Prehospital Emergency Care
ISSN: 1090-3127 (Print) 1545-0066 (Online) Journal homepage: http://www.tandfonline.com/loi/ipec20
Randomized Trial Comparing Two Mass Casualty Triage Systems (JumpSTART versus SALT) in a Pediatric Simulated Mass Casualty Event Nicole Jones MD, Marjorie Lee White MD, MPPM, MEd, Nancy Tofil MD, MEd, MeKeisha Pickens MS4, Amber Youngblood RN, Lynn Zinkan RN, MPH & Mark D. Baker MD, MPH To cite this article: Nicole Jones MD, Marjorie Lee White MD, MPPM, MEd, Nancy Tofil MD, MEd, MeKeisha Pickens MS4, Amber Youngblood RN, Lynn Zinkan RN, MPH & Mark D. Baker MD, MPH (2014) Randomized Trial Comparing Two Mass Casualty Triage Systems (JumpSTART versus SALT) in a Pediatric Simulated Mass Casualty Event, Prehospital Emergency Care, 18:3, 417-423 To link to this article: http://dx.doi.org/10.3109/10903127.2014.882997
Published online: 06 Mar 2014.
Submit your article to this journal
Article views: 229
View related articles
View Crossmark data
Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=ipec20 Download by: [University of Liverpool]
Date: 02 October 2015, At: 10:38
RANDOMIZED TRIAL COMPARING TWO MASS CASUALTY TRIAGE SYSTEMS (JUMPSTART VERSUS SALT) IN A PEDIATRIC SIMULATED MASS CASUALTY EVENT Nicole Jones, MD, Marjorie Lee White, MD, MPPM, MEd, Nancy Tofil, MD, MEd, MeKeisha Pickens, MS4, Amber Youngblood, RN, Lynn Zinkan, RN, MPH, Mark D. Baker, MD, MPH accuracy of 66% ±15%, an overtriage mean rate of 22 ± 16%, and an undertriage rate of 10 ± 9%. Twenty-six participants were assigned to the JumpSTART group with an overall accuracy of 66 ± 12%, an overtriage mean of 23 ±16%, and an undertriage rate of 11.2 ± 11%. Ease of use was not statistically different between the two systems (median Likert value of both systems = 2, p = 0.39) Time to triage per patient was statistically faster in the JumpSTART group (SALT = 34 ± 23 seconds, JumpSTART = 26 ± 19 seconds, p = 0.02). Both systems were prone to cognitive and affective error. Conclusion. SALT appears to be at least as good as JumpSTART in overall triage accuracy, overtriage, or undertriage rates in a simulated pediatric MCI. Both systems were considered easy to use. However, JumpSTART was 8 seconds faster per patient in time taken to assign triage designations. Key words: mass casualty triage; pediatric; simulation
Downloaded by [University of Liverpool] at 10:38 02 October 2015
ABSTRACT Purpose. Several field triage systems have been developed to rapidly sort patients following a mass casualty incident (MCI). JumpSTART (Simple Triage and Rapid Transport) is a pediatric-specific MCI triage system. SALT (Sort, Assess, Lifesaving interventions, Treat/Transport) has been proposed as a new national standard for MCI triage for both adult and pediatric patients, but it has not been tested in a pediatric population. This pilot study hypothesizes that SALT is at least as good as JumpSTART in triage accuracy, speed, and ease of use in a simulated pediatric MCI. Methods. Paramedics were invited and randomly assigned to either SALT or JumpSTART study groups. Following randomization, subjects viewed a 15-minute PowerPoint lecture on either JumpSTART or SALT. Subjects were provided with a triage algorithm card for reference and were asked to assign triage categories to 10 pediatric patients in a simulated building collapse. The scenario consisted of 4 children in moulage and 6 high-fidelity pediatric simulators. Injuries and triage categories were based on a previously published MCI scenario. One investigator followed each subject to record time and triage assignment. All subjects completed a post-test survey and structured interview following the simulated disaster. Results. Forty-three paramedics were enrolled. Seventeen were assigned to the SALT group with an overall triage
PREHOSPITAL EMERGENCY CARE 2014;18:417–423
INTRODUCTION A mass casualty incident (MCI) occurs when the need for care outstrips the current available resources. In this setting the focus changes to the greatest good for the greatest number of people as opposed to the greatest good for the individual. The use of an objective triage tool helps to provide a systematic approach to a very stressful and emotionally challenging triage process. Multiple prehospital triage systems exist, but there is a paucity of evidence to support the use of one system over another. Recently, the Centers for Disease Control and Prevention (CDC) formed a committee comprising emergency and disaster medicine experts to establish a consensus MCI system based on expert opinion and best available science.1 The proposed system, SALT (Sort, Assess, Lifesaving interventions, Treat or Transport), incorporates aspects from all of the existing triage systems to create a single overarching guide for unifying the mass casualty triage process across the United States.1 This system is designed for use in both adult and pediatric patients (Figure 1). JumpSTART (Simple Triage and Rapid Treatment) is the most widely used pediatric-specific MCI triage tool in the United States.2 JumpSTART is based on the START triage tool, which is currently the gold standard for field adult MCI triage in the United States.2 JumpSTART takes into account the unique aspects of pediatric physiology that theoretically should result in
