[Combination of Manchester Triage System and chief complaint: development and internal validation of a new algorithm for patient flow management in integrated emergency centres].

Abstract

Background: A central element of the emergency care reform in Germany is the establishment of integrated emergency centres (INZ) with a central point of assessment (ZE). This simulation study presents a new tool (L-MTS) combining the Manchester Triage System (MTS) with the Canadian Emergency Department Information System (CEDIS) chief complaint to identify patients with low urgency early, thereby decreasing the workload in the central emergency department (ZNA) while ensuring patient safety in the urgent care centre (AA).

Methods: Retrospective analysis of 38,639 adult self-presenting emergency patients (fNP) at two university hospitals (2024). Following randomization (70% training, n = 27,048/30% test, n = 11,591), chief-complaint-based outpatient rates were calculated as pretest probabilities. Using ROC analysis and the maximum Youden index, a 74% cutoff was defined for the allocation algorithm. This was refined into an optimized version (L-MTS opt.) by routing high-risk complaints directly to the ZNA. Internal validation within the test dataset was performed using decision curve analysis (DCA), calibration curves (Brier score), and bootstrapping (min. 5000 iterations).

Results: Outpatient rates varied substantially depending on the chief complaint. Using L-MTS opt., the ZNA workload could be reduced by up to 44.3% of all fNP within the test dataset. This includes 12.9% of patients requiring secondary transfer from the AA to the ZNA. The proportion of AA patients requiring monitoring or who died was 0.20% and 0.14%, respectively.

Conclusion: L-MTS opt. substantially reduces emergency department overcrowding while maintaining high patient safety. The broad availability of the variables used and the high adaptability of the allocation algorithms allow for wide application of the system. The findings support the need for close spatial and organizational integration of ZNA and AA to optimize patient flow.