See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/40040999

Use of SALT Triage in a Simulated Mass-Casualty Incident

Article  in  Prehospital Emergency Care · January 2010

DOI: 10.3109/10903120903349812 · Source: PubMed

CITATIONS READS
59 3,655

4 authors, including:

E Brooke Lerner Richard B Schwartz

Medical College of Wisconsin Augusta University
187 PUBLICATIONS   4,827 CITATIONS    18 PUBLICATIONS   536 CITATIONS   

SEE PROFILE SEE PROFILE

Ronald G Pirrallo
Medical College of Wisconsin
80 PUBLICATIONS   4,791 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

SALT Triage View project

Peds field triage View project

All content following this page was uploaded by Ronald G Pirrallo on 21 May 2014.

The user has requested enhancement of the downloaded file.

 USE OF SALT TRIAGE IN A SIMULATED MASS-CASUALTY INCIDENT
E. Brooke Lerner, PhD, Richard B. Schwartz, MD, Phillip L. Coule, MD,
Ronald G. Pirrallo, MD, MHSA

ABSTRACT difficult. After the drill, no one reported that SALT triage
was more difficult to use. Conclusion. We found that assess-
Objectives. To determine the accuracy of SALT (sort–assess– ments using SALT triage were accurate and made quickly
lifesaving interventions–treatment/transport) triage during during a simulated incident. The accuracy rate was higher
a simulated mass-casualty incident, the average time it takes than those published for other triage systems and of similar
to make triage designations, and providers’ opinions of speed. Providers also felt confident using SALT triage and
SALT triage. Methods. Seventy-three trainees participating found it was similar or easier to use than their current triage
in one of two disaster courses were taught to use SALT triage protocol. Using SALT triage during a drill improved confi-
during a 30-minute lecture. The following day they partici- dence. Key words: disaster; triage; emergency medical ser-
pated in teams, in one of eight simulated mass-casualty in- vices: triage; SALT triage
cidents. For each incident trainees were told to assess and
prioritize all victims. Each scenario comprised 28 to 30 vic- PREHOSPITAL EMERGENCY CARE 2010;14:21–25
tims, including 10 to 11 moulaged manikins and 18 to 20
moulaged actors. Each victim had a card that stated the INTRODUCTION
victim’s respiratory effort, pulse quality, and ability to fol-
low commands. Initial and final assigned triage categories The process of sorting multiple casualties for treat-
were recorded and compared with the intended category. Ten ment was first described over 200 years ago and to-
of the victims were equipped with stopwatches to measure day is known as mass-casualty triage.1,2 Civilian emer-
the triage time interval. Timing began when the trainee ap- gency medical services (EMS) providers are routinely
proached the victim and ended when the trainee verbalized trained in a method of prioritizing patients for treat-
his or her triage designation. The times were averaged and ment and/or transport. Triage becomes even more crit-
standard deviations were calculated. After the drill, trainees
ical whenever they are faced with more patients than
were asked to complete a survey regarding their experi-
ence. Results. There were 217 victim observations. The ini-
EMS providers. This can occur during a large-scale dis-
tial triage was correct for 81% of the observations; 8% were aster or may more commonly occur during a smaller
overtriaged and 11% were undertriaged. The final triage was event such as a multivehicle crash. Within the United
correct for 83% of the observations; 6% were overtriaged and States, the specific system of mass-casualty triage a
10% were undertriaged . The mean triage interval was 28 sec- prehospital care provider uses is largely dependent on
onds (±22; range: 4–94). Nine percent reported that prior to local or regional protocols, with little consistency or in-
the drill they felt very confident using SALT triage and 33% teroperability between jurisdictions.
were not confident. After the drill, no one reported not feel- A recent Centers for Disease Control and Pre-
ing confident using SALT triage, 26% were at the same level vention (CDC)-sponsored panel developed a pro-
of confidence, 74% felt more confident, and none felt less posed national guideline for mass-casualty triage
confident. Before the drill, 53% of the respondents felt SALT
called SALT (sort–assess–lifesaving interventions–
triage was easier to use than their current disaster triage
protocol, 44% felt it was similar, and 3% felt it was more
treatment/transport) triage (Fig. 1).3 They recom-
mended that across the United States a uniform stan-
dard be adopted so that all EMS providers would use
Received March 23, 2009, from the Department of Emergency a similar language and process when responding to a
Medicine, Medical College of Wisconsin (EBL, RGP), Milwaukee, mass-casualty event. The proposed guideline incorpo-
Wisconsin; and the Department of Emergency Medicine, Medical
rated aspects from all of the existing triage systems to
College of Georgia (RBS, PLC), Augusta, Georgia. Revision received
June 2, 2009; accepted for publication June 9, 2009. create a single overarching guide for unifying mass-
casualty triage. This nonproprietary guideline can be
Presented at the National Association of EMS Physicians annual
meeting, Jacksonville, Florida, January 2009. used for any patient regardless of age or physical or
mental limitations. The concept has been endorsed by
The authors would like to acknowledge the cooperation of the Na-
tional Disaster Life Support Foundation in the conduct of this study. many national organizations,4 but scientific validation
Dr Lerner was partially supported by CDC grant R49/CE001175. of the guideline is still needed. Key aspects of SALT
The authors report no conflicts of interest. The authors alone are re- triage include a global sorting of patients using voice
sponsible for the content and writing of this paper. commands so that individual assessment can be prior-
Address correspondence and reprint requests to: E. Brooke Lerner,
itized, lifesaving interventions are considered first dur-
PhD, Department of Emergency Medicine, Medical College of Wis- ing individual assessment, and an expectant category
consin, 9200 West Wisconsin Avenue, Milwaukee, WI 53226. e-mail: is included that is dependent on resource availability.3
eblerner@mcw.edu The objective of this study was to evaluate SALT
doi: 10.3109/10903120903349812 triage. This study had three specific aims: 1) to