Received June 24, 2013 from the University of Alabama – Birmingham, Department of Pediatrics, Division of Emergency Medicine, Birmingham, Alabama (NJ, MLW, MP, MDB), University of Alabama – Birmingham, Department of Pediatrics, Division of Critical Care Medicine, Birmingham, Alabama (NT), and University of Alabama – Birmingham, Department of Pediatrics, Children’s of Alabama, Pediatric Simulation Center, Birmingham, Alabama (MLW, NT, AY, LZ). Revision received November 26, 2013; accepted for publication December 2, 2013. The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper. The authors thank the University of Alabama at Birmingham Medical students, Pediatric Emergency Medicine fellows, and children who volunteered as actors for their assistance with this study. We would also like to thank Andrew Rucks, PhD, for his contributions and Chris Pruitt, MD, for reviewing the manuscript. Address correspondence to Nicole Jones, University of Alabama – Department of Pediatrics, Devision of Emergency Medicine, 1600 5th Avenue South, CPP1 Suite 210, Birmingham, AL 35233, USA. e-mail: [email protected] doi: 10.3109/10903127.2014.882997
417
PREHOSPITAL EMERGENCY CARE
JULY/SEPTEMBER 2014
VOLUME 18 / NUMBER 3
Downloaded by [University of Liverpool] at 10:38 02 October 2015
418
FIGURE 1. SALT algorithm.
more accurate triage assignments in children. In particular, JumpSTART addresses the fact that most pediatric arrests are respiratory in nature. Therefore, it mandates that 5 rescue breaths should be administered to an apneic child with a pulse. This is opposed to the SALT algorithm that provides an option to provide two rescue breaths to an apneic child. JumpSTART also accounts for the range of respiratory rates that are normal in the pediatric population (Figure 2). This pilot study was designed to compare SALT and JumpSTART in a simulated pediatric MCI. We hypothesize that SALT is not inferior to JumpSTART in regards to accuracy, speed, and ease of use.
METHODS Study Design and Setting The study design was formulated by the authors and was based on the design of the pilot test of the SALT MCI triage system.3 Triage categories for both JumpSTART and SALT were determined by a modified Delphi approach involving three of the study investigators (NJ, MB, MLW). Each investigator applied the respective algorithms to the individual patient scenarios and then met as a group to establish consensus. Permission was obtained from the
JUMPSTART VS. SALT IN A PEDIATRIC SIMULATED EVENT
Downloaded by [University of Liverpool] at 10:38 02 October 2015
N. Jones et al.
FIGURE 2. JumpSTART algorithm.
419
420 Birmingham Fire and Rescue emergency medical services (EMS) to allow recruitment of their emergency medical providers for participation in the study. The study took place at the Pediatric Simulation Center at Children’s of Alabama and was approved by the University of Alabama at Birmingham’s Institutional Review Board for Human Subjects.
PREHOSPITAL EMERGENCY CARE
JULY/SEPTEMBER 2014
VOLUME 18 / NUMBER 3
TABLE 1. Prior MCI knowledge between study groups
Knowledge of mass casualty triage,% (n) Prior participation in an MCI event,% (n) Familiar with JumpSTART,% (n) Familiar with SALT,% (n)
SALT% (n = 17)
JumpSTART% (n = 26)
100 (17) 29 (5) 12 (2) 5.9 (1)
92 (24) 19 (5) 3.8 (1) 0 (0)
Downloaded by [University of Liverpool] at 10:38 02 October 2015
Experimental Protocol The scenario was adapted with permission from a previously published pediatric MCI drill with similar injury patterns and assigned JumpSTART triage levels.4 The study scenario consisted of 10 patients (4 moulaged children, 5 high-fidelity simulation manikins, and 1 low-fidelity simulation manikin) in a simulated apartment building collapse. Efforts were made to replicate a true apartment collapse. Patient characteristics were identical for the two study groups. The children and manikins were distributed throughout 4 rooms and a hallway (Figure 3). Overturned furniture were placed throughout the scene, rooms were all darkened with tarp placed over hallway ceiling lights, and a flashing light and siren were placed in the hall. Triage assignment distribution consisted of 3 green (minor), 3 yellow (delayed), 3 red (immediate), and 1 black (expectant/dead) patient for both SALT and JumpSTART. The children were coached and dressed in appropriate moulage and the simulation manikins were programmed accordingly. The simulation manikins used were SimMan, SimBaby, and SimNewB (Laerdal), HAL 2yo (Guamard), MetiChild (METI), and a nonmechanical manikin that represented a deceased child. Informed consent was obtained from all participants. Participants were randomly assigned to either a SALT or JumpSTART study group based on a random numbers table. All participants viewed a 15-minute justin-time PowerPoint lecture with voice-over on either SALT or JumpSTART. Both lectures ended with clinical scenarios that allowed study participants to practice applying the triage algorithms. Each participant was provided with a standardized triage algorithm card for reference during the study. Subjects then participated individually in the simulated MCI. They were informed that there had been a building collapse and that they were to assess and assign triage designations to all victims. They were provided with 10 triage tags and told that the drill would be complete when they had no triage tags remaining.
