Diabetes Obes MetabBrain & NeurologicalCohortNovember 27, 2025

Comparative effectiveness of SGLT2 sodium-glucose cotransporter-2 inhibitors and GLP-1 glucagon-like peptide-1 receptor agonists for incident dementia: A retrospective multicohort study.

Anagnostakis F, Kokkorakis M, Nagarajan S, Anastasiou G, Mantzoros CS

Key Finding

After following 32,542 people with type 2 diabetes for over 6 years, researchers found no difference in dementia risk between those taking GLP-1 drugs versus SGLT2 inhibitors - both groups had about 3.5% develop dementia.

What This Study Found

Think of this study like comparing two different GPS routes to the same destination - researchers wanted to see if taking GLP-1 receptor agonists (like Ozempic, Wegovy) versus SGLT2 inhibitors (like Jardiance, Farxiga) led to different rates of dementia in people with type 2 diabetes. They followed over 32,000 people for a median of 6.3 years, carefully matching patients so the comparison would be fair - like making sure both groups had similar ages, health conditions, and other medications. The results were essentially a tie: 581 people (3.6%) in the GLP-1 group developed dementia compared to 572 people (3.5%) in the SGLT2 group. This is like two marathon runners crossing the finish line at almost exactly the same time - there's no meaningful winner. The researchers also looked specifically at vascular dementia (caused by blood vessel problems) and Alzheimer's disease, finding the same pattern of no significant differences between the two drug classes.

Statistics Decoded

Hazard ratio of 1.01 (95% CI: 0.90-1.13): This means GLP-1 users had essentially the same risk of dementia as SGLT2 users - it's like saying both groups ran the race at identical speeds. The confidence interval crossing 1.0 means we can't rule out that either drug could be slightly better. With 581 vs 572 cases out of over 16,000 in each group, we're looking at about 3.6% vs 3.5% developing dementia over 6+ years. The median 6.3-year follow-up means half the participants were followed longer than this, half shorter - giving us substantial time to see dementia develop.

Why This Matters

This reassures doctors and patients that neither major diabetes drug class appears to increase dementia risk compared to the other, which is crucial since diabetes itself raises dementia risk and millions take these medications long-term.

Original Abstract

This study provides real-world evidence on the comparative effectiveness of glucagon-like peptide-1 receptor agonists (GLP-1 RAs) and sodium-glucose cotransporter-2 inhibitors (SGLT2i) regarding the risk of incident dementia in adults with type 2 diabetes mellitus (T2DM). This cohort study utilised electronic health records from TriNetX (1 April 2013 to 31 December 2019). Adults with T2DM on metformin who initiated GLP-1 RAs or SGLT2i were analysed using Cox proportional hazards models in 1:1 propensity score-matched cohorts. The primary outcome was all-cause dementia; secondary outcomes included vascular and Alzheimer's dementia. Confounding variables were adjusted. Among 16 271 participants using GLP-1 RAs and 16 271 using SGLT2i over a median 6.3-year follow-up, 581 GLP-1 RA users and 572 SGLT2i users developed dementia. No significant differences were observed in all-cause dementia (hazard ratio: 1.01, 95% confidence interval: 0.90-1.13), vascular, or Alzheimer's dementia. Subgroup analyses confirmed these findings. SGLT2i use was not associated with increased risk of dementia compared with GLP-1RAs in patients with T2DM. Further research is warranted to explore the effects of (newer) incretin-based (co-)agonists on cognitive outcomes.