In the news: is increased alcohol intake associated with reduced dementia risk?

Last Updated : 07 October 2022
Table of contents

    Recent news stories reported that alcohol may cut the risk of dementia, stating that ‘beer IS good for you!’ and drinking two pints a day can ‘slash your risk.’ So, should we more regularly catch up with friends over pints or reach for a beer after a long day at work to boost our brain health? Before doing so, there are a few things to keep in mind when reading the headlines. 

    The study behind the headlines 

    The study behind the news was carried out by a group of international researchers from the so-called ‘Cohort Studies of Memory in an International Consortium (COSMIC)’ collaboration.1 Their aim was to synthesise international findings on the alcohol-dementia relationship, including representation from low- and middle-income countries.  

    The researchers compiled the results of 15 studies which had diverse representation, including the United States, Australia, Brazil, South Korea, Republic of Congo and several European countries. In total, 24,478 people were included (58% female) and had an average of 72 years at the start of the study. Most of the studies started in the 1990s or early 2000s with data collection on age, sex, smoking status and alcohol consumption (later converted into average grams of pure ethanol per day), as well as cholesterol, blood pressure, mental health and other illnesses for a selection of the 15 included studies. Based on the level of alcohol consumption, participants were categorised as abstainers (0 g alcohol per day), occasional alcohol use (<1.3 g/day), light-moderate alcohol use (1.3-24.9 g/day), moderate-heavy alcohol use (25-44.9 g/day) and heavy alcohol use (≥45 g/day). The researchers then followed these participants up over a period of one to two decades and measured cases of diagnosed dementia. After 10 years of follow-up, about 14% of all participants had developed dementia.  

    Table 1 - Amount of pure alcohol in various types of alcoholic drinks.2  

    Type of drink

    Typical serving size1

    Number of alcohol units2

    Single small shot of spirits (e.g., gin, rum, vodka, whisky, tequila and sambuca) 

    25 ml, ABV 40% 

    1 unit (8 g ethanol)

    Single large shot of spirits (e.g., gin, rum, vodka, whisky, tequila and sambuca) 

    35 ml, ABV 40%

    1.4 units (11.2 g ethanol)

    Small glass of red/white/rosé wine 

    125 ml, ABV 12%

    1.5 units (12 g ethanol)

    Bottle of lager/beer/cider 

    330 ml, ABV 5%

    1.7 units (13.6 g ethanol)

    Can of lager/beer/cider

    440 ml, ABV 5.5% 

    2.4 units (19.2 g ethanol) 

    Pint of lower-strength lager/beer/cider

    568 ml, ABV 3.6%

    2 units (16 g ethanol) 

    Standard glass of red/white/rosé wine 

    175 ml, ABV 12% 

    2.1 units (16.8 g ethanol) 

    Pint of higher-strength lager/beer/cider 

    568 ml, ABV 5.2%

    3 units (24 g ethanol)

    Large glass of red/white/rosé wine 

    250 ml, ABV 12% 

    3 units (24 g ethanol) 

    1 ABV (alcohol by volume) is a measure of the amount of pure alcohol as a percentage of the total volume of liquid in a drink. 
    2 One unit equals 10 ml or 8 g of pure alcohol, which is around the amount of alcohol the average adult can process in an hour.  

    The results showed that compared with abstainers, both the occasional drinkers and light-moderate drinkers had a 12% reduced risk of developing dementia, while the moderate-heavy and heavy drinkers had an even greater reduced risk of 38% and 19%, respectively. Within this association, the study has considered multiple factors such as age, sex and smoking status that could potentially influence the dementia outcome. Additional analysis considering education, BMI, depression and various illnesses such as stroke and diabetes found that the risk reduction slightly diminished. However, the same conclusion was still valid. The researchers note that although consistent evidence was found suggesting that abstinence from alcohol in later life was associated with increased dementia risk, ‘recommending abstainers to initiate drinking is not recommended.’  

    What to keep in mind when reading the study’s conclusions? 

    • The research does not provide firm evidence that alcohol is protective against dementia. 

