Ruben De Rouck - Presenter

Ives Hubloue - Contributor

Background: The lack of a universally accepted, reliable and validated mass casualty triage system remains an important gap. Moreover, the lack of a standard measure against which to measure triage outcomes makes it difficult to evaluate and compare triage systems. Computer simulation is an established methodology in operations research but remains underused in emergency management research. This study used a disaster medical management simulator to investigate the effect of several triage methods on the mortality of victims as performance outcome measure. Methods A previously developed comprehensive discrete event simulation model was updated and configured to simulate the prehospital response of an airplane crash at an international airport. The simulator manages the dynamic evolution of the health state of each victim based on time and treatment triggers and survival probability estimation. The mix of injury type and severity is based on published data of airplane crashes at airports. There were 250 occupants in the plane with 205 injured (26 T1, 62 T2, 113 T3 and 4 T4), 5 fatalities and 45 uninjured victims. Since airplane crashes are self-contained mass casualty incidents, there is no real lack of regional medical resources, but a problem of their effective mobilization and deployment. The ambulatory victims will self-evacuate out of the crashed plane and escorted to a non-urgent care area; the non-ambulatory survivors will be extricated at random by rescue teams and transported to a casualty collection point where they will be triaged. Victims are either transported directly to hospitals while treatment starts in the ambulances (scoop-and-run policy, ScR) or via a forward medical post were they receive stabilising treatment (stay-and-play policy, StP). When deciding between victims of the same triage category, Sacco's RPM score was used as a tiebreaker. During transport to the hospital, T1 victims are supervised by an emergency physician or nurse, T2 victims by an emergency nurse or medical technician. Victims are distributed among the hospitals based on their treatment capacity. Fifty replications of 4 configurations for each triage method (NATO, START, CareFlight, SIEVE and SALT) have been carried out. The results were analysed by one-way ANOVA and the post-hoc Scheffé test. Results Overall there is a mean mortality of 14.3 (95%CI: 13.8-14.8), with a minimum of 5 and a maximum of 23. START and CareFlight resulted in an average mortality of 12.9 and 13.2 respectively, (not statistically different). NATO resulted in a mortality of 14.3, SALT 15.2 and Sieve 16.0, all differences being statically significant (P< 0.001). On average StP resulted in 16.7 deaths, as opposed to 11.9 for ScR. For ScR, START (8.8 deaths) performed significantly better than CareFlight (9.3), NATO (12.8) and SALT (13.6), all P<0.001. In case of StP, NATO (15.8) performed better than SALT (16.9), SIEVE(16.9), START (17.0) CareFlight (17.0). The latter were not significantly different in post hoc analysis. Conclusions Our results suggest a link between the triage method and mortality. Further research is needed to clarify the influencing factors for this effect.
10 Sep 2018

Event (Conference)

TitleEuropean Society for Emergency Medicine (EUSEM) 12th Emergency Medicine Congres
Abbrev. TitleEUSEM Congress
Period8/09/1812/09/18
Web address (URL)
LocationScottish Event Campus
CityGlasgow
CountryUnited Kingdom

ID: 41837804