Comparing the National Early Warning Score and the Manchester Triage System in Emergency Department Triage: A Multi-Outcome Performance Evaluation.

Emergency department (ED) triage systems aim to prioritize patients based on clinical severity, ensuring timely intervention for high-risk cases. Recently, the National Early Warning Score (NEWS) has been proposed as an alternative to traditional triage systems, but its efficacy across multiple clinical outcomes remains unclear. This study aimed to compare the predictive performance of the NEWS and the Manchester Triage System (MTS) across multiple clinical outcomes. We conducted a retrospective, single-center study at Merano Hospital, Italy, from 1 June 2022 to 30 June 2023, comparing the performance of the NEWS and the Manchester Triage System (MTS). All adult ED patients (≥18 years) were included, while exclusions applied to those on fast-track pathways, non-residents, and pregnant patients. Primary outcomes included 30-day mortality, hospitalization, and ICU admission. A random 5% subgroup was analyzed for secondary outcomes, including the need for life-saving interventions (LSIs), physician-defined clinical priority, and severity. Predictive performance was assessed using Receiver Operating Characteristic (ROC) curves, area under the ROC curve (AUROC) comparisons, and Decision Curve Analysis (DCA). Among 27,238 patients, the NEWS predicted 30-day mortality more accurately than the MTS (AUROC 0.745 vs. 0.701, < 0.001). However, the MTS outperformed the NEWS for hospitalization (AUROC 0.733 vs. 0.609, < 0.001), ICU admission (AUROC 0.862 vs. 0.672, < 0.001), and all secondary outcomes. DCA further confirmed MTS's superiority across clinically relevant ED probability thresholds (20-40%). The NEWS, while effective for predicting mortality, it is inadequate in comprehensive triage decision-making. The MTS remains the superior system for prioritizing high-risk patients based on clinical severity. Rather than replacing triage with the NEWS, efforts should focus on refining existing systems to improve risk stratification. Future multi-center prospective studies are necessary to validate these findings.