Metformin‐based oral antihyperglycemic combination therapy and risk of dementia in patients with newly diagnosed type 2 diabetes: A population‐based study

INTRODUCTION This study examined the association between metformin‐based oral antihyperglycemic combination therapy and the risk of developing dementia in patients with newly diagnosed type 2 diabetes mellitus (T2DM). METHODS We conducted a retrospective cohort study using data from a random sample of 2,000,000 individuals in Taiwan's National Health Insurance Research Database from 2000 to 2018. A total of 44,073 patients with newly diagnosed T2DM were identified and categorized into four groups based on initial treatment: metformin (MET)‐sulfonylureas (SU), MET–dipeptidyl peptidase‐4 inhibitor (DPP4i), MET–thiazolidinedione (TZD), and MET–glinides. The participants were followed until December 2018. RESULTS The hazard ratio of MET–TZD was 0.73 (95% confidence interval [CI]: 0.57–0.94), of MET–DPP4i was 0.64 (95% CI: 0.52–0.79), and of MET–glinides was 1.14 (95% CI: 0.94–1.39). DISCUSSION The MET–DPP4i group had the lowest risk of dementia, followed by the MET–TZD group. Highlights Newly diagnosed type 2 diabetes mellitus (T2DM) patients using metformin (MET)–glinides and MET–sulfonylureas are at an increased risk of dementia. The MET–glinides group had the highest risk of dementia. The MET–dipeptidyl peptidase‐4 inhibitor group had the lowest risk of dementia, followed by the MET–thiazolidinedione group.