Measures Before watching the educational video, subjects completed a pretest survey assessing prior MCI triage knowledge. After the drill, participants underwent a structured and scripted debriefing by one of the in-
vestigators (MLW). They also completed a post-test survey that included questions about their views on the simulation experience, confidence with pediatric MCI triage, and ease of use. Survey questions were rated on a Likert scale from 1, very easy, to 5, very difficult. The debriefing inquired about participants’ reasons for assigning an incorrect triage designation. The reasons for incorrect triage designations were categorized into 3 groups: affective errors, cognitive errors, and technical errors. A priori, two of the investigators (NJ, MB) defined affective errors as errors due to emotional bias. Cognitive errors were defined as incorrect application of the triage algorithm. Technical errors referred to malfunction or inherent limitations of the manikins. Triage times were recorded by a single member of the research team (MP) by digital stopwatch. Triage accuracy was assessed by comparing the assigned category with that assigned by the authors. Those assessing triage accuracy (NJ, MB) were blinded to group assignment. Triage accuracy was analyzed with Fisher’s exact test. Time was analyzed with an unpaired t-test. Ease of use data were compared with the Mann-Whitney Utest. All tests were two-tailed, with an alpha level of 0.05. Quantitative statistics were calculated using SPSS 20.0.0 (IBM, Chicago).
RESULTS Forty-four EMS providers were approached to participate in the study. One subject declined and 43 EMS providers enrolled in the study. Data regarding prior MCI experience and familiarity with triage systems are displayed in Table 1. Overall triage accuracy, overtriage, and undertriage rates were not statistically different between the SALT and JumpSTART groups (Table 2). Study subjects took less time to assign triage categories in the JumpSTART TABLE 2. Accuracy of MCI triage category assignment using SALT or JumpSTART SALT
Accuracy,% (SD) Overtriage,% (SD) Undertriage,% (SD)
66 (15) 23 (16) 10 (9)
JumpSTART
P value
66 (13) 23 (16) 11 (11)
1 0.91 0.87
N. Jones et al.
421
JUMPSTART VS. SALT IN A PEDIATRIC SIMULATED EVENT
Room
Room Two
Thr
riefing Room
Downloaded by [University of Liverpool] at 10:38 02 October 2015
3
Hallway
Elevator
Elevator
Room
FIGURE 3. Schematic of simulated apartment collapse.
Unused Room
Room
Unused Room
422
PREHOSPITAL EMERGENCY CARE
TABLE 3. Mean time to assign triage designations using SALT or JumpSTART SALT
Time to triage, sec (SD)
34 (13)
JumpSTART
P value
26 (8)
0.02
group (Table 3). Both systems were comparably easy to use, with the median of subjects rating both systems as “easy to use” (Likert value = 2, p = 0.39). Subjects’ explanations for assigning wrong triage levels are categorized in Table 4.
Downloaded by [University of Liverpool] at 10:38 02 October 2015
DISCUSSION There was no difference in triage accuracy between the SALT and JumpSTART groups, supporting our hypothesis that SALT is at least as good as JumpSTART as an MCI tool. In regards to overall accuracy our results are within the range of results obtained in prior MCI triage studies. A just-in-time study involving first-year medical students trained in START triage showed an accuracy rate of 64–65%, an undertriage rate of 13%, and an overtriage rate of 18%.5 JumpSTART training in school nursing personnel showed a higher posttraining accuracy rate of 80% with an under- and overtriage rate of 10%.4 Recent studies evaluating SALT found accuracy rates of 83 and 80%, respectively.6,7 In one of these studies participants assigned triage designations in groups that were allowed to confer, possibly contributing to the high rates of accuracy.6 Another study was computer based and untimed, which may also contribute to higher accuracy rates.7 There are no standards as to what are acceptable accuracy, overtriage, and undertriage rates but general consensus is that both overtriage and undertriage in a mass casualty incident can lead to negative outcomes. Overtriage can lead to unnecessary use of limited resources and distraction of medical attention from persons needing it.3 Reducing overtriage has been shown to lower health-care costs on a case-by-case basis.8 Undertriage is also concerning as it can result in delayed care for persons with critical injuries. Cognitive errors were common in both SALT and JumpSTART groups. Two participants in each of the groups stated that they used their intuition to make triage decisions rather than referring to the triage algoTABLE 4. Error types for inaccurate triage assignments
Affective error, % (n) Cognitive error, % (n) Technical error, % (n)
SALT (14)
JumpSTART (23)
29 (4) 64 (9) 43 (6)
41 (9) 50 (11) 27 (6)
More than one reason was given for some inaccurate triage assignments. Three participants in each group did not provide reasons for their inaccurate triage designations.
