Protective effect of metformin against dementia in patients with obesity: Results from a global federated health network analysis

Abstract Aims Metformin, a first‐line medication for Type 2 diabetes (T2D), has been suggested to reduce dementia risk. We investigated whether metformin use was associated with lower long‐term incidence of dementia and all‐cause mortality in obese patients. Materials and Methods We analysed electronic health record data from the TriNetX network. Patients were categorised into four body mass index (BMI) groups (25–29.9, 30–34.9, 35–39.9 and over 40). In each group, those prescribed metformin were compared with matched controls who were not prescribed metformin, using propensity score matching. Kaplan–Meier survival analysis was used to estimate dementia incidence and all‐cause mortality over a 10‐year follow‐up. Results The matched cohorts included 132 920 (BMI 25–29.9), 142 723 (30–34.9), 94 402 (35–39.9) and 82 732 (over 40) patients per group. After 10 years of follow‐up, metformin users exhibited significantly lower risks of both dementia and all‐cause mortality compared to controls. Specifically, the hazard ratios for dementia across each BMI group were 0.875 (95% confidence interval [CI]: 0.848–0.904), 0.917 (0.885–0.951), 0.878 (0.834–0.924) and 0.891 (0.834–0.953), respectively. For all‐cause mortality, the corresponding hazard ratios were 0.719 (0.701–0.737), 0.727 (0.708–0.746), 0.717 (0.694–0.741) and 0.743 (0.717–0.771). Conclusions In this large, multi‐centre cohort study, metformin use was associated with reduced risks of dementia and all‐cause mortality in obese patients. The protective effect was observed across all BMI groups, with variations noted by population. These findings support the potential of metformin in lowering dementia risk in patients with obesity. Further studies are needed to explore the underlying mechanisms.