The US Food and Drug Administration’s Adverse Event Reporting System (FAERS) is an indispensable pharmacovigilance resource, yet it is increasingly used in ways that go beyond its design, particularly in the academic literature and hospital formulary debates. Researchers and clinicians may treat FAERS outputs as if they were incidence data or comparative risk estimates, which can lead to overconfident conclusions about the relative safety of drugs or formulations. This editorial focuses on common methodological and interpretive errors made by researchers when using FAERS—especially in comparative safety assessment—and outlines how FAERS can be used appropriately as a hypothesis-generating tool rather than a definitive source for clinical decision-making. 1–8 FAERS aggregates individual case, spontaneous safety reports submitted by manufacturers, healthcare professionals and consumers, encompassing millions of reports that reflect real-world drug use. These reports include structured elements such as coded adverse events, patient demographics and drug details, alongside unstructured narratives that provide context, including onset timing and outcomes. In pharmacovigilance, FAERS excels at hypothesis generation for safety signals, which are potential new or intensified drug–event associations that warrant verification through other sources, such as electronic health records or registries. Healthcare providers and researchers frequently use FAERS during formulary reviews, risk assessments for high-risk populations such as elderly inpatients on polypharmacy, and adverse drug reaction monitoring programmes. Unlike controlled clinical trials, FAERS captures rare events that are elusive in pre-approval studies, including idiosyncratic reactions in off-label … Correspondence to Dr Andrej Belančić; a. belancic93atgmail. com
Gkrinia et al. (Mon,) studied this question.