, irrespective of whether former drinkers were included with lifetime abstainers. Furthermore, in studies that did not separate lifetime abstainers from former drinkers, the association that alcohol consumption has neuroprotective effects became stronger since former drinkers would be at an increased risk at baseline otherwise. Any alcohol consumption, however, predicted a 34% lower risk of cognitive impairment and dementia in a study of 2805 older individuals initially free of any cognitive impairment and followed for 16 years in Australia (Simons et al), where it was concluded that while excess alcohol consumption should be avoided, it appeared safe and reasonable to recommend the continuation of moderate alcohol consumption in those already imbibing. This current study provides some fine tuning on previous findings.”
References from Forum critique:
Archer M, Kampman O, Bloigu A, Bloigu R, Luoto K, et al. Assessment of alcohol consumption in depression follow-up using self-reports and blood measures including inflammatory biomarkers. Alcohol and Alcoholism 2019;54:243–250. doi: 10.1093/alcalc/agz002
Neafsey EJ, Collins MA. Moderate alcohol consumption and cognitive risk. Neuropsychiatr Dis Treat 2011;7:465–484.
Simons LA, Simons J, McCallum J, Friedlander Y. Lifestyle factors and risk of dementia: Dubbo Study of the Elderly. Med J Aust 2006;184:68-70.
This study was designed to assess the association between alcohol consumption and dementia and the roles of mild cognitive impairment (MCI) and apolipoproteinEε4 (APOEE4) genotype in modifying this association. It was based on data from the Ginkgo Evaluation of Memory Study, conducted from 2000 to 2008 among US community-dwelling participants (which did not demonstrate beneficial effects of ginkgo). The study analyzed 3,021 participants aged 72 years and older who were free of dementia at baseline.
The assessment of alcohol consumption included both the amount and frequency of intake, and the authors had a very broad assessment of potential confounders, including cardiovascular and other diseases, depression, race, ethnicity, educational level, social activity, medication use, and genotype for APOE-E4. There was a very complete and careful attention to the assessment of cognitive function, as subjects underwent numerous and repeated cognitive assessments. For example, if there was any suggestion of cognitive impairment, cerebral magnetic resonance imaging and referral to an expert panel of clinicians, who reviewed and validated presence of dementia, was carried out. No previous studies have had such an extensive assessment of cognition.
The main results include a finding that for the large majority of participants (the 2,548 without mild cognitive impairment at baseline), complete abstention from alcohol was associated with a slightly higher risk of dementia than seen in any of the groups consuming alcohol. For 473 subjects with MCI at baseline, there was no effect on risk of dementia for up to about 10 drinks/week, then a slight but insignificant increase in risk with greater intake. Daily low-quantity drinking was associated with more than 50% lower dementia risk than infrequent higher alcohol consumption. Many of the associations shown had a consistent pattern (a decrease in risk with light drinking) but did not reach statistical significance. The association was similar for all sub-groups, and was not modified by APOEE4 genotype.
The analyses seem well done, and Forum members agreed that, while some of the associations did not reach statistical significance, the data support most previous studies that show a J-shaped or U-shaped association for the relation of alcohol consumption to the risk of dementia. As has been shown in most previous studies, regular (even daily) light drinking was found to be preferable to infrequent higher levels of alcohol consumption in terms of the effect on the risk of dementia.
*** *** *** *** *** *** *** *** ***
This critique by the International Scientific Forum on Alcohol Research was based on comments provided by the following members:
Fulvio Mattivi, MSc, CAFE – Center Agriculture Food Environment, University of Trento, via E. Mach 1, San Michele all’Adige, Italy
Pierre-Louis Teissedre, PhD, Faculty of Oenology–ISVV, University Victor Segalen Bordeaux 2, Bordeaux, France
Giovanni de Gaetano, MD, PhD, Department of Epidemiology and Prevention, IRCCS Istituto Neurologico Mediterraneo NEUROMED, Pozzilli, Italy
Harvey Finkel, MD, Hematology/Oncology, Retired (Formerly, Clinical Professor of Medicine, Boston University Medical Center, Boston, MA, USA)
Linda Mcevoy, PhD, Department of Radiology, University of California at San Diego (UCSD), La Jolla, CA, USA
Creina Stockley, PhD, MSc Clinical Pharmacology, MBA; Adjunct Senior Lecturer at the University of Adelaide, Australia
Erik Skovenborg, MD, specialized in family medicine, member of the Scandinavian Medical Alcohol Board, Aarhus, Denmark
Tedd Goldfinger, DO, FACC, Desert Cardiology of Tucson Heart Center, University of Arizona School of Medicine, Tucson, AZ, USA
Imke Janssen, PhD, Department of Preventive Medicine, Rush University Medical Centre, Chicago, IL, USA.
Matilda Parente, MD, consultant in molecular pathology/genetics and emerging technologies, San Diego, CA, USA.
R. Curtis Ellison, MD, Professor of Medicine, Section of Preventive Medicine & Epidemiology, Boston University School of Medicine, Boston, MA, USA
Arne Svilaas, MD, PhD, general practice and lipidology, Oslo University Hospital, Oslo, Norway
Dag S. Thelle, MD, PhD, Department of Biostatistics, Institute of Basic Medical Sciences, University of Oslo, Norway; Section for Epidemiology and Social Medicine, Sahlgrenska Academy, University of Gothenburg, Sweden