EDMG541Wk3

Rawono1

Koenig_and_Schultz_s_Disaster_Medicine_Ch_12.pdf

Home >Law homework help >EDMG541Wk3

Triage

Christopher A. Kahn, E. Brooke Lerner, and David C. Cone

OVERVIEW

Introduction

One of the hallmarks of a disaster is that the immediate needs of the affected population exceed currently available resources. Intuitively, this leads to the question of how limited resources can be used to optimize patient outcomes. Triage is the alloca- tion of limited resources during a disaster. Although the concept of triage is applicable to all resources, the most commonly dis- cussed and most studied application is to patient care. In this context triage is the rapid evaluation of patients to determine the most appropriate level of care and treatment, given the limited resources at hand. For the remainder of this chapter, triage shall refer specifically to identifying the appropriate level of care for patients during a mass casualty incident.

Although researchers have investigated this type of triage more extensively than the triage of equipment or other resources, even patient triage is not well studied. As with many topics encompassed by disaster medical sciences, it is extremely dif- ficult to conduct randomized, controlled trials during an actual event; the use of other comparative study designs is also challeng- ing. As a consequence, there is no high-quality evidence indicat- ing which triage systems provide optimal resource utilization or maximal outcomes, or even whether existing triage systems are of any value in managing the scene, optimizing resource allocation among patients, or ensuring optimum outcomes. The majority of studies on disaster triage focus on how well triage systems can be applied, how well they work in drills, or how they can be modified for specific scenarios. These studies are usually conducted with simulated scenarios, and often performed on paper rather than using real or simulated victims.1–3 Studies and reports describ- ing the performance of triage systems during actual disasters are scarce and anecdotal; comparative and outcomes studies are lack- ing, although a few retrospective analyses have been conducted on individual patients comparing outcomes using different triage systems.4,5 Even some of the basic assumptions underlying the use of triage in mass casualty situations have not been tested. For example, there is no solid evidence to support the claim that triage systems provide improved outcomes related to morbidity or mortality when compared with randomly assigned or “first

come, first served” methods. Furthermore, it is not clear that various triage levels closely correlate with severity of illness or injury. These assumptions are, however, commonly accepted a priori, and provide a base from which a closer evaluation of triage systems can begin. Accordingly, this chapter will describe the existing triage systems but will be unable to recommend one system be used over the others due to the lack of scientific data.

History of Mass Casualty Triage It is generally accepted that triage was invented during the

Napoleonic Wars by Dominique Jean Larrey, surgeon-in-chief of Napoleon’s army from 1797–1815. Although the concept of “inventing” resource allocation is dubious, Larrey can be cred- ited with codifying a system for sorting battle casualties into categories based on urgency of evaluation. The word triage is derived from the French verb trier, meaning “to sort.”

Following this initial development of a casualty-sorting sys- tem, triage continued to evolve as a consequence of wartime experience for over a century. The first civilian triage systems were developed in the latter half of the twentieth century. In modern times, approximately 200 years after Larrey’s initial work in the field, several dozen systems exist worldwide, using more than 120 different types of triage labels and tools.

STATE OF THE ART

Vignette

On December 6, 2006 a manufacturing plant in Milwaukee, Wis- consin exploded due to a propane leak. Approximately 100 peo- ple worked in the affected building, which had begun evacuation just prior to the explosion. In the end, three workers died and dozens required emergency care. This industrial site had a large oxygen tank on the scene, making secondary explosion a concern. Furthermore, the plant used a significant number of chemicals, making the possibility of chemical contamination and the need for patient decontamination an additional consideration.

“When I arrived at the scene of the explosion there were a couple of hundred people walking towards us looking

174 Koenig and Schultz's Disaster Medicine : Comprehensive Principles and Practices, edited by Kristi L. Koenig, and Carl H. Schultz, Cambridge University Press, 2009. ProQuest Ebook Central, http://ebookcentral.proquest.com/lib/apus/detail.action?docID=564432. Created from apus on 2018-03-08 07:04:09.

C op

yr ig

ht ©

2 00

9. C

am br

id ge

U ni

ve rs

ity P

re ss

. A ll

rig ht

s re

se rv

ed .

TR I AG E ■ 175

shocked and dismayed over what had happened. The best analogy I can give is that it was like a scene from a zombie movie as these people moved towards us. We were stand- ing at the only exit from the factory compound and peo- ple were coming from all over the facility – narrowing into a single stream as they moved through the gate. While the majority were not injured, we could visually identify many with obvious injuries within the group of evacuees. Our first goal was to immediately separate the injured from the group so that we could triage them to trans- port units and receiving facilities. The most unexpected aspect of this event was that unlike many of the MCI drills I have witnessed, all of our patients were ambula- tory with the challenge being not to miss any patients as they were self-evacuating. The final patient count was 44, with two ending up in intensive care. None of the victims were lying on the ground waiting for us to assess them like they are during training. They moved towards us in a wave, asking for directions and aid.” – Battalion Chief Pepie Du De Voire, Milwaukee Fire Department –EMS Operations Chief

The preceding example illustrates several difficulties with the implementation and evaluation of mass casualty triage sys- tems. First, actual disasters and mass casualty events are often quite different from simulations and training exercises. Proto- cols developed without field testing may be difficult to apply or seem less relevant in a field situation. Second, the possibility of secondary events (such as additional explosions) or contamina- tion may complicate a triage situation. Third, the need to rapidly evaluate a large number of persons to promptly identify those most in need of immediate care almost certainly leads to a base- line level of inaccuracy in triage; there is no evidence to define an “acceptable miss rate,” which is likely both situation and agency dependent. Even the outcomes that would define accurate triage have not been identified. With these caveats in mind, individual triage systems are examined in more detail.

