Rethinking Your Drinks: How Alcohol Affects Long-Term Brain Health

What’s this about?

This study asks whether drinking alcohol affects your risk of developing dementia, and if so, whether “light” or “moderate” drinking might actually be protective (as some earlier studies implied). The researchers undertook a large investigation to get a clearer answer, combining traditional observational data with genetic methods.

The key message: While older observational studies suggested that light drinking might be protective, this new work raises serious doubts about that and indicates that any level of alcohol consumption may increase dementia risk. (PubMed)

When and Who?

This research draws on two large datasets:

• The US Million Veteran Program (MVP) and the UK Biobank, with a combined observational cohort of about 559,559 adults aged 56‑72 at baseline (mean follow‑up ~4 years in the US dataset, ~12 years in the UK dataset).
  
• Genetic (Mendelian randomisation) analyses using summary data from about 2.4 million participants in multiple genome‑wide association studies (GWAS). 

So, this is recent work (2025) with huge sample sizes, both observational and genetic, and includes diverse ancestry groups. 

What did they find?

The findings come in two parts (observational + genetic), and they tell somewhat different stories.

Observationally (self‑reported alcohol use):

In the combined cohorts, ~14,540 participants developed dementia and ~48,034 died during follow‑up. 

Their results show a U‑shaped association: people who reported no drinking, and those who reported heavy drinking (e.g., >40 drinks per week), and those with a diagnosed alcohol use disorder (AUD) had higher dementia risk compared with light drinkers. For example, heavy drinkers had a hazard ratio (HR) ~1.41 (95% CI 1.15–1.74) compared to light drinkers. AUD cases had HR ~1.51 (95% CI 1.42–1.60).

This pattern (seemingly lower risk among light drinkers) might suggest that moderate drinking is “protective”, but the authors caution that this pattern may be misleading.

Genetic (Mendelian randomisation) analyses:

When using genetic proxies (variants associated with higher alcohol consumption or with AUD) to approximate “lifetime” exposure and reduce bias from confounding/reverse causation, they found that the more genetically‑predicted alcohol use, the higher the dementia risk. Specifically, a 1 standard deviation increase in log‑drinks per week was associated with a 15% increased odds of dementia (OR 1.15, 95% CI 1.03–1.27). Also, a doubling in genetically predicted AUD prevalence was associated with ~16% increase in dementia risk (OR 1.16, 95% CI 1.03–1.30). 

Importantly, in the genetic non‑linear analyses there was no evidence of a “dip” or protective effect at low levels of alcohol intake, risk steadily increased with more consumption.

They also show evidence that people who later developed dementia tended to reduce their drinking over time, suggesting “reverse causation” (i.e., declining alcohol use is a consequence of early brain decline) might partly explain why non‑drinkers sometimes show higher risk in observational studies. 

The observational data suggest a U‑shape (non‑drinkers and heavy drinkers at higher risk, light/moderate at lower), but the genetic data suggest no safe threshold: any increase in alcohol intake is higher dementia risk.

Why does it matter?

This research is important for several reasons. It challenges the long‑held public belief and some earlier research that “a little wine/beer now and then” might be good for brain health or even protective against dementia. This study suggests that narrative may be misleading. 

Because the sample sizes are huge, and the genetic methods help address biases (e.g., reverse causation, confounding) that plague many observational studies, the evidence is relatively strong and hard to ignore.

Dementia is a major public health challenge globally. If even light or moderate drinking contributes to risk, then alcohol consumption becomes one of the modifiable risk factors we should pay attention to. Indeed the authors estimate that reducing the prevalence of alcohol use disorder in the population could potentially reduce dementia incidence by up to 16%.

For individuals and public‑health policy makers alike, this suggests that the advice around “safe” drinking for brain health might need revisiting. Instead of “some alcohol might be fine”, the message may shift to “less is better, and zero may be safest” when thinking about long‑term cognitive health.

What am I doing about it?

Luckily, I’ve just never liked the taste of alcohol. For 30 years, I tried to convince myself to enjoy wine, but when I turned 50, I finally made peace with the fact that I simply don’t. Honestly, I’d rather sip on a blueberry chia shake than a glass of wine any day.

That said, I never say never. There are rare occasions when I might want a beer or feel like I need a glass in front of me for social reasons, but it usually ends after a few sips.

I’m very aware of the links between alcohol and diseases of aging, not to mention how it affects aesthetics, sleep, hormone detoxification and overall energy, so staying away comes pretty easily. But I know not everyone is wired the same way. It’s not always easy to quit or cut back. But it’s worth considering and learning how, if you’re struggling with this. 

As for those people in the Blue Zones who enjoy a drink and still live long, healthy lives, I don’t think it’s because of the wine. It’s the people they share it with, the community, the laughter, the connection. Alcohol just happens to be along for the ride. That’s just my opinion.

Practical tips

Here are some concrete suggestions you can adapt:

• Consider keeping a simple “drinks diary” for a few weeks: how many standard drinks you consume, on which days, and whether it’s for pleasure, habit, stress, socializing, etc.
• Listen to this great episode on gray area drinkers with Sara Rusbatch
  
• Set a realistic limit for yourself, for example, choose alcohol‑free days each week (say 2‑3) or reduce the average number of drinks per week by a defined percent.
  
• On social occasions, consider non‑alcohol alternatives you enjoy (sparkling water, mocktails, a flavourful tea). Building pleasurable rituals around non‑alcoholic options helps shift the “we always drink” habit.
  
• If you have any patterns of heavy drinking, or if you’ve used alcohol to cope with stress/anxiety, it may be worth discussing that with a health professional, because the link with dementia risk was particularly strong for those with AUD.
  
• Combine alcohol reduction with other brain‑health habits:
  
  • Prioritize sleep (quality + duration). Good sleep supports brain recovery and reduces dementia risk.
    
  • Engage in regular physical activity (piilates, spin, weight lifting, perfect). Strength and aerobic exercise both support brain and vascular health.
    
  • Stimulate your brain: reading physical books, engaging in new skills/hobbies, socializing, all build cognitive reserve.
    
  • Eat brain‑friendly foods: a diet rich in vegetables, fruit, lean protein, healthy fats (e.g., Mediterranean‑style) supports brain health and may buffer other risks.
    
  • Manage other health risks: e.g., cardiovascular health, diabetes, hearing loss, social isolation, all are relevant for dementia risk.
    
• Finally: Keep in mind that changing habits takes time. If you dislike the idea of “never drinking”, start small: reduce frequency or quantity, and reflect on how you feel physically, mentally, socially as the weeks go by.