Critique 014 13 August 2010
Arntzen KA, Schirmer H, Wilsgaard T, Mathiesen EB. Moderate wine consumption is associated with better cognitive test results: a 7 year follow up of 5033 subjects in the Tromsø Study. Acta Neurol Scand 2010; Suppl 190:23-29.
BACKGROUND: The impact of moderate alcohol consumption on cognitive function and dementia is unclear. We examined the relationship between consumptionof different alcoholic beverages and cognitive function in a large population-based study.
METHODS: Subjects were 5033 stroke-free men and women who participated in a longitudinal population-based study in Tromsø, Norway. Alcohol consumption and other cardiovascular risk factors were measured at baseline and cognitive function was assessed after 7 years follow up with verbal memory test, digit-symbol coding test and tapping test.
RESULTS: Moderate wine consumption was independently associated with better performance on all cognitive tests in both men and women. There was no consistent association between consumption of beer and spirits and cognitive test results. Alcohol abstention was associated with lower cognitive performance in women.
CONCLUSIONS: Light-to-moderate wine consumption was associated with better performance on cognitive tests after 7 years follow up.
Findings from the present study: This study is based on more than 5,000 people in Tromsø, Norway, who had an average age of just over 58 years at baseline. The subjects were followed over 7 years when they were administered a battery of cognitive function tests. Among women, there was a lower risk of a poor testing score for those who stated that they consume wine at least 4 or more times over a 14-day period in comparison with those who drink < 1 time during this period. The highest risk of poor cognitive functioning tended to be among teetotalers. Among men, some test results suggest less cognitive dysfunction for consumers of wine and beer, although the clear dose-response pattern seen among women was not present for men. Since only the frequency of consumption, not the amount of alcohol consumed per day, was recorded, these differences by gender could relate to differing levels of drinking or other confounders.
The tests administered for cognitive function were those usually used: tests of short-term verbal memory, the Wechsler adult intelligence scale, and a tapping test for psychomotor tempo. The expected associations between other risk factors for poor cognitive functioning were seen: lower testing scores among people who were older, less educated, smokers, and those with depression, diabetes, or hypertension.
There are some potential concerns about methodology which were also expressed in the authors’ self critique of their research. It has long been known that “moderate people do moderate things.” The authors state the same thing: “A positive effect of wine . . . could also be due to confounders such as socioeconomic status and more favorable dietary and other lifestyle habits.” The sample also likely has some selection bias, as the authors state: “Some degree of selection is likely to have occurred due to lower participation rate among people with dementia at baseline.” Finally “sick quitters” are not accounted for and have been included with the teetotalers; as stated, “The abstainer group may differ from non-abstainers in lifestyle issues that we have not measured and may also include some former alcoholics.”
Previous research on the topic: In the last three decades, the association between moderate alcohol intake and cognitive function has been investigated in 68 studies comprising 145,308 men and women from various populations with various drinking patterns. Most studies showed an association between light to moderate alcohol consumption and better cognitive function and reduced risk of dementia, including both vascular dementia and Alzheimer’s Disease.
In many studies a protective effect against cognitive decline was found in association with a moderate consumption of red wine, but less so with other beverages. Truelsen et al1, using a nested case-control design within a population-based cohort study (the Copenhagen City Heart Study), describe a protective effect of moderate wine drinking on dementia of all types. A problem is that studies of this sort cannot adjust for every potential confounder. For example, Truelsen et al did not include dietary habits among covariates. The authors acknowledge that wine drinkers might have better dietary habits than beer and liquor drinkers; in fact, two studies from Denmark have concluded that wine drinking is associated with a healthy diet2,3.
The same association between wine preference and a healthy diet might well be present in Norway – a country with a tradition of schnapps binge-drinking during weekends. However, the modern Norwegian wine drinker in this study appears to have very moderate habits. The majority of participants in the present study4 reported low alcohol intake, and 20% of men and 30% of women drank <1 glass of alcohol per fortnight. The median intake of alcohol was one drink (range 0-34) per 14 days in women and three drinks (range 0-101) per 14 days in men. On average, only 1.9% of men and 0.2% of women drank >2 glasses of alcohol daily. A greater effect from wine could be attributed to specific favorable biological effects of wine, but could also be due to confounders, such as socioeconomic status and more favorable dietary and other lifestyle habits. Although Arntzen et al4 have adjusted for education, physical activity, and smoking habits, the authors had no information on diet, income, or profession.
