Supplementary Material for: Antidepressants and Risk of Mortality in People with Depressive Disorder and Co-occurring type 2 Diabetes Mellitus: a 20-year Population-based Cohort Study

Introduction: Depression and diabetes often co-occur and worsen clinical outcomes of both conditions. However, mortality risk among depression patients with diabetes exposed to antidepressant is under-studied. We investigated whether antidepressant would decrease mortality risk in people with depression and type-2 incident diabetes.
Methods: This population-based cohort study identified 11,137 depression patients with incident type-2 diabetes between 2002-2021 in Hong Kong who were exposed to antidepressants, using territory-wide electronic medical-record database. Association between antidepressant exposure and mortality risk was analyzed by Cox proportional-hazards models for any antidepressant, specific drug classes, and individual agents, with stratified-analysis by HbA1c level. A comprehensive array of covariates, including age, sex, calendar-year period, catchment-area, pre-existing physical-comorbidities, diabetic-complications, substance/alcohol use disorders, cardiovascular/antidiabetes medications, and presence of antidepressants other than the specified drug was adjusted. Three sets of sensitivity-analyses were conducted by restricting to patients (a) with cumulative drug exposure ≥90 days and ≥180 days, (b) with medication-possession-ratio ≥80% and (c) monotherapy.
Results: Lower risk of all-cause mortality was associated with exposure to any antidepressant (HR 0.79, 95%CI 0.70-0.90) compared with no antidepressant in depression patients with incident-diabetes. Lower mortality risk was associated with exposure to noradrenergic and specific-serotonergic antidepressants (0.77[0.66-0.90]) compared with no antidepressant, and to mirtazapine (0.76[0.65-0.88]) and trazodone (0.75[0.63-0.90]). Sensitivity-analyses affirmed that lower mortality risk was associated with mirtazapine.
Conclusion: Depression patients with comorbid type-2 diabetes with exposure to several antidepressant is at decreased mortality risk. Further research is warranted to confirm our findings and clarify the mortality-reducing mechanisms of antidepressant in this vulnerable population.