    The 15 included studies were not designed to look specifically at the link between alcohol and dementia. They varied considerably in terms of included number of participants, population characteristics, study duration, alcohol and dementia assessments, and health and lifestyle factors assessed. Notably, the findings were inconsistent across global regions, with some studies demonstrating a protective effect from alcohol (e.g., sub-analysis of Oceania) and others not (e.g., sub-analysis of United States). Hence caution is needed when interpreting the pooled results, and as the study authors themselves highlight, the findings should be considered exploratory only. 

    • The effect by type of drink or type of dementia is not demonstrated.  

    The assessment of alcohol intake was highly diverse, with some studies questioning specific drinks and others overall alcohol intake; some assessing units, others glasses and others just drinking frequency. Additionally, since alcohol was self-reported, which is prone to under-reporting, and drinking patterns may change over lifetime, the reported grams of alcohol per day are a cautious estimate. The proportion of abstainers across studies ranged from 12.0% to 80.7%, while heavy drinkers ranged from 0.5% to 8.6%. It is unknown whether these discrepancies reflect actual regional and cultural differences or variable accuracy in assessment. Some media outlets have taken ‘2 pints per day’ from an estimated 32g alcohol content (in the moderate-heavy range), but given the caveats, the findings cannot be applied to beer or other alcohol. Likewise, due to inconsistent dementia assessments, it was not possible to assess whether risk may differ by specific type of dementia, such as Alzheimer’s.    

    • The analysed population may not be representative.  

    Almost 10,000 people had to be excluded from the analysis - the majority because of incomplete assessments. There were significant differences between participants and non-participants in terms of socio-demographics, health status and alcohol use. Had the full sample been analysed, the results may have been different, adding a further caveat as to whether the findings give a true representation of risk. 

    • The health risks from higher alcohol consumption may outweigh any potential benefits.  

    As the authors rightly acknowledge that there are various known risks from high alcohol consumption, including cardiovascular disease, certain cancers and neurological health. This may far outweigh any potential benefit from a small decrease in dementia risk. Again, the authors note ‘advising those who currently abstain to initiate drinking is not recommended.’  

    What do authorities say? 

    • The European Commission summarises the different policy recommendations for dementia prevention, including from the World Health Organisation, Alzheimer’s Disease International, European and UK countries.3 The advice includes improved diagnosis and management of cardiometabolic conditions, increased physical activity and obesity prevention, improved social support and recognition of loneliness and isolation, and information campaigns to increase education and awareness.  
    • Alzheimer’s Europe provides links to each country’s national dementia strategy.4 Some countries (e.g., Northern Ireland) specify alcohol as a potential risk factor, alongside others such as smoking, obesity and high blood pressure.5  
    • WHO emphasises that harmful alcohol use is associated with over 200 diseases and injury conditions, is responsible for 5% of the global disease burden, and has a causal relationship with mental and behavioural conditions.6  
    • Dementia Support UK and Dementia UK are among the organisations highlighting that excessive alcohol consumption can cause alcohol-related brain damage and potentially dementia.7,8  


    1. Mewton, L., Visontay, R., Hoy, N., Lipnicki, D. M., Sunderland, M., Lipton, R. B., ... & Cohort Studies of Memory in an International Consortium. (2022). The relationship between alcohol use and dementia in adults aged over 60 years: A combined analysis of prospective, individual‐participant data from 15 international studies. Addiction, 1-13. 
    2. NHS. (2021). Alcohol units. Retrieved from
    3. European Commission. (2020). Health Promotion and Disease Prevention Knowledge Gateway. Retrieved from
    4. Alzheimer Europe. (n.d.) National Dementia Strategies. Retrieved from 
    5. Alzheimer Europe (n.d.). Ireland. Retrieved from 
    6. World Health Organization. (2022). Alcohol. Retrieved from 
    7. Dementia UK. (n.d.). Alcohol related brain damage. Retrieved from 
    8. Dementia UK. (n.d.). Alcohol and dementia. Retrieved from