Health & Medicinepreprint2026-11-18

Comparative Risk of Incident Autoimmune Diseases by First Add-on Antidiabetic Drug Class After Metformin: A Population-Based Active-Comparator, New-User Cohort Study

Open access0 citations

Abstract

Abstract Purpose To compare the incidence of autoimmune rheumatic diseases among patients with type 2 diabetes initiating different oral glucose-lowering drug classes as the first add-on to metformin. Methods This active-comparator, new-user cohort study used the Korean National Health Insurance Service–National Sample Cohort. Adults who initiated metformin monotherapy and subsequently added a dipeptidyl peptidase-4 (DPP-4) inhibitor, sodium–glucose cotransporter-2 (SGLT-2) inhibitor, sulfonylurea, or thiazolidinedione between January 2014 and June 2023 were included. Patients with a history of autoimmune rheumatic disease were excluded. The primary outcome was incident autoimmune rheumatic disease identified using diagnostic codes. Multivariable Cox regression was used to estimate adjusted hazard ratios (aHRs) and 95% confidence intervals (CIs), with DPP-4 inhibitors as the reference. Results The cohort included 14,463 patients: 9,279 receiving DPP-4 inhibitors, 2,339 receiving SGLT-2 inhibitors, 2,434 receiving sulfonylureas, and 411 receiving thiazolidinediones. The incidence rates of autoimmune rheumatic disease were 15.32, 13.16, 16.67, and 10.97 per 1,000 person-years, respectively. Compared with DPP-4 inhibitors, SGLT-2 inhibitors were associated with lower risks of autoimmune rheumatic disease (aHR, 0.80; 95% CI, 0.65–0.99) and rheumatoid arthritis (aHR, 0.79; 95% CI, 0.63–0.99). The corresponding aHRs for overall autoimmune rheumatic disease were 1.11 (95% CI, 0.94–1.29) for sulfonylureas and 0.65 (95% CI, 0.42–1.01) for thiazolidinediones. Conclusion Among patients adding a second oral agent to metformin, SGLT-2 inhibitor initiation was associated with a lower incidence of autoimmune rheumatic disease than DPP-4 inhibitor initiation. Residual confounding and claims-based outcome ascertainment limit causal interpretation.

// Source

View paper (DOI)Open access versionOpenAlexResearch SquarePublished 2026-11-18

Authors: Seung‐Hun You, Wonsuk Shin, Yil‐Seob Lee, Hyounggyoon Yoo

Institutions: CHA University Bundang Medical Center