Original ContributionsComparative analysis of multiple-casualty incident triage algorithms*,**
Introduction
The medical management of a major multiple-casualty incident (MCI) revolves around victim triage. Accurately identifying patients who will substantially benefit from early scene intervention or transport to definitive care may be the most important medical function at an MCI.1 However, in the setting of a major MCI, these patients are typically in the minority, with most persons being either uninjured, mildly injured, or deceased.2 Triage systems used in MCIs must therefore allow rapid identification of the critically injured without the need for detailed examinations of all involved persons.
Physiologic systems have been favored in the MCI setting because they aim to identify patients with current instability. Anatomic and mechanism of injury-based systems identify patients who have the potential to deteriorate; the triage priority is therefore based on potential, rather than actual, instability. This may result in a tendency to overtriage patients, thereby overwhelming the system.3 Physiologic systems, however, provide a snapshot of the patient's stability at the instant of triage and are based on the assumption that triage will be an ongoing process with frequent reassessments. Patients who are initially physiologically stable but deteriorate will therefore be identified in subsequent triage rounds.
Several systems have been advocated as triage tools designed to enable the rapid identification of critically injured persons from large numbers of patients who do not have immediately life-threatening injuries in an MCI.
The Triage Sieve methodology4 (Figure 1) has been widely advocated in the United Kingdom and has been adopted in parts of Australia.The physiologic variables used in Triage Sieve to stratify patients are respiratory rate and either capillary refill or heart rate, depending on the ambient weather and temperature conditions.
The Simple Triage and Rapid Treatment algorithm (Figure 2) is used widely in North America.Simple Triage and Rapid Treatment initially used the ability to obey commands, respiratory rate, and capillary refill to assign a triage category. Modifications were later recommended that substituted palpability of the radial pulse for capillary refill because data suggested it to be more reliable.5
The CareFlight Triage (Figure 3) algorithm assesses the ability to obey commands, the presence of respirations, and the palpability of the radial pulse.It differs from modified Simple Triage and Rapid Treatment in that there is no respiratory rate assessment, and level of consciousness is assessed first.
There are no published reports or studies addressing the accuracy of these systems. The aim of this study was to retrospectively determine the association between the physiologic variables used in these triage systems, both in isolation and when combined together as an algorithm, with severe injury requiring immediate life-saving intervention or urgent transport.
Section snippets
Materials and methods
Consecutive trauma patients presenting to the emergency departments of 2 trauma centers in New South Wales, Australia, were retrospectively identified from the hospital trauma registries. All patients transported in the calendar year of 1994 were included. Patients from later years could not be included because, starting in 1995, ambulances stopped collecting patient data on the basis of the Trauma Score and began using the Triage-Revised Trauma score. Because the new system did not use
Results
There were 1,192 patients who met the study inclusion criteria. Forty-eight patients were excluded because the ambulance case sheets had not been filed in the patients' medical record or the data were incomplete, leaving 1,144 patients who were included in the data analysis. Sixty-five percent of the sample were men, 35% were women, and there were 38 (3.4%) deaths. The median age of the sample population was 33 years (interquartile range, 21 to 53 years). Mechanism of injury is presented in
Discussion
The physiologic predictors with the strongest association with critical injury were the Motor Component of the Glasgow Coma Scale and systolic blood pressure in this population of designated adult trauma patients. This finding is similar to the findings of other studies that examined the relationship between physiologic variables and severe injury,10, 11 particularly the utility of a measure of level of consciousness.12, 13, 14 The odds ratios were high for the Motor Component of the Glasgow
Acknowledgements
Author contributions: AG conceived the study and all authors were involved in study design. AG performed the data abstraction. AG and KH managed the data, including quality control. AL provided statistical advice on study design and analyzed the data. AG drafted the manuscript and all authors contributed substantially to its revision. AG, KH, and CHS take responsibility for the paper as a whole.
We thank the trauma registries of Nepean and Westmead Hospitals for their assistance and Jack Chen
References (28)
- et al.
The lack of full correlation between the Injury Severity Score and the resource needs of injured patients
Ann Emerg Med
(1990) - et al.
The trauma triage rule: a new, resource-based approach to the prehospital identification of major trauma victims
Ann Emerg Med
(1990) - et al.
A prospective comparison of paramedic judgement and the trauma triage rule in the prehospital setting
Ann Emerg Med
(1994) - et al.
A review of the management of a major incident involving predominantly paediatric casualties
Injury
(1994) - et al.
Helicopter transport of injured children: system effectiveness and triage criteria
J Pediatr Surg
(1996) - et al.
Prehospital triage of the injured pediatric patient
J Pediatr Surg
(2000) Disaster planning and response
N Engl J Med
(1991)Disaster planning and organisation: casualty patterns in disasters
J World Assoc Emerg Disaster Med
(1987)Triage
Prehosp Immediate Care
(1997)
Disaster triage: START then SAVE—a new method of dynamic triage for victims of a catastrophic earthquake
Prehosp Disaster Med
Prehospital Trauma Life Support: Basic and Advanced
Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach
Biometrics
Multivariable Analysis: A Practical Guide for Clinicians
Cited by (0)
- *
Author contributions are provided at the end of this article.
- **
Address for reprints: Alan Garner, MSc, NRMA CareFlight, PO Box 159, Westmead 2145, Australia; E-mail [email protected].