Triage Systems

Despite the large number of triage systems extant throughout the world, most have several features in common. A major- ity of these systems use a “walking filter” to identify quickly less severely injured patients and remove them from the imme- diate disaster zone; these patients are usually tagged “minor” or “green.” Patients not expected to survive are usually tagged “expectant,” “morgue,” or “black.” Residual patients are then categorized into the remaining triage levels. The use of color codes, generally black, red, yellow, and green to identify differing severity levels, is common. The primary differences between sys- tems rest in how patients are triaged to each level. Additionally, some systems use additional levels, colors, or classifications to further stratify victims. To date, no system has been shown con- clusively to be better than any other in terms of patient outcomes, scene management, or resource allocation. Little information is publicly available about some triage systems known to be in use worldwide, particularly in Europe. What follows is a descrip- tion of the triage systems that had sufficient available infor- mation to describe and discuss. These systems include START (Simple Triage and Rapid Treatment), Homebush Triage Stan- dard, CareFlight Triage, Triage Sieve, the Sacco Triage Method, the CESIRA Protocol, MASS Triage, and Military/NATO Triage.

Included is also a brief discussion of the SALT system, which has not been widely adopted at the time of this writing. A separate discussion of the secondary triage systems SAVE and Triage Sort, as well as the pediatric-specific systems, JumpSTART and the Pediatric Triage Tape is then provided.

START

Simple Triage and Rapid Treatment (START) is the most com- monly used triage system for handling multicasualty emergencies in the United States and has been adopted by components of the federal government.6 START is also used in Canada, Saudi Ara- bia, and parts of Australia and Israel. The START system was developed by the Newport Beach Fire and Marine Department and Hoag Hospital in Orange County, California in the early 1980s and is based on the NATO triage classification system.

START uses physiological parameters and is designed so that the healthcare provide can complete a patient assessment within 60 seconds or less and identify patients with immediate medical needs. Each patient is assessed and assigned to one of four color categories depending on the injuries (Table 12.1). A visible triage tag or ribbon is placed on each victim identifying the patient’s category for rescuers who will collect, treat, and/or transport. START is based on the ability to obey commands, respiratory rate, and capillary refill (or radial pulse in the modified version as described by Schultz and Koenig11,20).

Following a mass casualty incident, START triage begins with directing victims who are ambulatory to move to a safe area. These patients are tagged as “minor” by using a green label and, typically, are not more thoroughly assessed until the remaining, more seriously injured patients have been treated. Triage con- tinues in a systematic manner for the remaining victims. Triage categorization is based on three observations: respiration, per- fusion (or pulse in the modified system), and mental status. The mnemonic, “RPM,” has been created as a memory aid. Patients with no spontaneous respirations receive airway repositioning; if they remain apneic, they are tagged “deceased” by using a black label and receive no further interventions. Patients with respira- tions greater than 30 breaths per minute, capillary refill longer than 2 seconds (or lack of a radial pulse in the modified system), or unable to follow simple commands are tagged “immediate”

Table 12.1

START Triage

RED Any of the following: IMMEDIATE Respirations >30 breaths/min Priority I No palpable radial pulse

(or, in some systems, capillary refill time >2 s) Not able to follow commands

YELLOW DELAYED

Nonambulatory patients who do not meet black or red criteria

Priority II

GREEN MINOR

Able to walk to a designated safe area for further assessment

Priority III

BLACK Not breathing despite one attempt to open the airway DECEASED Priority IV

Koenig and Schultz's Disaster Medicine : Comprehensive Principles and Practices, edited by Kristi L. Koenig, and Carl H. Schultz, Cambridge University Press, 2009. ProQuest Ebook Central, http://ebookcentral.proquest.com/lib/apus/detail.action?docID=564432. Created from apus on 2018-03-08 07:04:09.

C op

yr ig

ht ©

2 00

9. C

am br

id ge

U ni

ve rs

ity P

re ss

. A ll

rig ht

s re

se rv

ed .

176 ■ CH R I S TO P H E R A. KA H N, E. BRO O K E LE R N E R, A N D DAV I D C. CO N E

by using a red label. The remaining patients are tagged “delayed” by using a yellow label.

Some areas use variations of the START triage system. For example, the Israeli triage system uses two additional categories and colors: blue for children, and gray for combined injuries such as chemical contamination and physical trauma.7 Additionally, most agencies use a “no radial pulse” (modified system) crite- rion rather than “capillary refill time more than 2 seconds” to compensate for difficulties in determining capillary refill time in cold or dark conditions; few agencies continue to use capillary refill for determination of circulatory status.

START only allows for two interventions to be made during the triage process: direct pressure for bleeding control (preferably applied by a bystander or another victim to keep the rescuer free for further triage), and basic airway opening maneuvers. It is also recommended that repeated assessments are made as often as possible since patient conditions may change.

Use of START triage has been described for two terrorist incidents: the attack on the World Trade Center in New York in 2001, and the bombing of the Alfred P. Murrah Federal Building in Oklahoma City in 1995.8–10 Use of START for two U.S. disas- ters, Hurricane Andrew in 1992 and the Northridge earthquake in 1994, has also been described.11 There are, however, no data in these descriptive papers regarding whether the system was used correctly or improved outcomes for patients in any of these events; a description of the 2001 World Trade Center attacks notes limitations of START due to concerns regarding personnel and structural safety, but does not detail triage accuracy for the few rescued patients.8

Homebush Triage Standard

The Homebush Triage Standard methodology was developed in Australia in 1999 as an attempt to unify varying triage proto- cols across the country.12 It is based on the START and SAVE20

(Secondary Assessment of Victim Endpoint) triage systems. It includes a fifth triage category called “dying,” which is given a white label. This category is meant to separate the dead (labeled black) from the dying, so that comfort care can be provided to those patients who are dying in an area where they are not surrounded by the deceased. The red category is assigned to patients who have no palpable radial pulse, are unable to fol- low commands, or have respirations more than 30 breaths per minute. Nonurgent and urgent patients are determined identi- cally to START’s minor and delayed patients, respectively. This system also uses geographical location rather than triage tags to indicate the patients’ conditions. In other words, patients are physically moved to the “level white” area, rather than placing a tag on their bodies to indicate their triage assign- ments. In addition to a color, each category has a designated standard phonetic alphabet code (e.g., alpha, bravo, charlie, delta, and echo) to facilitate radio communications (Table 12.2). Finally, in addition to primary triage, this system includes a secondary patient assessment to evaluate the extent of injuries and consider them in light of the available resources. This sec- ondary system is used to prioritize order of transport to the hospital.