JULY/SEPTEMBER 2014
VOLUME 18 / NUMBER 3
rithm cards. One participant stated, “In the real world you use your gut.” Another contributing factor to the errors of cognition may be due to inadequate triage training. It is possible that if more time was given for training or a more detailed talk was given prior to participation in the simulation that overall accuracy may increase. However, in real MCI events, it may be impossible to achieve extensive training prior to the event. Just-in-time training, similar to what was done in our study, is more likely to occur. Technical errors ranked second in the SALT group and last in the JumpSTART group. Some of these errors were unavoidable due to mechanical aspects of the manikins. An example of this is the blue light around the SimBaby’s mouth that was intended to represent cyanosis. Two study participants stated that they did not realize that the light represented cyanosis and this likely affected their assigned triage designation. Affective errors ranked second in the JumpSTART group and last in the SALT group. MCI triage in general and pediatric MCI specifically lends itself to emotional bias as most health-care providers have a goal of doing the most good for the individual patient. In MCI triage the goal changes to the greatest good for the greatest number.6 Many of the study participants stated that they would upgrade a child’s triage designation simply because he/she was a child. Statements made by study participants include “futility is hard in kids” and “I hate to just mark off children.” These types of errors bring to light the issue of theoretical benefits of a triage system versus what is actually done in practice. Further research is needed to investigate the application and effectiveness of MCI triage algorithms in disasters.
STUDY LIMITATIONS This study has several limitations. Overall study subject numbers were small. It is possible that larger study groups would have demonstrated differences in triage accuracy or ease of use. Another weakness was the randomization process. A random numbers table was used but, likely due to the small number of study participants, there was a considerable difference between participants in the SALT (17) and JumpSTART groups (26). While the overall number of study subjects was small, lending itself most simply to simple randomization, this inequality renders our study less powerful to detect a difference between the two groups, which is a considerable limitation in this noninferiority design. Further, while both groups had similar prior knowledge of mass casualty triage, considerably more subjects in the SALT group had prior experience in MCI events. This could create an allocation bias, impacting the primary endpoint of triage accuracy for participants in the SALT group.
N. Jones et al.
In addition, some participants misinterpreted features of the simulators. This limit could likely be decreased if participants would have had a prestudy orientation to the simulator features. Finally, an optimal, “real-world” approach to compare these triage systems might be a knowledge retention study, where subjects learn one system and are then studied at a later time. While this is a promising future direction for further investigation, this was beyond the scope of this initial study.
CONCLUSION
Downloaded by [University of Liverpool] at 10:38 02 October 2015
423
JUMPSTART VS. SALT IN A PEDIATRIC SIMULATED EVENT
Mass casualty triage algorithms are used worldwide without substantial evidence to support the use of any one system over another. JumpSTART is the most commonly used pediatric MCI triage tool and is designed to account for the unique physiology of children. Theoretically, JumpSTART should allow for more appropriate triage of children in comparison to other MCI triage systems. SALT has recently been proposed as the national MCI triage system to be used for both adult and pediatric triage. However, it has not previously been evaluated in a pediatric population. Our preliminary research suggests that while it may take responders longer to assign triage categories when using SALT versus JumpSTART, these two MCI triage systems do not appear to differ in regards to accuracy or ease of use in a pediatric simulated mass casualty event. Given the limitations of our study fu-
ture research is needed to further investigate these findings.
References 1.
Lerner E, Schwartz, R, Coule, P, Winstein E, Cone DC, Hunt RC, Sasser SM, Liu JM, Nudell N, Wedmore I, Hammond J, Bulger E, Salomone J, Sanddal T, Lord G, Markenson D, O’Conner R. Mass casualty triage: an evaluation of the data and development of a proposed national guideline. Disaster Med Public Health Prep. 2008;2:S25–S33. 2. Romig L. Pediatric triage: a system to JumpSTART your triage of young patients at MCIs. JEMS. 2002;27:52–63. 3. Cone D, Serra J, Burns K, MacMillan D, Kurland L, Van Gelder C. Pilot test of the SALT mass casualty triage system. Prehosp Emerg Care. 2009;13:536–40. 4. Sanddal T, Loyacono T, Sanddal N. Effect of JumpSTART training on immediate and short-term pediatric triage performance. Pediatr Emerg Care. 2004;20(11):749–53. 5. Sapp R, Brice J, Myers J, Hinchey P. Triage performance of first year medical students using a multiple casualty scenario, paper exercise. Prehosp Disaster Med. 2010;25(3): 239–45. 6. Lerner E, Schwartz R, Coule P, Pirrallo G. Use of SALT triage in a simulated mass-casualty incident. Prehosp Emerg Care. 2010;14(1):21–5. 7. Deluhery M, Lerner B, Pirrallo R, Schwartz R. Paramedic accuracy usiing SALT triage after a brief initial training. Prehosp Emerg Care. 2011;15:526–32. 8. Faul M, Wald M, Sullivent E, Sasser S, Kapil V, Lerner B, et al. Large cost savings realized from the 2006 field triage guideline: reduction in overtriage in U.S. trauma centers. Prehosp Emerg Care. 2012;16:222–9. 9. Burstein J. Mostly dead: can science help with disaster triage? Ann Emerg Med. 2009;54: 431.
