Evaluating the National Early Warning Score (NEWS) in triage: A machine learning perspective

Area of Science:

• Emergency Medicine
• Clinical Assessment
• Patient Triage

Background:

• The National Early Warning Score (NEWS) is a standard tool in Emergency Departments (EDs) for predicting patient mortality.
• Its efficacy in evaluating other critical triage parameters like urgency and severity requires further investigation.

Purpose of the Study:

• To assess the predictive performance of the National Early Warning Score (NEWS) for various clinical outcomes beyond mortality.
• To evaluate NEWS's accuracy in reflecting physician-defined patient urgency and severity.

Main Methods:

• A prospective observational study involving 1,270 patients in the Merano Emergency Department.
• Categorization of NEWS into low, intermediate, and high-risk groups.
• Decision tree and Receiver Operating Characteristic (ROC) curve analyses to evaluate NEWS against 30-day mortality, ICU admission, general hospitalization, and physician-defined urgency/severity.

Main Results:

• NEWS showed strong predictive value for 30-day mortality (ROC=0.82) and ICU admission (ROC=0.72).
• Decision tree analysis confirmed NEWS as a key predictor for mortality and ICU admissions.
• NEWS demonstrated lower reliability for assessing physician-defined urgency and severity, with a tendency to misclassify critically ill patients.

Conclusions:

• The National Early Warning Score (NEWS) is a valuable tool for predicting mortality and ICU admissions in the ED setting.
• Limitations in assessing broader clinical outcomes indicate NEWS should complement, not replace, existing triage protocols.
• Further research may explore integrating NEWS with other clinical indicators for comprehensive patient assessment.