Use of the Homebush triage system was documented in the Bali bombing on October 12, 2002, but again, only descriptive information is provided, with no data regarding triage accuracy or effects on any particular outcomes.13

Table 12.2

Homebush Triage Standard

RED Immediate

ALPHA Any of the following: Respirations > 30 breaths/min No palpable radial pulse Not able to follow commands

YELLOW Urgent

BRAVO Nonambulatory patients who do not meet black, white, or red criteria

GREEN Nonurgent

CHARLIE Able to walk to a designated safe area for further assessment.

WHITE Dying

DELTA Dying patients; may have a pulse, but no spontaneous respirations

BLACK Dead

ECHO Not breathing despite one attempt to open the airway

CareFlight Triage

The CareFlight system is a triage tool used in parts of Australia. Presence of breathing, level of consciousness, and presence of radial pulse determine the triage priority. This system is similar to START, with the notable exceptions that respiratory rate is not evaluated in CareFlight and that assessment of mental sta- tus (ability to follow commands) is done prior to assessment of circulation. CareFlight also uses a four-color system to iden- tify patients who should be triaged as unsalvageable, immediate, urgent, and delayed (Table 12.3). A retrospective 2001 study by Garner et al., compared START, modified START, and Triage Sieve with CareFlight and determined CareFlight to be more specific for critical injury (as defined by the modified Baxt cri- teria) and faster to administer.5 This difference was minimal, however; the difference between the upper limit of the 95% CI for specificity of the modified START system (using radial pulse) and the lower limit of that for the CareFlight being only 1%. Although the Triage Sieve was significantly less sensitive (with roughly equivalent specificity) for critical injury in this study, some have noted that failure to include the Triage Sort (the sec- ondary triage system that is meant to follow the Triage Sieve) as part of this algorithm may limit the applicability of these results to actual disasters.

Triage Sieve

Triage Sieve has been widely adopted in the United Kingdom, parts of Europe, and parts of Australia, and is accepted by NATO.

Table 12.3

CareFlight Triage

RED Immediate

Any of the following: Not able to follow commands No palpable radial pulse

YELLOW Urgent

Nonambulatory patients who do not meet black or red criteria

GREEN Delayed

Able to walk to a designated safe area for further assessment

BLACK Unsalvageable

Not breathing despite one attempt to open the airway

Courtesy of NRMA CareFlight.

C op

yr ig

ht ©

2 00

9. C

am br

id ge

U ni

ve rs

ity P

re ss

. A ll

rig ht

s re

se rv

ed .

TR I AG E ■ 177

Table 12.4

Triage Sieve

Priority I Immediate

Any of the following: Respirations <10 or >29 breaths/min Capillary refill time >2 s OR pulse >120 beats/min

Priority II Urgent

Nonambulatory patients who do not meet dead or immediate criteria

Priority III Delayed

Able to walk to a designated safe area for further assessment

Priority IV Dead

Not breathing despite one attempt to open the airway

Triage Sieve is similar to START in that a preliminary walking filter is followed by the use of respiratory rate and capillary refill or heart rate to classify patients into triage categories. Patients able to walk are classified as “delayed” priority III; patients who do not breathe following an attempt to open the airway are classified “dead” priority IV; and patients with a respiratory rate of less than 10 or more than 29 breaths per minute, capillary refill time of more than 2 seconds, or heart rate of more than 120 beats per minute are classified “immediate” priority I. All other patients are considered “urgent” priority II. Triage Sieve does not measure level of consciousness (Table 12.4).

Use of the Triage Sieve was documented in the London bomb- ings of July 7, 2005.14 In general, Triage Sieve is used as a primary “triage-to-treatment” algorithm and is followed by a secondary “triage-to-transportation” algorithm, the Triage Sort.

Sacco Triage Method

The Sacco Triage Method was developed in the United States, using a novel development method. The Delphi technique was used to estimate the chances of victim deterioration by obtaining consensus among a group of experts based on changes in physio- logical parameters of the patient.15 This triage system is intended to account for both the patient’s physiological parameters and the available resources.

The Sacco Triage Method uses a computer program to collect available resources in a database. A physiological score is then computed mathematically for each patient. This score considers the patient’s respiratory rate, pulse rate, and best motor response, assigns a coded value, and then sums these values to calculate the Sacco score.16 The developers report that this score can be calculated and a triage category assigned within 45 seconds once all data have been entered.

