Overview

• Published in BMJ Evidence-Based Medicine, the study pooled observational data from 559,559 adults in the UK Biobank and US Million Veteran Program with genetic evidence from about 2.4 million people.
• Observational results showed the familiar U-shaped curve, yet genetic analyses found no protective effect at low intake and indicated risk rises steadily as lifetime exposure increases.
• Genetically predicted increases of just 1–3 drinks per week were associated with roughly a 15% higher lifetime dementia risk, and doubling genetic liability to alcohol dependence with about a 16% higher risk.
• Across the cohorts, 14,540 people developed dementia; in observational comparisons, non-drinkers and very heavy drinkers (≥40 drinks per week) had a 41% higher risk than light drinkers (<7 per week), and alcohol dependence was linked to a 51% higher risk.
• Researchers observed that people later diagnosed with dementia often reduced drinking before diagnosis, supporting reverse causation, and they urged alcohol-reduction strategies while noting limits including European-ancestry dominance, self-reported intake, and assumptions in genetic methods.