# Identifying Predictors of Serious Adverse Events in Antidepressant Treatment from a Decade-Long Nationwide Pharmacovigilance Study: Impact of Dementia and Parkinson’s Disease Treatment

## TL;DR

This study finds that certain antidepressants and combinations with other medications increase the risk of serious side effects, especially in men.

## Contribution

The study identifies fluoxetine, male sex, and concomitant antiparkinsonian and antidementia treatments as key predictors of serious adverse events.

## Key findings

- Fluoxetine is the only antidepressant with a high likelihood of serious adverse events.
- Concomitant antiparkinsonian treatment increases SAE risk the most (OR 8.29).
- Vascular and liver-related ADEs are most likely to be serious.

## Abstract

Backgrounds and Objectives: This study aims to characterize the prevalence and severity of antidepressant-associated adverse drug events (ADEs) and to identify predictors strongly associated with serious adverse events (SAEs). Materials and Methods: Disproportionality analysis on antidepressant-related ADEs spontaneously reported to the Korea Adverse event Reporting System (KIDS KAERS DB) from 2014 to 2023 was performed. Multiple logistic regression was conducted to identify predictors associated with SAEs. Sensitivity analysis was performed to validate the overall findings and assess the robustness of associations across subgroups defined by completeness of demographic data (age and sex), elderly age-stratification, and causality assessment. The study protocol was approved by the Kyung Hee University institutional review board. Results: Among 21,103 antidepressant-related ADEs, duloxetine was the most etiologic medication, followed by amitriptyline and escitalopram. Fluoxetine is the only agent with a high likelihood of reporting SAEs. ADEs involving vascular (extracardiac) disorders (ROR 42.42, 95% CI 13.19–136.42) and liver and biliary system disorders (ROR 7.84, 95% CI 3.77–16.29) were most likely to be SAEs. The predictors associated with substantial increased SAE risk were fluoxetine use (OR 2.71, 95% CI 1.68–4.39), male sex (OR 1.48, 95% CI 1.11–1.98), and concomitant administration of antiparkinsonian treatment (OR 8.29, 95% CI 3.61–19.06) and antidementia treatment (OR 2.94, 95% CI 1.34–6.05). Sensitivity analyses demonstrated similar and consistent findings. However, reversed trends in the association between SOC-based ADEs and sex were observed in the sensitivity analysis restricted to cases with “certain” and “probable” causality. Conclusions: The type of antidepressant, concomitant medications, and sex are major predictors for SAE risk. Further controlled studies on the impact of comorbidities and polypharmacy on antidepressant-related SAEs are warranted.