The victims are tagged and organized into three groups according to the score. Triage tags have a large clock face with numbers representing the score and can be easily seen by emer- gency medical services providers. The triage officer contacts a central dispatcher and provides information on number of vic- tims, the Sacco scores, ambulance-processing rate at the scene, and number of landing sites for helicopters. These data are entered into incident command software, which then produces the optimal triage strategy. This strategy defines the order in which victims are transported and treated and to which hospitals they are sent. The system also alerts the hospitals to the num- ber, severity, and scheduled arrival of patients. The Sacco Triage Method is proprietary, and accordingly, the specific details of how triage categories are determined are not publicly available

Table 12.5

CESIRA Protocol

Red Coscienza Unconscious

Emorragie Hemorrhaging

Shock Shock

Insufficienza respiratia Insufficient respiration

Yellow Rotture ossee Broken bones

Altro Other injuries

Green Walking

for review, research, or independent confirmation. Reliable infor- mation on current deployment and field success is not available, although implementation of this method in parts of Florida has been reported.17 The developers report that the Sacco score accu- rately predicts a patient’s survivability from trauma, although this has not been prospectively validated.17

CESIRA Protocol

The CESIRA Protocol, developed in 1990, has three basic cate- gories: red, which includes patients who are unconscious, hem- orrhaging, in shock, or having respiratory insufficiency; yellow for patients with broken bones and other injuries; and green for victims able to walk. CESIRA is the Italian acronym for the words describing these injuries (Table 12.5). CESIRA does not contain a dead category; nonphysicians are not legally autho- rized to declare death in Italy, and the system is designed for prehospital use when physicians are not present.

MASS Triage

MASS (Move, Assess, Sort, Send) is a disaster triage system that utilizes the U.S. military triage categories with a simple system to triage large numbers of casualties quickly in a mass casualty incident. It was developed and is now taught by participants in the National Disaster Life Support Foundation. Although MASS is based on START, it sorts patients into a triage category before individual assessments are done. The first, or “Move” stage directs victims who are able to walk to go to a designated area; these victims are marked “minimal”/green. The victims who can- not walk are asked to move an arm or leg; those able to follow the command and move an extremity are “delayed”/yellow. If vic- tims cannot move when asked, they are assessed and assigned to either the “immediate” or “expectant” groups. The Assess stage of MASS is compatible with other triage systems, such as START. Additionally, the Assess stage may include a subjective compo- nent directing persons with an expected fatal injury (regardless of expected duration of survival, such as fatal doses of radiation or 100% total body surface area burns) to the expectant category. The “Sort” stage is a further, subjective categorization. “Send” is the transport phase. Transport may be conducted according to priorities determined in the Sort phase.

Military Triage/NATO Triage

The main objective of military triage is to treat and return injured soldiers to the front lines as soon as possible. For those who

C op

yr ig

ht ©

2 00

9. C

am br

id ge

U ni

ve rs

ity P

re ss

. A ll

rig ht

s re

se rv

ed .

178 ■ CH R I S TO P H E R A. KA H N, E. BRO O K E LE R N E R, A N D DAV I D C. CO N E

Table 12.6

Military Triage

P1 T1 IMMEDIATE: Life-threatening injuries must be treated within first hour. Good chance of survival

P2 T2 DELAYED: Delay in treatment, can wait a few hours. Stabilization

P3 T3 MINIMAL: Walking, treatment may be delayed for several hours.

P1 – Hold T4 EXPECTANT: Significant resources needed to treat patient. Signs of impending death

Dead Dead Dead

cannot return to duty, medical focus is on wound debridement, limb salvage, and preservation of life. Military triage is based on the North American Treaty Organization (NATO) triage classi- fication, a subjective categorization based on expected survival and resource utilization. All NATO member countries follow a standardized triage system for their military operations provid- ing consistency for multinational operations.

Military triage begins with the immediate sorting of patients according to type and severity of injury and likelihood of sur- vival, and the establishment of priorities for treatment and evac- uation to ensure medical care of the greatest benefit to the largest number. Most military triage systems use the “T” (treatment) system, T1, T2, T3, T4, and dead. Others such as the British Military system use the “P” (priority) system – P1, P2, P3 and P1-hold.18 Holding areas are developed for victims according to their injuries following initial evaluation. Patients are treated and stabilized until they can be transported to a medical facility. The classification scheme is subjective, based on the experience of the triage provider rather than specified physiological criteria (see Table 12.6).

SALT Triage

A more recent U.S. federally funded project to examine exist- ing triage systems has resulted in the development of the SALT (Sort, Assess, Lifesaving measures, Treat/Transport) system. After determining that no existing system has sufficient scientific sup- port to recommend its use, a workgroup used the limited existing data and expert opinion to craft the SALT system, attempting to incorporate the best features of the existing systems. SALT is intended to serve as a national all-hazard mass casualty initial triage standard for all patients (e.g., adults, children, special pop- ulations).19

SALT begins with a global sorting of patients to prioritize them for individual assessment. Patients who are able to walk are instructed to walk to a designated area, and are assigned last priority for individual assessment. Those who remain are asked to wave or are observed for purposeful movement. Those who do not move (i.e., are still) and those with obvious life threats are assessed first because they are the most likely to need life-saving interventions.

The individual assessment begins with limited rapid life- saving interventions (LSIs). These are only performed if the intervention is within the responder’s scope of practice, and only if the necessary equipment is immediately available.

The recommended LSIs include: controlling major hemor- rhage, opening the airway, providing chest decompression, or using autoinjector antidotes. If the patient is a child and not breathing, the provider can consider giving two rescue breaths.

After any needed LSIs are provided, patients are prioritized for treatment and/or transport by assigning them to one of five categories. Patients who are not breathing even after LSIs are attempted are triaged as dead. Patients who have mild injuries that are self-limited if not treated and can tolerate a delay in care without increasing their risk of mortality are triaged as minimal. Patients who do not obey commands, do not have a peripheral pulse, are in respiratory distress, or have uncontrolled major hemorrhage are triaged as immediate. Within this group of immediate patients providers should also determine whether a patient has injuries that are likely to be incompatible with life given the currently available resources. If so, then the provider triages that patient as expectant rather than immediate. The remaining patients are triaged as delayed (Figure 12.1). To assist with identification of patients SALT recommends that dead be symbolized by the color black, expectant by gray, immediate by red, delayed by yellow, and minimal by green.

