Machine Learning--Enhanced Pharmacovigilance Assessment of Cardiac Adverse Events Associated With Minoxidil

This study investigated cardiac adverse events associated with oral minoxidil using data from the U.S. FDA Adverse Event Reporting System (FAERS). Among 61,223 reports, 614 met inclusion criteria, with 276 (45%) involving cardiac events. Disproportionality analysis showed a strong association between minoxidil and cardiac events, with high Reporting Odds Ratio (ROR) and Proportional Reporting Ratio (PRR) values. Machine learning models, particularly Random Forest, outperformed traditional methods in distinguishing cardiac from non-cardiac events, highlighting patient age as a key factor. The study suggests that integrating machine learning with traditional methods can enhance drug safety assessments, though findings are hypothesis-generating and require further investigation.