21
22 PREHOSPITAL EMERGENCY CARE JANUARY/MARCH 2010 VOLUME 14 / NUMBER 1

FIGURE 1. The SALT triage guideline. LSI = lifesaving intervention; SALT = sort–assess–lifesaving interventions–treatment/transport.

determine trainee accuracy when using SALT triage The day after the SALT training was provided,
during a simulated mass-casualty incident, 2) to all trainees participated in one of eight simulated
determine the average time required to make triage mass-casualty incidents. The trainees were placed in
designations during individual patient assessment, teams of eight to 11 people and were told that they
and 3) to determine trainees’ opinions of SALT triage. were responding to a bomb blast at a community
concert. Each incident included between 28 and 30
bomb blast victims. Moulaged manikins represented
METHODS between 10 and 11 of the victims and 18 to 20 moulaged
This prospective observational study was conducted actors were used to represent the remaining victims.
during two Advanced Disaster Life Support (ADLS) Each victim had a card that stated the victim’s respi-
courses. One course was conducted in Augusta, ratory effort, pulse quality, and ability to follow com-
Georgia, in July 2008 and the other in Milwaukee, mands. The actors were also given instructions on how
Wisconsin, in December 2008. The courses had open to act out their symptoms, and several were told to dis-
enrollment and were widely advertised within each rupt the scene by yelling for help, acting intoxicated,
community. The Augusta course was held on a mil- or demanding assistance for their friend. The drills
itary base and had a combination of military and were made as realistic as possible with noise, sirens,
civilian trainees. The Milwaukee course was held at and other real-life distractions such as interruptions
a Veterans Affairs hospital and had primarily civilian by members of the media and having to deal with a
trainees. secondary device on one of the victims (i.e., a bomb
During the mass-triage section of each course, the or gun). The simulated incident in Augusta was con-
trainees were taught to use the SALT triage method. ducted outside and used a group of teenaged boys as
The SALT triage training was provided as a 30-minute the actors. The Milwaukee session was conducted in a
lecture. The training in Augusta was provided by an gymnasium with limited lighting and used a variety of
ADLS-certified instructor who was also a member community volunteers as victims. Each incident used
of the CDC panel that developed SALT. The training the same scenario, a bomb blast at a local concert, and
in Milwaukee was provided by an ADLS-certified the same victims. Because two actors were not avail-
instructor who had had limited previous experience able for all drills, a few of the incidents were not able
with the SALT triage method. Both presentations used to use all of the patient scenarios.
a similar set of slides differing only in the correction An observer who was identified as an instructor
of minor typographical errors and improvement in monitored the drill and recorded the initial and final
formatting. assigned triage categories for each simulated victim.
Lerner et al. SALT TRIAGE IN SIMULATED MASS-CASUALTY INCIDENT 23

