Effectual Emergency Severity Adaptation For Improved Triage Care Operations

Studies have shown that triage nurses often have difficulty correctly predicting triage level when using the Emergency Severity Index (ESI), even after having been previously trained on the algorithm. This paper examines nurses' difficulties with applying the ESI and previous attempts at electronic triage support systems. United States emergency department data obtained from the National Center for Health Statistics website is analyzed using the random forest algorithm and C5.0 decision tree algorithm to determine if refinements can be made to the current ESI algorithm so that it can be more easily implemented by nurses. A model electronic triage application is produced based on the current ESI algorithm and the results of the decision tree analysis.