Abstract: Medication overdose is a significant public health concern, leading to thousands of preventable deaths annually. This study focuses on the development and application of predictive analytics using machine learning models to identify potential medication overdoses, thereby enhancing patient safety. By integrating structured patient data and electronic health records, the proposed approach utilizes logistic regression and random forest algorithms for risk prediction and feature importance analysis. The research demonstrates how these predictive models can accurately identify high-risk patients, offering actionable insights for proactive interventions. Our findings underscore the potential of predictive analytics to transform overdose prevention strategies in healthcare, paving the way for real-time decision support systems.
Rampure et al. (Mon,) studied this question.