Secondary Triage Systems In situations where projected out-of-hospital time is exten-

sive, secondary triage systems may be used to further catego- rize victims for prioritization of transport. In severely resource- constrained environments, these systems are designed to consider likelihood of a positive outcome in addition to the urgency of treatment and the amount of projected resources that are needed. Consequently, the resulting priority decisions for transport (should an opportunity become available) and other resource utilization may not be in complete concordance with the primary triage.

SAVE Triage

SAVE, or Secondary Assessment of Victim Endpoint, is used with the START triage algorithm. This system uses objective and subjective individualized considerations for likelihood of sur- vival and resource utilization to direct limited treatment options in the field and prioritize transport of victims who are most likely to benefit from advanced care. SAVE is designed to limit use of medical resources by defining those victims with poor prognoses and those whose outcomes are unlikely to improve with immediate care. Also, consideration is given to healthcare workers and other special categories of victims who, with mini- mal treatment (e.g., splinting of a sprained ankle), can assist in the disaster response by increasing the available resources (i.e., skilled personnel) supporting medical care. Treatment is pri- oritized toward victims whose likelihood of survival (if given medical care) is more than 50% and who would benefit from immediate intervention. These likelihoods are calculated based on prognostic tools including a limb salvage score, the Glasgow Coma Scale (GCS), and data on survivability after burns. The full details of this system are too extensive to present here but can be found in the original 1996 manuscript.20

Triage Sort

Often paired with the Triage Sieve, the Triage Sort is a secondary triage system that uses the Revised Trauma Score (RTS) to catego- rize patients into immediate, urgent, and delayed categories. The RTS is derived from the GCS, blood pressure, and respiratory

C op

yr ig

ht ©

2 00

9. C

am br

id ge

U ni

ve rs

ity P

re ss

. A ll

rig ht

s re

se rv

ed .

TR I AG E ■ 179

Figure 12.1. SALT Mass Casualty Triage.

rate. This system is generally first applied to patients initially “sieved” into the immediate category to further stratify them when transportation is limited. Attention is then given to urgent and finally delayed patients.

Pediatric Triage Systems The physiological and anatomical differences between chil-

dren and adults are significant. Children are more prone to head injuries, airway obstruction, and hypothermia. They have pro- portionally less blood volume than adults, and very young chil- dren may be unable to walk, communicate verbally, or cooperate with instructions. Care providers may have difficulty obtain- ing blood pressure readings on children, and the act of triaging children may cause emotional challenges to rescuers beyond the already stressful scenario of disaster triage. With these differences in mind, several triage systems have been developed specifically for use with pediatric patients; however, like the general triage systems discussed previously, there are presently no validated pediatric triage tools.

JumpSTART

JumpSTART is designed to be an objective, physiologically appro- priate tool for triaging children younger than 8 years of age. JumpSTART was developed by Romig in 1995 and modified in 2001, and is a modification of the START triage system.21 Three key changes were made to the START system based on children being more likely to suffer respiratory arrest than adults, having different respiratory rates, and young children being unable to follow commands.

In the JumpSTART system, when a child is identified as having a pulse but not breathing, the rescuer is instructed to

reposition the airway and give five rescue breaths (called “jump- start” breaths). A child who is still not breathing after the rescue breaths would be labeled black, whereas a child whose breathing is present at this point would be labeled red.

When deciding whether a child should be triaged into a red or yellow category, responders must recognize the different values for age-adjusted normal and abnormal respiratory rates and the limited ability to follow commands in the pediatric population. For children, a respiratory rate less than 15 or greater than 45 indicates that a red label should be assigned. If the respiratory rate is between 15 and 45, a yellow label is applied. Although normal respiratory rates in children vary by age, this simplified rule was chosen to minimize confusion and maximize utility. For assessing mental status, young children may lack the capa- bility of responding to commands. Therefore, JumpSTART uses the AVPU (Alert/Verbal/Painful/Unresponsive) tool rather than response to commands for the mental status component of the algorithm.

Pediatric Triage Tape

The Pediatric Triage Tape (PTT) is derived from Triage Sieve. It is used throughout the United Kingdom and parts of Europe, India, Australia, and South Africa. The PTT is designed to complement any existing triage labeling system. If the child is walking or an infant is alert and moving all limbs, the tape is not necessary, and the patient is labeled “delayed” (green). If a child is not walking or moving appropriately, the tape is used to measure the length of the child, similar to the Broselow Tape or other length- based algorithms. The PTT is divided into five length blocks; each block contains the algorithm for Triage Sieve, modified for age-appropriate respiratory and heart rate parameters.22,23

C op

yr ig

ht ©

2 00

9. C

am br

id ge

U ni

ve rs

ity P

re ss

. A ll

rig ht

s re

se rv

ed .

180 ■ CH R I S TO P H E R A. KA H N, E. BRO O K E LE R N E R, A N D DAV I D C. CO N E

Table 12.7

JumpSTART Triage

RED IMMEDIATE

Any of the following: Respirations <15 or >45 breaths/min No palpable radial pulse Inappropriate response to pain, posturing, or unresponsive to stimuli

YELLOW DELAYED

Nonambulatory patients who do not meet black or red criteria

GREEN MINOR

Able to walk to a designated safe area for further assessment. This includes children who are developmentally unable to walk; however, these children must be the first to be assessed when the green patients are reevaluated

BLACK DECEASED

Not breathing despite one attempt to open the airway and, in patients with a palpable pulse, after provision of five rescue breaths

Courtesy of Dr. Lou E. Romig.

