Introduction: Pre-hospital emergency medicine is crucial for the treatment of timely critical ill patients. Accurate telephone triage performed by 144 operators is therefore of crucial impor- tance in determining patient priority, in order to determine which resources should be sent to which patient to provide the best quality care. Under-triage on dispatch at the time of the call could, depending on the case and the severity of the under-triage, determine a worse clinical outcome than normally expected, leading to situations of serious and avoidable disability and death.
Objective: With this study we aim to analyze cases of under-triage by the 144-emergency me- dical dispatch center in order to highlight the overall performance, and based on the addi- tional elements analyzed, understand which factors are more closely linked than others to under-triage. Thanks to the results, we hope to provide objective feedback to the 144 dispatch centers, and, at the same time, elements for potential improvement.
Methods: This is a retrospective population study, where we’ve analyzed data provided by the FUS (Fondation Urgences Santé) about cases of under-triage that occurred in the period between 01.06. and 31.12.2021. The inclusion criteria we used in the study included all cases triated with a P2 or P3 priority (< P1) that than had an on-scene NACA scores between 4 and 7 (> 3) which relates to a more important problem.
Results: The total percentage of under-triage in the indicated period amounts to 5.18%, of the- se the majority (84.50%) was moderately under-triaged (P2), while a lower percentage (15.50%) was severely under-triaged (P3). The main problems encountered in cases of under-triage (P2 and P3 combined) were respiratory distress (23.72%) and shock (12.11%). We analyzed other results such as the correlation between under-triage and patient age, NACA-scores, and other factors.
Conclusions: Overall, the 144-dispatch system was found to have an average performance comparable to those analyzed in other studies (5.18% versus an average of 4.86%). However, the study also revealed a difficulty in recognizing certain types of problems (respiratory di- stress, shock, reduced state of consciousness, ...), which leaves room for improvement that can be achieved through training as well as the implementation of new technologies and software with AI integration available on the market that have shown their effectiveness.