The initial assigned triage category was the category TABLE 1. First Assigned Triage Category Compared with the
that was assigned by the first trainee who assessed the Intended Category
patient. The final assigned triage category was the cat- First Assigned Triage Category
egory that was assigned at the end of the drill. Dur-
Intended Category Dead Expectant Immediate Delayed Minimal
ing the drill the victims’ conditions did not change,
but frequently other members of the trainee respon- Dead 13∗ 3 0 0 0
der group would identify what they perceived as er- Expectant 1 12∗ 1 0 0
Immediate 1 3 48∗ 6 0
rors in triage and change the victim’s designation. Delayed 0 0 8 37∗ 8
The initial and final triage categories were then com- Minimal 0 0 0 10 66∗
pared with the intended triage category for each sim- ∗
Indicates the correct assignment.
ulated victim. The percentage of correct assignments
was determined along with 95% confidence intervals
ported that they had heard of SALT triage prior to tak-
(CIs).
ing the course.
Ten of the victims were selected to measure the time
During the eight simulated mass-casualty triage in-
required to individually assess the patient. The actors
cidents, 235 victim observations were studied. Eigh-
operated stopwatches themselves, while the manikins
teen were excluded because the role was incorrectly
were timed by an instructor who was observing the
acted by the actor victim (3), the victim was incorrectly
drill and providing information on the manikin’s con-
moulaged (4), the victim was carrying a secondary de-
dition. Timing began when the trainee approached the
vice (5), the victim was not triaged during the allotted
victim and ended when the trainee verbalized his or
time (5), or the observer did not record the triage cate-
her triage designation or applied a triage tag. The
gory (1).
recorded times were averaged and the standard devia-
Overall, the initial triage was correct for 81% (95%
tions were calculated. Trainees were told that the drill
CI: 75%–86%) of the observations; 8% were overtriaged
was being studied, but they were not told that they
and 11% were undertriaged (Table 1). The final triage
were being timed.
was correct for 83% (95% CI: 78%–88%) of the observa-
After the drill debriefing, trainees were asked to
tions; 6% were overtriaged and 10% were undertriaged
complete an optional retrospective before-and-after
(Table 2). During the course in Augusta, the final triage
survey measuring the degree of self-reported compe-
designation was correct for 86% of the observations;
tence before and after the drill. The survey explored
7% were overtriaged and 7% were undertriaged. Dur-
trainees’ confidence with SALT and ease of use as well
ing the course in Milwaukee, the final triage designa-
as basic demographic information. The survey had
tion was correct for 80% of the observations; 13% were
been previously pilot-tested. Results were analyzed
overtriaged and 7% were undertriaged.
using descriptive statistics.
There were a total of 58 timed victim observations
This study was considered exempt from institutional
because six times were not recorded; three victims
review board review by the institutional review boards
were not triaged, so no time could be determined; and
at the Medical College of Georgia and the Medical Col-
one victim was excluded from the time segment of
lege of Wisconsin.
the study. This victim participated in the Milwaukee
course and was excluded from the analysis because
RESULTS the actor was told by the course observer/controllers
to demand that the providers help her friend first. The
Seventy-three (73) trainees participated in the two
actor was observed as being very insistent and distract-
ADLS courses. There were 43 trainees at the Augusta
ing to the trainees, leading to extremely long triage
course, including 16 physicians, 10 nurses, five prehos-
times that ranged from 133 to 180 seconds. The over-
pital care providers, five physician’s assistants, three
all mean triage interval was 28 seconds (standard de-
pharmacists, and four people from other backgrounds.
viation 22; minimum 4, maximum 94) (Fig. 2). If the
There were 30 trainees at the Milwaukee course, in-
cluding 11 physicians, six nurses, eight prehospital
TABLE 2. Final Assigned Triage Category Compared with
care providers, one nurse/prehospital care provider,
the Intended Category
and four people from other backgrounds.
Overall, 63% of the trainees reported having prior Final Assigned Triage Category
drill experience. The mean number of prior drills for Intended Category Dead Expectant Immediate Delayed Minimal
those with prior experience was 7 (minimum 1 and
Dead 15∗ 2 0 0 0
maximum 60). Twenty-nine percent of the trainees had Expectant 2 11∗ 0 0 0
prior actual mass-casualty incident experience. The Immediate 1 2 40∗ 4 0
mean number of prior mass casualty incidents for Delayed 0 0 9 47∗ 8
those with prior experience was 3 (minimum 1 and Minimal 0 0 0 8 68∗
maximum 15). Twenty-one percent of the trainees re- ∗
Indicates the correct assignment.
24 PREHOSPITAL EMERGENCY CARE JANUARY/MARCH 2010 VOLUME 14 / NUMBER 1