Comparison of Pediatric Triage Systems

CareFlight triage can be used in both children and adults. In a study by Wallis et al., CareFlight, PTT, START, and Jump- START were compared in 3,461 injured children presenting to the Trauma Unit of the Red Cross Children’s Hospital in Cape Town, South Africa.4 The same modified Baxt criteria used in the Garner comparative study were used to define critical injury. Overall, CareFlight showed the best performance in sensitivity and specificity; 95% confidence intervals for sensitivity over- lapped for PTT, CareFlight, and START, whereas specificity 95% confidence intervals overlapped only for PTT and CareFlight, with JumpSTART very slightly lower. Sensitivity was very poor (approximately 1%) for JumpSTART, compared with approxi- mately 39%–46% for the other algorithms. This is the only pub- lished report that could be identified in which pediatric triage systems were compared. Therefore, it is currently not possible to recommend any of the pediatric triage systems as superior to any of the others.

Use of Common Scales for Triage

One might ask whether any of the individual assessment items contained in some of the aforementioned triage systems might provide comparable sensitivity and specificity in identifying the more seriously injured patients. It seems plausible that assess- ment of a single “score” or feature may prove as accurate as assessment of several parameters within a multifeature triage system. If validated, the use of a single value would be sim- pler and more rapid, and thus preferable. The most studied of these is the motor component of the GCS. A 1998 investigation demonstrated that the major discriminatory power of the GCS lies in the motor component. In this retrospective study corre- lating mortality with the first GCS score obtained in the field in approximately 1,200 patients, the motor score alone performed slightly better than the summed three-part GCS.24 Several large studies have subsequently supported this finding.25–28 A retro- spective study of almost 30,000 patients from a state trauma registry suggests that separating patients into those who can fol- low simple commands (Glasgow motor score = 6) from those who cannot (Glasgow motor score from 1 to 5) provides the best

discriminatory capability, perhaps providing a simple and ele- gant alternative to a multistep triage algorithm.26 Research has demonstrated that a GCS motor component score of 5 or less is correlated with worse outcomes in trauma patients.27

Respiratory rate is another commonly used trauma triage criteria. There is limited evidence to support its use in the trauma setting. The frequently used cutoff of 30 breaths per minute as the respiratory criterion is arbitrary and not supported in current literature; it appears that lowering this cutoff to the mid-20s may improve overall triage performance.29

Outcome Measures for Triage Research

Although a large number of mass casualty triage systems exist worldwide, many with similar features, there are no currently accepted reference standards that define key outcome measures. There are essentially two categories of outcomes that could be used in assessing how triage affects patient outcome: patient- based scoring systems, such as Injury Severity Score (ISS), (e.g., does triage system “x” appropriately assign all patients with an ISS greater than “y” to the “red” category?) and resource- based systems (e.g., were all patients who required surgery or a blood transfusion in the first 2 hours after arrival at the hospital correctly assigned to the “red” category?). In 1990, Baxt et al., defined a set of criteria based on resource utilization within 2 days of arrival at the hospital to define a patients’ level of serious injury, in an attempt to correlate resource utilization to ISS.30

This was refined by Garner et al., in 2001 to reflect more accu- rately which patients in the prehospital setting were truly in need of immediate care. Although this landmark study retrospectively examined individual trauma patients to see how each of four systems (CareFlight, START, modified START, and Triage Sieve) performed, at the time of this writing, only one study has been published that examines the real-time performance of any mass casualty triage system with reference to these a priori criteria. The article discusses the performance of START triage at a train collision involving approximately 150 victims; the authors found that the “walking filter” appeared to function well in defining less severely injured patients.31

Some researchers suggest that scored scales, such as the ISS and RTS, could be used as reference criteria (gold standards) to assess the accuracy of triage instruments. It remains unclear which of these scores, if any, accurately predict outcomes or the need for resource utilization in a mass casualty setting. Further- more, there is no agreement within specific scores as to which cutoffs should be used to define “correct” use of a triage instru- ment (e.g., what would “y” be in the previous example using ISS?). Although the reproducibility and applicability to a wide range of trauma patients makes tools such as ISS and RTS attrac- tive, their validity in assessing outcomes in the mass casualty setting is unknown.

Measures other than clinical patient outcome may be impor- tant in judging and comparing trauma systems. How efficiently scarce field resources are used, how quickly patients are trans- ported to hospitals, and costs of training, program implemen- tation, and competency maintenance for personnel in a given system may be of tremendous importance. Essentially no work has been done to examine or define these nonclinical outcomes.

At this time, no specific recommendation based on strong evidence can be made to support any one triage system over another. Although not studied, it seems likely that use of a sin- gle standard system across an entire region is likely to improve

C op

yr ig

ht ©

2 00

9. C

am br

id ge

U ni

ve rs

ity P

re ss

. A ll

rig ht

s re

se rv

ed .

TR I AG E ■ 181

Table 12.8

Incident Type Triage Goal Difference

Single patient incident Optimize individual patient outcome by providing all resources required to meet the patient’s needs

No consideration of other patients

Multiple casualties Prioritize patients for appropriate treatment/transport so they receive needed resources in sufficient time to reduce morbidity and mortality

Resources are devoted to patients according to priority, but patients receive all the care they need

Disaster Prioritize patients for appropriate treatment/transport so they receive needed resources in sufficient time to reduce morbidity and mortality, but also ensure that scarce resources are utilized to provide the best outcome at a population level

Resources are devoted to patients according to priority, but patients who are unlikely to survive, given the available resources, are given a low priority for treatment/transport

interoperability during mass casualty responses. In the absence of additional evidence to recommend a particular methodology, choice of this system will likely be based on existing resources, the need for retraining, and flexibility in the face of various disaster scenarios. Specific attention to chemical, biological, and radio- logical/nuclear (CBRN) events is a critical component of state- of-the-art triage systems and must be considered when choosing a triage methodology. Although proposed, no CBRN-capable mass casualty triage system has been adequately studied.32