As stated, in this study other types of alcoholic beverage showed less of an effect than did wine. It is likely that phenolic compounds present in wine may have contributed to the overall effect. However, it is never clearly stated which type of wine subjects tended to consume (red, white or both), or the origin of the wines. In general, white wines contain lower levels of polyphenols than do red wine; levels in white wines may be similar to the concentrations found in some beers. (The presence in beer of other phytochemicals coming from hops such as lupulone or humulone, may also contribute to its effects.) Further, levels of tannins and overall polyphenols in wine vary according to the origin of the wine and also the grapes used for its production.
Potential mechanisms: While the present study4 was an observational study, and potential residual confounding by other lifestyle factors is always a possibility, the results are concordant with many other prospective studies that suggest lower risk of cognitive dysfunction and dementia for moderate drinkers, especially of wine. Mechanisms of such an effect are unclear, but Collins et al5 found that the literature supports less cognitive decline to be associated from both alcohol and polyphenols. They state that such protection may relate to “cardiovascular and ⁄or hematological effects, but there is also experimental evidence that moderate alcohol levels can exert direct ‘neuroprotective’ actions — pertinent are several studies in vivo and rat brain organotypic cultures, in which antecedent or preconditioning exposure to moderate alcohol neuroprotects against ischemia, endotoxin, b-amyloid, a toxic protein intimately associated with Alzheimer’s, or gp120, the neuroinflammatory HIV-1 envelope protein5.”
Summary: A large prospective study of men and women in northern Norway reported that moderate wine consumption was independently associated with better performance on cognitive tests after 7 years of follow up. There was no consistent association between consumption of beer or spirits and cognitive test results. The authors also reported that abstinence was associated with significantly lower cognitive performance in women. As noted by the authors, in any observational study there is the possibility of residual confounding by other lifestyle habits affecting cognitive function, and the present study was not able to adjust for certain ones (e.g., diet, income, or profession) but did adjust for age, education, body mass index, depression, and cardiovascular disease and its major risk factors.
The results of this study support findings from a number of observational, prospective studies that have shown that the moderate consumption of alcohol, especially of wine, may have favorable effects on cognitive function. Such effects could relate to the presence in wine of a number of polyphenols and other substances that reduce the risk of cognitive decline from ageing. Mechanisms that have been suggested for such protection against cognitive dysfunction include effects on atherosclerosis, coagulation, inflammation, as well as direct neuroprotective effects.
Reference from Forum review:
- Truelsen T, Thudium D, Grønbæk M. Amount and type of alcohol and risk of dementia. The Copenhagen City Heart Study. Neurology 2002;59:1313-1319.
- Tjønneland A, Grønbæk M, Stripp C, Overvad K. Wine intake and diet in a random sample of 48,763 Danish men and women. Am J Clin Nutr 1999;69:49-54.
- Johansen D, Friis K, Skovenborg E, Grønbæk M. Food buying habits of people who buy wine or beer: cross sectional study. BMJ 2006;332:519-522.
- Arntzen KA, Schirmer H, Wilsgaard T, Mathiesen EB. Moderate wine consumption is associated with better cognitive test results: a 7 year follow up of 5,033 subjects in the Tromsø Study. Acta Neurol Scand Suppl 2010;190:23-29.
- Collins MA, Neafsey EJ, Mukamal KJ, Gray MO, Parks DA, Das DK, Korthuis RJ. Alcohol in moderation, cardioprotection, and neuroprotection: Epidemiological considerations and mechanistic studies. Alcohol Clin Exp Res 2009;33:206–219.
* * *
Contributions to these comments from the International Scientific Forum on Alcohol Research were from the following:
R. Curtis Ellison, MD, Section of Preventive Medicine & Epidemiology, Boston University School of Medicine, Boston, MA, USA
Ross McCormick PhD, MSC, MBChB, Associate Dean, Faculty of Medical and Health Sciences, The University of Auckland, Auckland, New Zealand
Erik Skovenborg, MD, Scandinavian Medical Alcohol Board, Practitioner, Aarhus, Denmark
Gordon Troup, MSc, DSc, School of Physics, Monash University, Victoria, Australia
David Vauzour, PhD, Dept. of Food and Nutritional Sciences, The University of Reading, UK