Elsevier

Injury

Volume 47, Issue 5, May 2016, Pages 988-992
Injury

Performance characteristics of five triage tools for major incidents involving traumatic injuries to children

https://doi.org/10.1016/j.injury.2015.10.076Get rights and content

Abstract

Context Triage tools are an essential component of the emergency response to a major incident. Although fortunately rare, mass casualty incidents involving children are possible which mandate reliable triage tools to determine the priority of treatment.

Objective

To determine the performance characteristics of five major incident triage tools amongst paediatric casualties who have sustained traumatic injuries.

Design, setting, participants

Retrospective observational cohort study using data from 31,292 patients aged less than 16 years who sustained a traumatic injury. Data were obtained from the UK Trauma Audit and Research Network (TARN) database.

Interventions Statistical evaluation of five triage tools (JumpSTART, START, CareFlight, Paediatric Triage Tape/Sieve and Triage Sort) to predict death or severe traumatic injury (injury severity score >15).

Main outcome measures Performance characteristics of triage tools (sensitivity, specificity and level of agreement between triage tools) to identify patients at high risk of death or severe injury.

Results

Of the 31,292 cases, 1029 died (3.3%), 6842 (21.9%) had major trauma (defined by an injury severity score >15) and 14,711 (47%) were aged 8 years or younger. There was variation in the performance accuracy of the tools to predict major trauma or death (sensitivities ranging between 36.4 and 96.2%; specificities 66.0–89.8%). Performance characteristics varied with the age of the child. CareFlight had the best overall performance at predicting death, with the following sensitivity and specificity (95% CI) respectively: 95.3% (93.8–96.8) and 80.4% (80.0–80.9). JumpSTART was superior for the triaging of children under 8 years; sensitivity and specificity (95% CI) respectively: 86.3% (83.1–89.5) and 84.8% (84.2–85.5). The triage tools were generally better at identifying patients who would die than those with non-fatal severe injury.

Conclusion

This statistical evaluation has demonstrated variability in the accuracy of triage tools at predicting outcomes for children who sustain traumatic injuries. No single tool performed consistently well across all evaluated scenarios.

Introduction

The term ‘big bang’ major incident is used to describe a major incident caused by sudden catastrophic events with little or no warning, where the number of casualties is relatively constant from the time of the incident but has the potential to outstrip resources [1], [2]. Such incidents test the response of emergency medical services and hospitals and it is essential that resources are used in an optimal way to target those with greatest need [3]. In order to achieve this, one of the first priorities is to undertake rapid and accurate triage to prioritise and provide care to as many casualties as possible with the intention of minimising loss of life and suffering, moderated by the available resources. However, there is uncertainty around the efficacy of commonly used triage systems, particularly in children

[4], and a recent systematic review of the literature concluded that there is limited evidence of the validity of triage tools in major incidents of this nature [1].

This study aims to assess the performance accuracy of five manual/paper based triage tools when assessing paediatric casualties and to compare the level of agreement between them. The tools assessed are: JumpSTART (age ≤8 years) [5], START (age >8 years) [6], CareFlight [7], Paediatric Triage Tape/Sieve [8], Triage Sort [9].

Section snippets

Study design and data collection

A retrospective observational cohort study was undertaken. Approval was obtained from the Trauma Audit and Research Network (TARN; www.tarn.ac.uk) to analyse data from the TARN database. TARN collects and records data from hospitals across England and Wales for patients who sustain injury resulting in hospital admissions for >3 days, critical care admission or death. A dataset was obtained in August 2009 containing 31,560 paediatric trauma patient records for patients aged less than 16 years,

Results

A total of 31,292 patients aged less than 16 years were included in the study; 10,048 females (32.1%) and 21,244 males (67.9%), with mean ages 7.9 years (standard deviation 4.9 years) and 8.7 years (standard deviation 4.8 years), respectively. A total of 1029 patients (3.3%) died and the median ISS was 9 (IQR 5–13), with 6842 (21.9%) having an ISS >15. Within the group of patients who survived, 19.4% (5878/30,263) had an ISS >15 compared to 93.7% (964/1029) of those in the non-survivor group.

Discussion

There are two key issues in assessing the performance of a triage tool in paediatric major incidents. The tool must be sensitive enough to identify patients at greatest need, but at the same time must ensure the best use of available resources by delaying treatment for patients who do not require immediate attention [17]. This study has used a large dataset to evaluate the performance of five paediatric triage tools against two separate outcomes, survival (dead or alive) and injury severity

Conclusion

There is variation in the performance of existing triage tools used for predicting patient outcome and treatment priority in children. No single tool performed consistently well across all evaluated scenarios.

Conflict of interest statement

The authors declare: GDP, NS, SJBM and CLP had financial support from Department of Health for the submitted work; no other financial relationships with any organisations that might have an interest in the submitted work in the previous three years; no other relationships or activities that could appear to have influenced the submitted work.

Funding

This work was supported by a grant from the Department of Health Emergency Department. The funder had no role in the design, analysis, interpretation of the results, or the writing of the manuscript.

Acknowledgement

We would like to thank Fiona Lecky, Research Director, and Antoinette Edwards, Projects and Research Manager, as the Trauma Audit and Research Network for facilitating access to the TARN database.

References (20)

There are more references available in the full text version of this article.

Cited by (16)

  • Just-in-Time Training for Disaster Response in the Austere Environment

    2019, Clinical Pediatric Emergency Medicine
    Citation Excerpt :

    Patient evaluation and management will be dependent on available resources and local practices. Disaster principles of care to do the most good for the greatest number begins with accurate triage40-43 and often requires making difficult decisions throughout care. Diagnosis is often based only on history and physical exam given limited availability of laboratory testing and imaging.

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

    2017, Injury
    Citation Excerpt :

    In keeping with the literature on mortality following trauma, a surrogate systolic blood pressure measurement of 90 mmHg was used to represent the presence of a radial pulse for purposes of classification using START and Careflight [22,23]. We acknowledge that the use of surrogates (Table 5) is a limitation of our study, but one that is shared with other major incident triage publications and due to the nature of the JTTR is unavoidable [5,10,30]. In summary, we present the modified physiological triage tool, the first example of a statistically derived physiological triage tool for use in the military major incident setting.

  • A Scoping Review of Pediatric Mass-Casualty Incident Triage Algorithms

    2023, Disaster Medicine and Public Health Preparedness
View all citing articles on Scopus
View full text