Short-Term Risk of Type 2 Diabetes in Patients Using Various Antidepressants Compared with Patients Using Fluoxetine

Abstract

Background: The objective is to compare the risk of developing type 2 diabetes (T2D) within a year in patients prescribed various antidepressants (ADs) and those prescribed fluoxetine as a control group.

Methods: This study used standardized data from the Health Insurance Review and Assessment Service claims database (n=1,456,489). Patients aged ≥10 years with no previous use of ADs and no history of diabetes mellitus, regardless of whether they were diagnosed with any depressive disorder, were eligible for this study. Among these eligible patients, those who had used ADs for the first time or had never used them between January 2017 and December 2017 were selected for this study. I compared the short-term (<12 months) risk of T2D in patients using various ADs, excluding tricyclic ADs, with those using fluoxetine as a control. The Cox proportional hazards model was used to calculate hazard ratios (HRs).

Results: The HRs (95% confidence intervals) for T2D incidence in the various AD groups compared with that in the fluoxetine group are as follows: 0.84 (0.67-1.06, P = .15), bupropion; 0.91 (0.77- 1.07, P=.25), tianeptine; 0.91 (0.77-1.07, P=.25), escitalopram; 0.96 (0.82-1.13, P = .63), paroxetine; 0.97 (0.70-1.35, P=.87), fluvoxamine; 1.07 (0.85-1.36, P=.55), vortioxetine; 1.07 (0.91-1.25, P=.42), sertraline; 1.14 (0.99-1.31, P = .07), duloxetine; 1.17 (0.97-1.41, P = .09), mirtazapine; 1.17 (1.00-1.38, P=.05), trazodone; 1.22 (1.04-1.45, P=.02), venlafaxine; and 1.29 (1.03-1.61, P = .03), milnacipran.

Conclusion: The short-term risk of T2D was significantly higher in the milnacipran and venlafaxine groups than in the fluoxetine group. All other ADs except milnacipran and venlafaxine showed no difference in the risk of developing T2D compared to fluoxetine. These results suggest that clinicians should be mindful of the risk of developing T2D when administering milnacipran and venlafaxine to patients.