PLoS One. 2025 Jul 11;20(7):e0327844. doi: 10.1371/journal.pone.0327844. eCollection 2025.
ABSTRACT
BACKGROUND: While antidepressants are effective in alleviating symptoms, their association with mortality remains unclear. This research investigated the link between antidepressant usage and all-cause mortality among depressed patients.
METHODS: We performed a real-world study on 5,947 adults with depression using a dataset from the National Health and Nutrition Examination Survey (2005-2018). Depression was identified by a Patient Health Questionnaire-9 score ≥10, or the use of antidepressants, with all-cause mortality assessed through the National Death Index. Covariates included demographics, socioeconomic status, lifestyle factors, and chronic conditions. The study performed weighted Cox proportional-hazards models, propensity score methods, and inverse probability of treatment weighting (IPTW) to estimate hazard ratios (HRs) and 95% confidence intervals (CIs) for comparing mortality risk between patients treated with antidepressants and those who were not. We conducted sensitivity analyses to evaluate the robustness of our findings.
RESULTS: During the median 82-month follow-up period, 15.0% of participants (n = 894) died. Antidepressant users (n = 3,925) had a crude mortality rate of 16.5%, compared to 12.2% in non-users (n = 2,022). The crude Cox proportional-hazards analysis indicated that antidepressant use was linked to a non-significant elevation in mortality (HR = 1.18, 95% CI 0.95-1.47, P = 0.126). This association attenuated completely after covariate adjustment (adjusted HR = 0.92, 95% CI 0.75-1.13). Propensity score analyses indicated no significant link between antidepressant use and mortality (IPTW, HR = 0.96, 95% CI 0.80-1.16, P = 0.707). Across all methods, no statistically significant association was observed.
CONCLUSION: All-cause mortality is not significantly affected by the overall use of antidepressants in individuals with depression; however, future studies should investigate safety differences between specific drug classes.
PMID:40644427 | DOI:10.1371/journal.pone.0327844
Recent Comments