Abstract
With the advancement of global healthcare and drug research, drug-induced adverse reactions, particularly drug-induced adverse effects of depression (DIAEs), are garnering increased attention. DIAEs significantly impact patients' quality of life, treatment adherence, and prognosis, potentially exacerbating conditions, leading to medication discontinuation, or triggering extreme events like suicide. Therefore, exploring DIAEs mechanisms and effects is crucial. The aim of this study was to comprehensively explore and analyze drug-induced depressive adverse events (DIAEs) using the US Food and Drug Administration's (FDA's) Adverse Event Reporting System (FAERS) database, which provides a scientific basis for drug safety monitoring, clinical medication guidance, and improvement of drug development. In this study, we analyzed in-depth the time of onset, drug class, drug interactions, and demographic characteristics of DIAEs by data cleaning and preprocessing the reports of depression-related adverse events (AEs) in the FAERS database from the first quarter of 2004 to the third quarter of 2024, using statistical methods and data mining techniques. The disproportionate analysis (DPA) algorithm was used to combine multiple algorithms (e.g., ROR, PRR, BCPNN, EBGM) for signal detection of DIAEs-related drugs. It was found that the number of depression-related drug reports in the FAERS database increased year by year, and the trend of depressive AEs showed a polynomial growth curve with high R2 values. The analysis showed that drugs such as Varenicline, Isotretinoin and Adalimumab were highly associated with depressive AEs and that the risk associated with depression was not mentioned in the labeling of 12 drugs, revealing new drug signals. Analysis of the time to onset of DIAEs revealed an "early failure curve" for many drugs, with a median time to onset of depression of 27 days for Varenicline. The in-depth analysis of DIAEs in the FAERS database revealed the epidemiological characteristics, demographic distribution, and potential risk factors of depressive AEs, which provides an important basis for drug safety monitoring, clinical decision-making, and drug labeling updates. The study also identified multiple signals of drug depression that were not labeled in drug labels, suggesting that the monitoring of high-risk drugs should be strengthened in clinical practice.