FIGURE 2. Frequency of observation for each range of time to triage.

excluded victim’s data had been included, the overall Simple Treatment and Rapid Transport (START) triage
average would have increased to 34 seconds. During has been studied. These studies have shown that
the Augusta course the mean triage interval was 27 accuracy ranged from 48%5 to 75%.6 Further, 79% of
seconds (standard deviation 23). During the Milwau- the trainees had not heard of SALT triage prior to
kee course the mean triage interval was 30 seconds the training that was provided as part of the studied
(standard deviation 21). courses, and 37% had never before participated in a
Of the 73 trainees who participated in the two disaster drill. This indicates that SALT triage has the
courses, 70 (96%) completed the retrospective survey. potential of being readily learned and correctly used
Prior to the drill, 33% did not feel confident using SALT with minimal training.
triage and 32% were confident or very confident using The mean of 28 seconds to make a triage designa-
SALT. After the drill, none of the respondents did not tion while individually assessing a victim is compa-
feel confident using SALT triage, 26% were at the same rable to, if not better than, other systems. A previous
level of confidence, 74% felt more confident, and none study found that START triage had a mean triage time
felt less confident (Table 3). Before the drill, more than of 30 seconds and the Sacco triage method using the
half, 53%, thought SALT was easier to use than their RPM (respiratory rate, pulse, motor) score had a mean
current disaster triage protocol and 3% thought it was triage time of 45 seconds.7 It is hypothesized that the
more difficult to use than their current disaster triage triage time using SALT triage might be reduced be-
protocol (Table 4). After the drill, no respondents re- cause it does not require an estimation of the victim’s
ported that SALT was more difficult to use than their pulse or respiratory rate. However, additional research
current disaster triage protocol. Further, 77% did not is needed to make this determination.
change how easy they felt SALT triage was to use, 18% Two-thirds of trainees thought SALT triage was eas-
thought it was easier after the drill, and 5% thought ier to use than their current triage protocol. More im-
it was similar rather than easier to use after the drill. portantly, of those who had prior actual mass-casualty
Twenty-three participants had prior actual disaster ex- incident experience, none of them felt that SALT triage
perience, and of those none thought SALT triage was was harder to use than their current triage protocol.
harder to use than their current triage protocol. This indicates that providers will not have difficulty
changing to meet the SALT triage guideline.
This study also determined that after initial didac-
DISCUSSION tic training, a minority of trainees felt very confident
This study found that trainees who were taught to using SALT. After using SALT triage in a drill, all of
use SALT triage had a high rate of accuracy. The 83% the participants had some level of confidence using
accuracy rate that was seen in this study is comparable SALT triage, and almost a third felt very confident us-
to, if not better than, what has been reported when ing it. This indicates that it is likely very important for
trainees to be given experience using a triage system

