Abstract
OBJECTIVE: To compare incident dementia risk among patients who initiated treatment with metformin or sulfonylurea in Veterans Health Affairs (VHA) patients with replication in Kaiser Permanente Washington (KPW) patients to determine whether first-choice antidiabetic medications are associated with reduced risk of dementia. PATIENTS AND METHODS: Cohorts contained 75,187 VHA patients and 10,866 KPW patients, 50 years and older, who initiated monotherapy with metformin or sulfonylurea. Patients were free of dementia diagnoses and any diabetes treatment for 2 years before cohort entry. Variables were extracted from electronic health data from VHA (1999-2015) and KPW (1996-2015), which included diagnosis codes, pharmacy data, laboratory values, and demographic characteristics. Propensity scores and inverse probability of treatment weighting controlled for confounding. RESULTS: Veterans Health Affairs patients were 60.8±6.8 years of age on average, and KPW patients were 63.1±9.5 years of age. In the VHA sample, 72,769 (96.8%) were male; and in the KPW sample, 5480 (50.4%). After adjusting for confounding, metformin initiation was associated with a significantly (P=.02) lower risk of dementia in VHA (hazard ratio, 0.9; 95% CI, 0.9-1.0), with a similar point estimate in KPW (hazard ratio, 0.9; 95% CI, 0.7-1.1). Metformin was not associated with dementia risk in patients 75 years and older. CONCLUSION: Existing epidemiological studies of metformin and incident dementia have been inconsistent. Using a similar study design in 2 patient populations that differed in clinical and demographic characteristics, our results provide robust evidence that metformin use is associated with a modestly lower risk of incident dementia.