ISSN: 1090-3127 (Print) 1545-0066 (Online) Journal homepage: http://www.tandfonline.com/loi/ipec20
Randomized Trial Comparing Two Mass Casualty Triage Systems (JumpSTART versus SALT) in a Pediatric Simulated Mass Casualty Event Nicole Jones MD, Marjorie Lee White MD, MPPM, MEd, Nancy Tofil MD, MEd, MeKeisha Pickens MS4, Amber Youngblood RN, Lynn Zinkan RN, MPH & Mark D. Baker MD, MPH To cite this article: Nicole Jones MD, Marjorie Lee White MD, MPPM, MEd, Nancy Tofil MD, MEd, MeKeisha Pickens MS4, Amber Youngblood RN, Lynn Zinkan RN, MPH & Mark D. Baker MD, MPH (2014) Randomized Trial Comparing Two Mass Casualty Triage Systems (JumpSTART versus SALT) in a Pediatric Simulated Mass Casualty Event, Prehospital Emergency Care, 18:3, 417-423 To link to this article: http://dx.doi.org/10.3109/10903127.2014.882997
Published online: 06 Mar 2014.
Submit your article to this journal
Article views: 229
View related articles
View Crossmark data
Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=ipec20 Download by: [University of Liverpool]
Date: 02 October 2015, At: 10:38
RANDOMIZED TRIAL COMPARING TWO MASS CASUALTY TRIAGE SYSTEMS (JUMPSTART VERSUS SALT) IN A PEDIATRIC SIMULATED MASS CASUALTY EVENT Nicole Jones, MD, Marjorie Lee White, MD, MPPM, MEd, Nancy Tofil, MD, MEd, MeKeisha Pickens, MS4, Amber Youngblood, RN, Lynn Zinkan, RN, MPH, Mark D. Baker, MD, MPH accuracy of 66% ±15%, an overtriage mean rate of 22 ± 16%, and an undertriage rate of 10 ± 9%. Twenty-six participants were assigned to the JumpSTART group with an overall accuracy of 66 ± 12%, an overtriage mean of 23 ±16%, and an undertriage rate of 11.2 ± 11%. Ease of use was not statistically different between the two systems (median Likert value of both systems = 2, p = 0.39) Time to triage per patient was statistically faster in the JumpSTART group (SALT = 34 ± 23 seconds, JumpSTART = 26 ± 19 seconds, p = 0.02). Both systems were prone to cognitive and affective error. Conclusion. SALT appears to be at least as good as JumpSTART in overall triage accuracy, overtriage, or undertriage rates in a simulated pediatric MCI. Both systems were considered easy to use. However, JumpSTART was 8 seconds faster per patient in time taken to assign triage designations. Key words: mass casualty triage; pediatric; simulation
Downloaded by [University of Liverpool] at 10:38 02 October 2015
ABSTRACT Purpose. Several field triage systems have been developed to rapidly sort patients following a mass casualty incident (MCI). JumpSTART (Simple Triage and Rapid Transport) is a pediatric-specific MCI triage system. SALT (Sort, Assess, Lifesaving interventions, Treat/Transport) has been proposed as a new national standard for MCI triage for both adult and pediatric patients, but it has not been tested in a pediatric population. This pilot study hypothesizes that SALT is at least as good as JumpSTART in triage accuracy, speed, and ease of use in a simulated pediatric MCI. Methods. Paramedics were invited and randomly assigned to either SALT or JumpSTART study groups. Following randomization, subjects viewed a 15-minute PowerPoint lecture on either JumpSTART or SALT. Subjects were provided with a triage algorithm card for reference and were asked to assign triage categories to 10 pediatric patients in a simulated building collapse. The scenario consisted of 4 children in moulage and 6 high-fidelity pediatric simulators. Injuries and triage categories were based on a previously published MCI scenario. One investigator followed each subject to record time and triage assignment. All subjects completed a post-test survey and structured interview following the simulated disaster. Results. Forty-three paramedics were enrolled. Seventeen were assigned to the SALT group with an overall triage
PREHOSPITAL EMERGENCY CARE 2014;18:417–423
INTRODUCTION A mass casualty incident (MCI) occurs when the need for care outstrips the current available resources. In this setting the focus changes to the greatest good for the greatest number of people as opposed to the greatest good for the individual. The use of an objective triage tool helps to provide a systematic approach to a very stressful and emotionally challenging triage process. Multiple prehospital triage systems exist, but there is a paucity of evidence to support the use of one system over another. Recently, the Centers for Disease Control and Prevention (CDC) formed a committee comprising emergency and disaster medicine experts to establish a consensus MCI system based on expert opinion and best available science.1 The proposed system, SALT (Sort, Assess, Lifesaving interventions, Treat or Transport), incorporates aspects from all of the existing triage systems to create a single overarching guide for unifying the mass casualty triage process across the United States.1 This system is designed for use in both adult and pediatric patients (Figure 1). JumpSTART (Simple Triage and Rapid Treatment) is the most widely used pediatric-specific MCI triage tool in the United States.2 JumpSTART is based on the START triage tool, which is currently the gold standard for field adult MCI triage in the United States.2 JumpSTART takes into account the unique aspects of pediatric physiology that theoretically should result in
