Major incident triage: Derivation and comparative analysis of the Modified Physiological Triage Tool (MPTT).

Abstract

BACKGROUND: Triage is a key principle in the effective management at a major incident. There are at least three different triage systems in use worldwide and previous attempts to validate them, have revealed limited sensitivity. Within a civilian adult population, there has been no work to develop an improved system. METHODS: A retrospective database review of the UK Joint Theatre Trauma Registry was performed for all adult patients (>18years) presenting to a deployed Military Treatment Facility between 2006 and 2013. Patients were defined as Priority One if they had received one or more life-saving interventions from a previously defined list. Using first recorded hospital physiological data (HR/RR/GCS), binary logistic regression models were used to derive optimum physiological ranges to predict need for life-saving intervention. This allowed for the derivation of the Modified Physiological Triage Tool-MPTT (GCS≥14, HR≥100, 12<RR≥22). A comparison of the MPTT and existing triage tools was then performed using sensitivities and specificities with 95% confidence intervals. Differences in performance were assessed for statistical significance using a McNemar test with Bonferroni correction. RESULTS: Of 6095 patients, 3654 (60.0%) had complete data and were included in the study, with 1738 (47.6%) identified as priority one. Existing triage tools had a maximum sensitivity of 50.9% (Modified Military Sieve) and specificity of 98.4% (Careflight). The MPTT (sensitivity 69.9%, 95% CI 0.677-0.720, specificity 65.3%, 95% CI 0.632-0.675) showed an absolute increase in sensitivity over existing tools ranging from 19.0% (Modified Military Sieve) to 45.1% (Triage Sieve). There was a statistically significant difference between the performance (p<0.001) between the MPTT and the Modified Military Sieve. DISCUSSION & CONCLUSION: The performance characteristics of the MPTT exceed existing major incident triage systems, whilst maintaining an appropriate rate of over-triage and minimising under-triage within the context of predicting the need for a life-saving intervention in a military setting. Further work is required to both prospectively validate this system and to identify its performance within a civilian environment, prior to recommending its use in the major incident setting.