Researchers analyzed data from 12,000 adults with type 2 diabetes who received semaglutide, a GLP‑1 receptor agonist, and compared them to matched controls. Using a validated dementia‑prediction model, they estimated the likelihood of developing dementia over the next five years. The analysis found that participants on semaglutide had a 26% lower predicted risk of dementia compared with controls, a reduction that remained significant after adjusting for age, sex, education, baseline cognition, and comorbidities.

The study was observational and relied on electronic health records and pharmacy claims to identify treatment exposure and risk factors. The predicted outcomes were generated with a machine‑learning algorithm trained on population‑level data, allowing a large sample size but not direct measurement of incident dementia cases.

These findings suggest a potential neuroprotective effect of semaglutide, but the authors caution that the results are associative and not causal. They call for randomized controlled trials to confirm whether semaglutide can truly reduce dementia incidence, especially given its established benefits for cardiovascular health and weight management. The study was published in the journal Alzheimer's & Dementia in 2024, adding to a growing body of evidence linking metabolic therapies to brain health.