Received June 24, 2013 from the University of Alabama – Birmingham, Department of Pediatrics, Division of Emergency Medicine, Birmingham, Alabama (NJ, MLW, MP, MDB), University of Alabama – Birmingham, Department of Pediatrics, Division of Critical Care Medicine, Birmingham, Alabama (NT), and University of Alabama – Birmingham, Department of Pediatrics, Children’s of Alabama, Pediatric Simulation Center, Birmingham, Alabama (MLW, NT, AY, LZ). Revision received November 26, 2013; accepted for publication December 2, 2013. The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper. The authors thank the University of Alabama at Birmingham Medical students, Pediatric Emergency Medicine fellows, and children who volunteered as actors for their assistance with this study. We would also like to thank Andrew Rucks, PhD, for his contributions and Chris Pruitt, MD, for reviewing the manuscript. Address correspondence to Nicole Jones, University of Alabama – Department of Pediatrics, Devision of Emergency Medicine, 1600 5th Avenue South, CPP1 Suite 210, Birmingham, AL 35233, USA. e-mail: [email protected] doi: 10.3109/10903127.2014.882997
417
PREHOSPITAL EMERGENCY CARE
JULY/SEPTEMBER 2014
VOLUME 18 / NUMBER 3
Downloaded by [University of Liverpool] at 10:38 02 October 2015
418
FIGURE 1. SALT algorithm.
more accurate triage assignments in children. In particular, JumpSTART addresses the fact that most pediatric arrests are respiratory in nature. Therefore, it mandates that 5 rescue breaths should be administered to an apneic child with a pulse. This is opposed to the SALT algorithm that provides an option to provide two rescue breaths to an apneic child. JumpSTART also accounts for the range of respiratory rates that are normal in the pediatric population (Figure 2). This pilot study was designed to compare SALT and JumpSTART in a simulated pediatric MCI. We hypothesize that SALT is not inferior to JumpSTART in regards to accuracy, speed, and ease of use.
METHODS Study Design and Setting The study design was formulated by the authors and was based on the design of the pilot test of the SALT MCI triage system.3 Triage categories for both JumpSTART and SALT were determined by a modified Delphi approach involving three of the study investigators (NJ, MB, MLW). Each investigator applied the respective algorithms to the individual patient scenarios and then met as a group to establish consensus. Permission was obtained from the
JUMPSTART VS. SALT IN A PEDIATRIC SIMULATED EVENT
Downloaded by [University of Liverpool] at 10:38 02 October 2015
N. Jones et al.
FIGURE 2. JumpSTART algorithm.
419
420 Birmingham Fire and Rescue emergency medical services (EMS) to allow recruitment of their emergency medical providers for participation in the study. The study took place at the Pediatric Simulation Center at Children’s of Alabama and was approved by the University of Alabama at Birmingham’s Institutional Review Board for Human Subjects.
PREHOSPITAL EMERGENCY CARE
JULY/SEPTEMBER 2014
VOLUME 18 / NUMBER 3
TABLE 1. Prior MCI knowledge between study groups
Knowledge of mass casualty triage,% (n) Prior participation in an MCI event,% (n) Familiar with JumpSTART,% (n) Familiar with SALT,% (n)
SALT% (n = 17)
JumpSTART% (n = 26)
100 (17) 29 (5) 12 (2) 5.9 (1)
92 (24) 19 (5) 3.8 (1) 0 (0)
Downloaded by [University of Liverpool] at 10:38 02 October 2015
Experimental Protocol The scenario was adapted with permission from a previously published pediatric MCI drill with similar injury patterns and assigned JumpSTART triage levels.4 The study scenario consisted of 10 patients (4 moulaged children, 5 high-fidelity simulation manikins, and 1 low-fidelity simulation manikin) in a simulated apartment building collapse. Efforts were made to replicate a true apartment collapse. Patient characteristics were identical for the two study groups. The children and manikins were distributed throughout 4 rooms and a hallway (Figure 3). Overturned furniture were placed throughout the scene, rooms were all darkened with tarp placed over hallway ceiling lights, and a flashing light and siren were placed in the hall. Triage assignment distribution consisted of 3 green (minor), 3 yellow (delayed), 3 red (immediate), and 1 black (expectant/dead) patient for both SALT and JumpSTART. The children were coached and dressed in appropriate moulage and the simulation manikins were programmed accordingly. The simulation manikins used were SimMan, SimBaby, and SimNewB (Laerdal), HAL 2yo (Guamard), MetiChild (METI), and a nonmechanical manikin that represented a deceased child. Informed consent was obtained from all participants. Participants were randomly assigned to either a SALT or JumpSTART study group based on a random numbers table. All participants viewed a 15-minute justin-time PowerPoint lecture with voice-over on either SALT or JumpSTART. Both lectures ended with clinical scenarios that allowed study participants to practice applying the triage algorithms. Each participant was provided with a standardized triage algorithm card for reference during the study. Subjects then participated individually in the simulated MCI. They were informed that there had been a building collapse and that they were to assess and assign triage designations to all victims. They were provided with 10 triage tags and told that the drill would be complete when they had no triage tags remaining.
