INTRODUCTION: Non-chemotherapy drug-induced agranulocytosis (NCDIA) is a serious adverse reaction that significantly increases the risk of life-threatening infections. Although the association between certain non-chemotherapy drugs and agranulocytosis has been documented, a comprehensive analysis using a large-scale pharmacovigilance database is lacking. This study aimed to systematically identify and characterize NCDIA by analyzing adverse event reports from the FAERS database. METHODS: We conducted a retrospective analysis of NCDIA reports from the FAERS database spanning from 2004 to 2024 Q1. Drugs were classified using the Anatomical Therapeutic Chemical (ATC) classification system, with chemotherapy agents (ATC code L01) excluded. The Reporting Odds Ratio (ROR) method was employed to detect potential adverse event signals. Positive signals were defined as cases with at least three reports and a lower 95% confidence interval (CI) of ROR greater than one. Time-to-event analysis was also performed to examine onset patterns across different demographic groups and drugs. RESULTS: A total of 10,913 NCDIA reports were identified from the FAERS database. Disproportionality analysis revealed significant signals for 166 non-chemotherapy drugs related to agranulocytosis, which were systematically classified into three risk categories: known (n = 111), possible (n = 25), and new potential risks (n = 30). This classification system enables us to identify drugs with known risks, those that might pose a risk, and new risks warranting further investigation. Demographic analysis revealed that females, children (<
18 years), and the elderly (≥65 years) experienced earlier onset of agranulocytosis. Drug-specific onset timing analysis provided evidence for optimizing monitoring protocols. Notably, NCDIA-associated mortality rates showed a significant decrease from 11.91% (2004-2010) to 7.28% (2021-2024) ( CONCLUSION: This comprehensive pharmacovigilance study not only confirmed previously known NCDIA associations but also identified new potential risk drugs. The novel risk classification system and detailed onset timing analysis provide valuable insights for clinical monitoring. The findings of earlier onset in specific populations and declining mortality trends have important implications for developing targeted surveillance strategies and improving patient safety management.