BMJ Evid Based Med

Even light drinking may raise dementia risk, large-scale study finds

September 26, 2025

Study details: This multi-method study combined observational cohort, case-control, and Mendelian randomization analyses to assess the relationship between alcohol use and dementia risk. Data were drawn from over 559,000 adults in the U.S. Million Veteran Programme and UK Biobank, with genetic data from 2.4 million individuals across 45 genome-wide association studies (GWAS) cohorts. Participants were aged 56 to 72 at baseline and followed for up to 12 years.

Results: Observational findings revealed a U-shaped curve, with both abstainers and heavy drinkers showing elevated dementia risk. Compared with light drinkers (<7 drinks/week), heavy drinkers (≥40 drinks/week) had a 41% increased risk, rising to 51% among those with alcohol use disorder (AUD). However, Mendelian randomization showed a linear, dose-dependent increase in dementia risk with genetically predicted alcohol use. A 1-SD increase in weekly drinks was associated with a 15% higher risk, and a twofold increase in AUD prevalence correlated with a 16% increase in dementia risk.

Clinical impact: These findings challenge the long-held belief that light-to-moderate alcohol consumption may be neuroprotective. Instead, they support a causal link between alcohol use and increased dementia risk, suggesting that reducing alcohol intake—even at low levels—may be a viable strategy for dementia prevention.

Source:

Topiwala A, et al. (2025, September 23). BMJ Evid Based Med. Alcohol use and risk of dementia in diverse populations: evidence from cohort, case-control and Mendelian randomisation approaches. https://pubmed.ncbi.nlm.nih.gov/40987604/