Measures Before watching the educational video, subjects completed a pretest survey assessing prior MCI triage knowledge. After the drill, participants underwent a structured and scripted debriefing by one of the in-
vestigators (MLW). They also completed a post-test survey that included questions about their views on the simulation experience, confidence with pediatric MCI triage, and ease of use. Survey questions were rated on a Likert scale from 1, very easy, to 5, very difficult. The debriefing inquired about participants’ reasons for assigning an incorrect triage designation. The reasons for incorrect triage designations were categorized into 3 groups: affective errors, cognitive errors, and technical errors. A priori, two of the investigators (NJ, MB) defined affective errors as errors due to emotional bias. Cognitive errors were defined as incorrect application of the triage algorithm. Technical errors referred to malfunction or inherent limitations of the manikins. Triage times were recorded by a single member of the research team (MP) by digital stopwatch. Triage accuracy was assessed by comparing the assigned category with that assigned by the authors. Those assessing triage accuracy (NJ, MB) were blinded to group assignment. Triage accuracy was analyzed with Fisher’s exact test. Time was analyzed with an unpaired t-test. Ease of use data were compared with the Mann-Whitney Utest. All tests were two-tailed, with an alpha level of 0.05. Quantitative statistics were calculated using SPSS 20.0.0 (IBM, Chicago).
RESULTS Forty-four EMS providers were approached to participate in the study. One subject declined and 43 EMS providers enrolled in the study. Data regarding prior MCI experience and familiarity with triage systems are displayed in Table 1. Overall triage accuracy, overtriage, and undertriage rates were not statistically different between the SALT and JumpSTART groups (Table 2). Study subjects took less time to assign triage categories in the JumpSTART TABLE 2. Accuracy of MCI triage category assignment using SALT or JumpSTART SALT
Accuracy,% (SD) Overtriage,% (SD) Undertriage,% (SD)
66 (15) 23 (16) 10 (9)
JumpSTART
P value
66 (13) 23 (16) 11 (11)
1 0.91 0.87
N. Jones et al.
421
JUMPSTART VS. SALT IN A PEDIATRIC SIMULATED EVENT
Room
Room Two
Thr
riefing Room
Downloaded by [University of Liverpool] at 10:38 02 October 2015
3
Hallway
Elevator
Elevator
Room
FIGURE 3. Schematic of simulated apartment collapse.
Unused Room
Room
Unused Room
422
PREHOSPITAL EMERGENCY CARE
TABLE 3. Mean time to assign triage designations using SALT or JumpSTART SALT
Time to triage, sec (SD)
34 (13)
JumpSTART
P value
26 (8)
0.02
group (Table 3). Both systems were comparably easy to use, with the median of subjects rating both systems as “easy to use” (Likert value = 2, p = 0.39). Subjects’ explanations for assigning wrong triage levels are categorized in Table 4.
Downloaded by [University of Liverpool] at 10:38 02 October 2015
DISCUSSION There was no difference in triage accuracy between the SALT and JumpSTART groups, supporting our hypothesis that SALT is at least as good as JumpSTART as an MCI tool. In regards to overall accuracy our results are within the range of results obtained in prior MCI triage studies. A just-in-time study involving first-year medical students trained in START triage showed an accuracy rate of 64–65%, an undertriage rate of 13%, and an overtriage rate of 18%.5 JumpSTART training in school nursing personnel showed a higher posttraining accuracy rate of 80% with an under- and overtriage rate of 10%.4 Recent studies evaluating SALT found accuracy rates of 83 and 80%, respectively.6,7 In one of these studies participants assigned triage designations in groups that were allowed to confer, possibly contributing to the high rates of accuracy.6 Another study was computer based and untimed, which may also contribute to higher accuracy rates.7 There are no standards as to what are acceptable accuracy, overtriage, and undertriage rates but general consensus is that both overtriage and undertriage in a mass casualty incident can lead to negative outcomes. Overtriage can lead to unnecessary use of limited resources and distraction of medical attention from persons needing it.3 Reducing overtriage has been shown to lower health-care costs on a case-by-case basis.8 Undertriage is also concerning as it can result in delayed care for persons with critical injuries. Cognitive errors were common in both SALT and JumpSTART groups. Two participants in each of the groups stated that they used their intuition to make triage decisions rather than referring to the triage algoTABLE 4. Error types for inaccurate triage assignments
Affective error, % (n) Cognitive error, % (n) Technical error, % (n)
SALT (14)
JumpSTART (23)
29 (4) 64 (9) 43 (6)
41 (9) 50 (11) 27 (6)
More than one reason was given for some inaccurate triage assignments. Three participants in each group did not provide reasons for their inaccurate triage designations.
