(046) Effect of SGLT2 Inhibitors on Decreasing the Risk of Depression in Patients with Diabetes: A Nationwide Active-Comparator Cohort Study in Taiwan

Background: Type 2 diabetes mellitus (T2DM) and depression are both common diseases that can significantly decrease patients' quality of life. Prior research has already indicated an association between T2DM and depression. However, it is still unclear the impact of sodium-glucose co-transporter 2 inhibitors (SGLT2is), on the risk of subsequent depression in patients with T2DM.

Objectives: To investigate whether the risk of developing depression among T2DM patients is reduced after using SGLT2is.

Methods: This study analyzed National Health Insurance data in Taiwan. Using a cohort study with active comparator user design, we investigated the risk of developing depression in patients with T2DM who are using SGLT2is. After excluding individuals under 20 years old, those diagnosed with prior medical history of depression, anxiety disorders, dementia, bipolar disorder, schizophrenia, we identified patients who used SGLT2is (ATC code: A10BK) between January 1, 2016, and 2018 as exposure group, and patients who used DPP-4is (ATC code: A10BH) during the study period as active comparator group. Each selected patients were followed to ascertain whether he or she received a depression diagnosis (ICD-9-CM codes: 296.2、, 296.3、, 296.82、, 296.99、, 298.0、, 300.4、, 301.12、, 309.1、, 311; ICD-10-CM: F34.1、, F43.21、, F32、, F33). Cox proportional hazards models were used to estimate hazard ratios (HR) and 95% confidence intervals (CI).

Results: Among the diabetic population in Taiwan, new-onset depression developed 0.93% in SGLT2i users, 2.03% in DPP4i users. Furthermore, Cox proportional hazards regression indicated that SGLT2i users had lower risk of depression compared with DPP4i users (HR: 0.439; 95% CI: 0.412-0.468; p < 0.001). Most subgroup analyses also displayed the inverse risk of depression in SGLT2is users.

Conclusions: The risk of developing depression is lower with SGLT2i users compared to DPP-4i users. We will consider more confounders in the regression models to demonstrate the relationship between SGLT2is and subsequent risk of depression in the future. In addition, sensitivity analyses will be performed.