Other Triage Considerations

Differentiation of Disaster Triage from Other Triage Modalities

It is important to distinguish disaster triage from other types of triage, including daily emergency department triage, single- patient trauma triage, and multicasualty incident triage when resources are not exceeded and transportation and communica- tion infrastructures are intact. Single-patient trauma triage (the process of determining whether a given single trauma patient, such as a motor vehicle crash victim, needs to be transported to a trauma center or not, often using the National Consensus Field Triage developed by the CDC) attempts to match the patient’s clinical needs with the correct resources (trauma center vs. a hospital that is not a trauma center).33 Unlike the mass casualty setting, however, this type of triage does not weigh the relative needs of several patients to determine who will get a relatively scarce resource, but instead helps determine whether a single patient might need the resources of the trauma center.

When personnel in the prehospital setting or emer- gency department are confronted with multiple patients, these providers prioritize care for individuals based on acuity. In these multicasualty situations, as well as in daily emergency depart- ment triage, the goal is to determine which patients can wait for treatment without increasing their risk of morbidity and mor- tality and which need immediate attention.

In disaster situations when resources are exceeded, the need to prioritize patients must include a rationing of supplies, shifting the focus from ensuring that each individual receives the best possible care to ensuring the population as a whole experiences the best possible outcome. In other words, the focus is on pro- viding the most efficient care for the greatest number of people (Table 12.8). In some cases, this will include identifying patients

who are not expected to survive and making the decision that no resources beyond comfort measures will be used for these individuals.

Context-Specific Triage Although the majority of disasters to date have resulted in

traumatic injuries from explosions, collisions, collapses, and other releases of kinetic energy, other scenarios that involve CBRN agents, or a combination of agents, are also possible; responders must be prepared to manage any type of event. Most of the triage systems described previously focus on triage of trauma victims in which kinetic or thermal energy is the only cause of their injury. It is possible that applying these triage sys- tems to other types of events may not be feasible and may not improve patient outcomes. Developing an all-hazard triage algo- rithm may be difficult because it must be scientifically valid for all types of threats while remaining simple to use, accurate, repro- ducible, and rapid. Cone and Koenig proposed a basic triage algo- rithm, based on START triage, with simple modifications made for each category of CBRN event. These modifications address the need for further triage to maximize patient outcomes while protecting response personnel.33 The SALT triage system is also intended to be an all-hazard triage method. In addition, timing of decontamination versus transport may be of critical impor- tance in balancing patient care needs with personnel protection and is an important component of mass casualty triage protocols for CBRN events.

Biological mass casualty triage is likely to present the most radical departure from “routine” mass casualty trauma triage. Rather than a single, large peak of acutely injured individuals, response personnel may be faced with a prolonged presenta- tion of victims in addition to the baseline number of patients already in the system. Victims are likely to have no trauma and there may not be a specific disaster scene. Rather than triage by acuity, a system that groups patients by exposure status may be necessary. Such a strategy has been proposed by Burkle and advocates triaging patients into five groups: susceptible but not exposed, exposed but not yet infectious, infectious, removed by death or recovery, and protected by vaccination or prophylactic medication.34 Triage tools for biological events require develop- ment in concert with infectious disease, public health, and other experts in epidemiology and mass patient care. These tools may eventually take the form of telephone screening or broadcast

C op

yr ig

ht ©

2 00

9. C

am br

id ge

U ni

ve rs

ity P

re ss

. A ll

rig ht

s re

se rv

ed .

182 ■ CH R I S TO P H E R A. KA H N, E. BRO O K E LE R N E R, A N D DAV I D C. CO N E

messages for self-evaluation in affected areas. It is possible that referral to hospitals for triage evaluation may not be prudent due to risk of spreading the contagion. Given the broad uncer- tainty in biological mass casualty triage and the substantial dif- ferences between triage tools useful for trauma and those useful for purely biological events, a further discussion of these tools is deferred.

RECOMMENDATIONS FOR FURTHER RESEARCH

Current triage systems suffer from significant limitations.

■ Lack of scientific validation ■ Lack of standardization and interoperability ■ Absence of flexibility in addressing nontraumatic disaster

scenarios

It is vital that methodologically sound, outcomes-based research be conducted to address these limitations. In particular, specific questions that should be addressed within the field of disaster triage include

■ What is the appropriate outcome measure for studying triage systems?

■ Using this measure, which system (if any) is superior? ■ What is the best way to rapidly sort large numbers of non-

traumatic victims, such as victims of chemical, biological, or radiological attacks, for prioritization of treatment and transport?

■ Is it possible, or even desirable, to have a “one size fits all” algorithm that would cover all of these possibilities, or are the differences between event types too pronounced?

In the meantime communities should select the triage system that is most appropriate for their circumstances, keeping in mind that interoperability within a region will be enhanced by using a stan- dardized triage system. Furthermore, whichever triage system is selected, it should be used and practiced regularly, and whatever tools are needed for implementation should be readily accessible and familiar to the providers.

REFERENCES

1. Sanddal TL, Loyacono T, Sanddal ND. Effect of JumpSTART training on immediate and short-term pediatric triage perfor- mance. Pediatr Emerg Care. 2004;20(11):749–753.

2. Risavi BL, Salen PN, Heller MB, et al. A two-hour interven- tion using START improves prehospital triage of mass casualty incidents. Prehosp Emerg Care. 2001;5(2):197–199.

3. Schenker JD, Goldstein S, Braun J, et al. Triage accuracy at a multiple casualty incident disaster drill: the Emergency Medical Service, Fire Department of New York City experience. J Burn Care Res. 2006;27(5):570–575.

