A major new study published in BMJ Evidence-Based Medicine has found that alcohol consumption, including moderate drinking, may increase the risk of all-cause dementia. The research analyzed data from more than half a million adults enrolled in the U.S. Million Veteran Program (MVP) and the UK Biobank (UKB), combining observational, longitudinal, and genetic approaches to investigate the link between alcohol use and cognitive decline.
Alcohol affects brain neurons in multiple ways. It alters cell membranes, ion channels, enzymes, and receptors, directly interacting with those for acetylcholine, serotonin, GABA, and glutamate (NMDA receptors). Its sedative effects stem from activating GABA and glycine receptors while inhibiting NMDA receptor function. Outside the brain, ethanol can act as an anti-infective by disrupting osmotic balance across cell membranes.
Observational analyses initially suggested a “U-shaped” relationship i.e. heavy drinkers and people with alcohol use disorder (AUD) had the highest dementia risk, while moderate drinkers appeared somewhat protected. However, further investigation revealed that this apparent benefit likely results from reverse causation, where individuals reduce drinking as cognitive decline begins.
The study followed 559,559 adults (average age 56–72 years) for up to 12 years, during which there were 14,540 dementia cases and 48,034 deaths. Observational data showed higher dementia risk among non-drinkers, heavy drinkers consuming more than 40 drinks per week, and those with AUD, while moderate intake sometimes appeared protective. Longitudinal MVP data suggested that future dementia cases reduced alcohol consumption prior to diagnosis, explaining the misleading protective effect seen in some observational studies.
Genetic analyses contradicted the U-shaped pattern, revealing a monotonic increase in dementia risk with higher alcohol consumption. A one standard deviation increase in alcohol intake raised dementia odds by 15%, while doubling AUD prevalence increased risk by 16%. Non-linear Mendelian randomization analyses confirmed that even low levels of drinking provide no protective effect, with risk rising steadily. Negative-control checks supported the robustness of the findings.
While the results were most pronounced in participants of European ancestry, consistent patterns were observed across all groups studied. The research underscores that AUD is strongly associated with higher dementia risk, and no level of alcohol intake is entirely safe.
Lead researcher Zheng et al. concluded that alcohol exposure, measured as drinks per week, problematic alcohol use, or AUD, causally increases dementia risk in current drinkers. Observational U-shaped associations likely reflect reverse causation and confounding, rather than any true protective effect of moderate drinking.
For individuals, these findings challenge the longstanding notion that moderate drinking is beneficial for brain health. For clinicians and policymakers, reducing AUD prevalence could significantly reduce the burden of dementia, with estimates suggesting that halving AUD could prevent up to 16% of cases. Experts emphasize the need for clear public messaging that highlights dementia risk at all levels of alcohol consumption, particularly in older adults.
References
- Topiwala, A., Levey, D. F., Zhou, H., Deak, J. D., Adhikari, K., Ebmeier, K. P., Bell, S., Burgess, S., Nichols, T. E., Gaziano, M., Stein, M., & Gelernter, J. (2025). Alcohol use and risk of dementia in diverse populations: evidence from cohort, case–control and Mendelian randomisation approaches. BMJ Evidence-Based Medicine.
- Ethanol: Uses, Interactions, Mechanism of Action | DrugBank Online [Internet]. [cited 2025 Sept 29]. Available from: https://go.drugbank.com/drugs/DB00898
- Zheng L, Liao W, Luo S, Li B, Liu D, Yun Q, Zhao Z, Zhao J, Rong J, Gong Z, Sha F, Tang J. Association between alcohol consumption and incidence of dementia in current drinkers: linear and non-linear mendelian randomization analysis. EClinicalMedicine. 2024 Sep 5;76:102810.