TABLE 3. Change in Confidence before the Drill Compared

with after the Drill TABLE 4. Change in Ease of Use before the Drill Compared
with after the Drill
Confidence after the Drill
After the Drill
Confidence Prior Very Somewhat Not
to the Drill Confident Confident Confident Confident SALT SALT SALT Is More
Prior to the Drill Is Easier Is Similar Difficult
Very confident 6 0 0 0
Confident 6 10 0 0 SALT is easier 32 3 0
Somewhat confident 7 16 2 0 SALT is similar 10 19 0
Not confident 2 7 14 0 SALT is more difficult 2 0 0
Lerner et al. SALT TRIAGE IN SIMULATED MASS-CASUALTY INCIDENT 25

rather than simply being provided with a didactic lec- for timing including two that needed lifesaving inter-
ture. However, additional research should be done to ventions, in a real incident these times may be longer
determine the optimal required training to be prepared if more lifesaving interventions are applied.
to respond to a mass-casualty incident.
Interestingly, minimal differences in the results were
found between the Augusta and the Milwaukee drills, CONCLUSION
even though the instructors had very different experi-
Trainee victim assessments made using SALT triage
ence levels with SALT. The Augusta instructor was one
during a simulated disaster drill were found to be
of the creators of SALT and the Milwaukee instructor
accurate. Providers with minimal experience and
had only limited experience with SALT. This may in-
training were able to make quick and accurate triage
dicate that SALT is relatively easy to teach, although
decisions. The accuracy rate was higher than those
further research and evaluation are needed. However,
published for the START triage system and of similar
this may also be due to the differences in the circum-
speed. Providers also felt confident using SALT triage
stances of the two drills (e.g., the differences in light-
and found it was similar to or easier to use than their
ing) or some other unknown factor.
current triage protocol. Using SALT triage during a
drill improved confidence in its use. More work is
LIMITATIONS needed, but SALT triage appears to be a promising
triage tool for mass-casualty incidents.
This study was conducted during a simulated exer-
cise. The findings may not be the same as what would
be seen during a real mass-casualty incident. Further, References
the accuracy of the triage categories was based on the
SALT triage category definitions, which have not been 1. Hoey BA, Schwab CW. Level I center triage and mass casualties.
Clin Orthop Relat Res. 2004 May;(422):23–9.
correlated to survival or patient outcome. This study
2. Kennedy K, Aghababian RV, Gans L, Lewis CP. Triage: tech-
was not capable of correlating triage category with niques and applications in decision making. Ann Emerg Med.
patient outcome, so it was not possible to determine 1996;28:136–44.
whether using SALT triage will improve patient out- 3. Lerner EB, Schwartz RB, Coule PL, et al. Mass casualty triage: an
come. This study was also limited to trainees in one evaluation of the data and development of a proposed national
guideline. Disaster Med Public Health Prep. 2008;2(suppl 1):S25–
of two ADLS courses. These trainees may not be gen-
S34.
eralizable to all of the types of responders who might 4. SALT mass casualty triage. Disaster Med Public Health Prep.
be trained to use SALT triage, and we cannot report 2008;2(4):245–246.
on the accuracy of different provider types. The ret- 5. Risavi BL, Salen PN, Heller MB, Arcona S. A two-hour interven-
rospective self-assessment survey is designed to re- tion using START improves prehospital triage of mass casualty
incidents. Prehosp Emerg Care. 2001;5:197–9.
duce “response shift” that can confound the validity 6. Kahn C, Schultz C, Miller K, Anderson C. Does START triage
of traditional self-ratings. This instrument is a well- work? An outcomes-level assessment of use at a mass casualty
established instrument to assess the effectiveness of event [abstract]. Acad Emerg Med. 2007;14(suppl 1):S12a–S13a.
teaching interventions.8 The victims who timed the 7. Sacco WJ, Navin DM, Fiedler KE, Waddell RK 2nd, Long
triage decision making were told to begin the timer WB, Buckman RF Jr. Precise formulation and evidence-based
application of resource-constrained triage. Acad Emerg Med.
when the responder “approached them”; this may 2005;12:759–70.
have introduced some variability in timing since vic- 8. Pirrallo RG, Wolff M, Simpson DE, Hargarten SW. Analysis of an
tims may have interpreted this instruction differently. international EMS train-the-trainer program. Ann Emerg Med.
Further, although we used a variety of patient types 1995;25:656–9.
Copyright of Prehospital Emergency Care is the property of Taylor & Francis Ltd and its content may not be
copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written
permission. However, users may print, download, or email articles for individual use.

View publication stats