4. Wallis LA, Carley S. Comparison of paediatric major incident primary triage tools. Emerg Med J. 2006;23(6):475–478.

5. Garner A, Lee A, Harrison K, Schultz CH. Comparative analysis of multiple-casualty incident triage algorithms. Ann Emerg Med. 2001;38(5):541–548.

6. Domestic Preparedness Training Program Instructor Guide. Chemical and Biological Defense Command, Department of Defense; 1998.

7. Mor M, Waisman Y. Triage principles in multiple casualty sit- uations involving children – the Israeli experience. Pediatric Emergency Medicine Database (serial online) August 2002. Avail- able at: http://www.pemdatabase.org/files/triage.pdf. Accessed February 26, 2009.

8. Asaeda G. The day that the START triage system came to a STOP: observations from the World Trade Center disaster. Acad Emerg Med. 2002;9(3):255–256.

9. Cook L. The World Trade Center attack. The paramedic response: an insider’s view. Crit Care (London). 2001;5(6):301– 303.

10. Teague DC. Mass casualties in the Oklahoma City bombing. Clin Orthop Rel Res. 2004;(422):77–81.

11. Schultz CH, Koenig KL, Noji EK. A medical disaster response to reduce immediate mortality after an earthquake. N Engl J Med. 1996;334(7):438–444.

12. Nocera A, Garner A. An Australian mass casualty incident triage system for the future based upon triage mistakes of the past: the Homebush Triage Standard. Aust NZ J Surg. 1999;69(8):603– 608.

13. Tran MD, Garner AA, Morrison I, Sharley PH, et al. The Bali bombing: civilian aeromedical evacuation. Med J Aust. 2003;179(7):353–356.

14. Hines S, Payne A, Edmondson J, Heightman AJ. Bombs under London. The EMS response plan that worked. JEMS. 2005;30(8):58–60, 62, 64–57.

15. Sacco WJ, Navin DM, Fiedler KE, et al. Precise formulation and evidence-based application of resource-constrained triage. Acad Emerg Med. 2005;12(8):759–770.

16. Lindsey J. New triage method considers available resources. JEMS. 2005;30(7):92–94.

17. King D. New triage method seeks consistency, reduced mortal- ity. Merginet; February 2004. Available at: http://www.merginet .com/clinical/trauma/saccotriage.cfm. Accessed February 26, 2009.

18. Hodgetts TJ. Triage: a position statement. Available at: http:// ec.europa.eu/environment/civil/prote/pdfdocs/disaster med final 2002/d6.pdf. Accessed February 26, 2009.

19. Centers for Disease Control and Prevention. SALT Triage. For: Terrorism Injuries Information, Dissemination, and Exchange Project. Atlanta, GA; September 22, 2007.

20. Benson M, Koenig KL, Schultz CH. Disaster triage: START, then SAVE – a new method of dynamic triage for victims of a catastrophic earthquake. Prehosp Disaster Med. 1996;11(2):117– 124.

21. Romig LE. Pediatric triage. A system to JumpSTART your triage of young patients at MCIs. JEMS. 2002;27(7):52–58, 60– 53.

22. Wallis LA, Carley S. Validation of the Paediatric Triage Tape. Emerg Med J. 2006;23(1):47–50.

23. Hodgetts TJ, Hall J, Maconochie I, et al. Paediatric triage tape. Prehosp Immediate Care. 1998;2:155–159.

24. Jagger J, Jane JA, Rimel R. The Glasgow Coma Scale: to sum or not to sum? Lancet. 1983;2(8341):97.

25. Al-Salamah MA, McDowell I, Stiell IG, et al. Initial emer- gency department trauma scores from the OPALS study: the case for the motor score in blunt trauma. Acad Emerg Med. 2004;11(8):834–842.

26. Meredith W, Rutledge R, Hansen AR, et al. Field triage of trauma patients based upon the ability to follow commands: a study in 29,573 injured patients. J Trauma. 1995;38(1):129–135.

C op

yr ig

ht ©

2 00

9. C

am br

id ge

U ni

ve rs

ity P

re ss

. A ll

rig ht

s re

se rv

ed .

TR I AG E ■ 183

27. Ross SE, Leipold C, Terregino C, O’Malley KF. Efficacy of the motor component of the Glasgow Coma Scale in trauma triage. J Trauma. 1998;45(1):42–44.

28. Healey C, Osler TM, Rogers FB, et al. Improving the Glas- gow Coma Scale score: motor score alone is a better predictor. J Trauma. 2003;54(4):671–678; discussion 678–680.

29. Husum H, Gilbert M, Wisborg T, et al. Respiratory rate as a pre- hospital triage tool in rural trauma. J Trauma. 2003;55(3):466– 470.

30. Baxt WG, Upenieks V. The lack of full correlation between the Injury Severity Score and the resource needs of injured patients. Ann Emerg Med. 1990;19(12):1396–1400.

31. Kahn CA, Schultz CH, Miller K, Anderson CL. Does START triage work? An outcomes-level assessment of use at a mass casualty event. Ann Emerg Med 2009; in press. DOI: 10.1016/j.annemergmed.2008.12.035

32. Cone DC, Koenig KL. Mass casualty triage in the chemical, biological, radiological, or nuclear environment. Eur J Emerg Med. 2005;12(6):287–302.

33. Resources for Optimal Care of the Injured Patient: 2006: American College of Surgeons Committee on Trauma; 2006.

34. Burkle FM Jr. Population-based triage management in response to surge-capacity requirements during a large-scale bioevent disaster. Acad Emerg Med. 2006;13(11):1118–1129.

C op

yr ig

ht ©

2 00

9. C

am br

id ge

U ni

ve rs

ity P

re ss

. A ll

rig ht

s re

se rv

ed .