Decades after a diabetes prevention trial, the metformin group is showing less dementia
STUDY: Wander PL et al, medRxiv preprint 2026
STUDY TYPE: Long-term follow-up of a randomized trial (secondary analysis)
FUNDING: National Institutes of Health
Background
Metformin is the standard first-line drug for type-2 diabetes. For high-risk adults, it cuts the diabetes risk in half. Whether it prevents dementia, though, has been unclear, with observational studies pointing opposite ways and often confounded by the reason it was prescribed.
This analysis used decades of follow-up from a randomized trial to test whether metformin assignment affected dementia risk and cognition years later.
Design
- 1,483 adults randomized to metformin, placebo, or lifestyle intervention in the 1996-1999 Diabetes Prevention Program, followed for cognitive outcomes 22 to 24 years later (median age 74).
- Although patients in all groups received metformin a some point, the original metformin group had far more total drug exposure (about 15 years) than placebo or lifestyle groups (under 5).
- Dementia and mild cognitive impairment were tested with a standard battery; memory and executive tests repeated over 15 years.
Results
Metformin use was associated with a 60-62% lower risk of dementia compared to placebo and lifestyle interventions (in adjusted analyses).
Metformin was also linked to modestly better memory scores over time, with no sign of harm.
Limitations
Outcomes uncorrected for multiple comparisons. Many participants crossed over between arms over decades, diluting the comparison. Dementia cases were few, and cognitive testing wasn’t done at baseline.
Practice Implications
- Metformin in pre-diabetes prevents long-term cognitive decline and dementia.
- The finding makes biological sense, as insulin resistance in the brain is linked to dementia.
—Chris Aiken, MD
Director, Psych Partners
Editor in Chief, Carlat Psychiatry Report
