Ala’ Karajeh¹ and Rasit Eskicioglu², ¹Independent Researcher, Canada, ²Atlas University, Turkey

Older patients often present with multiple comorbidities and distinct clinical patterns, making their severity assessment more challenging in emergency settings. This study analyzes two clinical databases from Beth Israel Deaconess Medical Center to examine the triage characteristics and disposition outcomes of older emergency patients. A framework of four machine-learning models was developed and compared to a baseline logistic regression model to predict whether a patient is likely to be hospitalized or discharged based on triage information. The models demonstrated reasonable predictive performance and highlight the potential of using machine learning-based triage tools to support early risk identification and improve decision-making for this patient group.

Emergency Medicine, Hospitalization Prediction by Machine Learning, Emergency Older Patients Classification, and Emergency Older Patients Data Analytics