JULY/SEPTEMBER 2014
VOLUME 18 / NUMBER 3
rithm cards. One participant stated, “In the real world you use your gut.” Another contributing factor to the errors of cognition may be due to inadequate triage training. It is possible that if more time was given for training or a more detailed talk was given prior to participation in the simulation that overall accuracy may increase. However, in real MCI events, it may be impossible to achieve extensive training prior to the event. Just-in-time training, similar to what was done in our study, is more likely to occur. Technical errors ranked second in the SALT group and last in the JumpSTART group. Some of these errors were unavoidable due to mechanical aspects of the manikins. An example of this is the blue light around the SimBaby’s mouth that was intended to represent cyanosis. Two study participants stated that they did not realize that the light represented cyanosis and this likely affected their assigned triage designation. Affective errors ranked second in the JumpSTART group and last in the SALT group. MCI triage in general and pediatric MCI specifically lends itself to emotional bias as most health-care providers have a goal of doing the most good for the individual patient. In MCI triage the goal changes to the greatest good for the greatest number.6 Many of the study participants stated that they would upgrade a child’s triage designation simply because he/she was a child. Statements made by study participants include “futility is hard in kids” and “I hate to just mark off children.” These types of errors bring to light the issue of theoretical benefits of a triage system versus what is actually done in practice. Further research is needed to investigate the application and effectiveness of MCI triage algorithms in disasters.
STUDY LIMITATIONS This study has several limitations. Overall study subject numbers were small. It is possible that larger study groups would have demonstrated differences in triage accuracy or ease of use. Another weakness was the randomization process. A random numbers table was used but, likely due to the small number of study participants, there was a considerable difference between participants in the SALT (17) and JumpSTART groups (26). While the overall number of study subjects was small, lending itself most simply to simple randomization, this inequality renders our study less powerful to detect a difference between the two groups, which is a considerable limitation in this noninferiority design. Further, while both groups had similar prior knowledge of mass casualty triage, considerably more subjects in the SALT group had prior experience in MCI events. This could create an allocation bias, impacting the primary endpoint of triage accuracy for participants in the SALT group.
N. Jones et al.
In addition, some participants misinterpreted features of the simulators. This limit could likely be decreased if participants would have had a prestudy orientation to the simulator features. Finally, an optimal, “real-world” approach to compare these triage systems might be a knowledge retention study, where subjects learn one system and are then studied at a later time. While this is a promising future direction for further investigation, this was beyond the scope of this initial study.
CONCLUSION
Downloaded by [University of Liverpool] at 10:38 02 October 2015
423
JUMPSTART VS. SALT IN A PEDIATRIC SIMULATED EVENT
Mass casualty triage algorithms are used worldwide without substantial evidence to support the use of any one system over another. JumpSTART is the most commonly used pediatric MCI triage tool and is designed to account for the unique physiology of children. Theoretically, JumpSTART should allow for more appropriate triage of children in comparison to other MCI triage systems. SALT has recently been proposed as the national MCI triage system to be used for both adult and pediatric triage. However, it has not previously been evaluated in a pediatric population. Our preliminary research suggests that while it may take responders longer to assign triage categories when using SALT versus JumpSTART, these two MCI triage systems do not appear to differ in regards to accuracy or ease of use in a pediatric simulated mass casualty event. Given the limitations of our study fu-
ture research is needed to further investigate these findings.
References 1.
Lerner E, Schwartz, R, Coule, P, Winstein E, Cone DC, Hunt RC, Sasser SM, Liu JM, Nudell N, Wedmore I, Hammond J, Bulger E, Salomone J, Sanddal T, Lord G, Markenson D, O’Conner R. Mass casualty triage: an evaluation of the data and development of a proposed national guideline. Disaster Med Public Health Prep. 2008;2:S25–S33. 2. Romig L. Pediatric triage: a system to JumpSTART your triage of young patients at MCIs. JEMS. 2002;27:52–63. 3. Cone D, Serra J, Burns K, MacMillan D, Kurland L, Van Gelder C. Pilot test of the SALT mass casualty triage system. Prehosp Emerg Care. 2009;13:536–40. 4. Sanddal T, Loyacono T, Sanddal N. Effect of JumpSTART training on immediate and short-term pediatric triage performance. Pediatr Emerg Care. 2004;20(11):749–53. 5. Sapp R, Brice J, Myers J, Hinchey P. Triage performance of first year medical students using a multiple casualty scenario, paper exercise. Prehosp Disaster Med. 2010;25(3): 239–45. 6. Lerner E, Schwartz R, Coule P, Pirrallo G. Use of SALT triage in a simulated mass-casualty incident. Prehosp Emerg Care. 2010;14(1):21–5. 7. Deluhery M, Lerner B, Pirrallo R, Schwartz R. Paramedic accuracy usiing SALT triage after a brief initial training. Prehosp Emerg Care. 2011;15:526–32. 8. Faul M, Wald M, Sullivent E, Sasser S, Kapil V, Lerner B, et al. Large cost savings realized from the 2006 field triage guideline: reduction in overtriage in U.S. trauma centers. Prehosp Emerg Care. 2012;16:222–9. 9. Burstein J. Mostly dead: can science help with disaster triage? Ann Emerg Med. 2009;54: 431.
