Salvatore Silvestri1, Adam Field2, Neal Mangalat3, Tory Weatherford4, Christopher Hunter5, Zoe McGowan6, Zachary Stamile7, Trevor Mattox8, Tanner Barfield8, Aarian Afshari9, George Ralls10, Linda Papa11. 1. Program Director, Orlando Health Emergency Medicine Residency, Orlando Regional Medical Center, Orlando, Florida; Associate EMS Medical Director, Orange County EMS System, Orlando, Florida. 2. Resident, Emergency Medicine, University of Arizona, Tucson, Arizona. 3. Clinical Instructor, Department of Emergency Medicine, University of Texas-Houston, Texas, Houston. 4. Fellow, Pediatric Emergency Medicine, Orlando Health, Orlando, Florida. 5. Health Services Director, Orange County Government, Orlando, Florida; Associate EMS Medical Director, Orange County EMS System, Orlando, Florida. 6. Pediatric Emergency Medicine, Orlando Health, Orlando, Florida. 7. Emergency Department, Tuba City Regional Health Care, Tuba City, Arizona; Emergency Department, Flagstaff Medical Center, Flagstaff, Arizona. 8. University of Central Florida College of Medicine, Orlando, Florida. 9. Resident, Emergency Medicine, Emory University, Atlanta, Georgia. 10. Public Safety Director, Orange County Government, Orlando, Florida; EMS Medical Director, Orange County EMS System, Orlando, Florida. 11. Director of Academic Clinical Research, Attending Emergency Physician, Orlando Regional Medical Center, Orlando, Florida; Professor, University of Central Florida College of Medicine, Orlando, Florida; Associate Professor, Florida State University College of Medicine, Orlando, Florida.
Abstract
OBJECTIVES: We compared Sort, Assess, Lifesaving Intervention, Treatment/Transport (SALT) and Simple Triage and Rapid Treatment (START) triage methodologies to a published reference standard, and evaluated the accuracy of the START method applied by emergency medical services (EMS) personnel in a field simulation. DESIGN: Simulated mass casualty incident (MCI). Paramedics trained in START triage assigned each victim to green (minimal), yellow (delayed), red (immediate), or black (dead) categories. These victim classifications were recorded by investigators and compared to reference standard definitions of each triage category. The victim scenarios were also compared to the a priori classifications as developed by the investigators. SETTING: MCI field simulation. MAIN OUTCOME MEASURE: Comparison of the correlation of START and SALT triage methodologies to reference standard definitions. Another outcome measure was the accuracy of the application of START triage by EMS personnel in the field exercise. RESULTS: The strongest correlation to the reference standard was SALT with an r = 0.860 (p < 0.001) and κ = 0.632 (p < 0.001). START and SALT triage systems agreed 100 percent on both black and green classifications. There were significant correlations between the field triage and both START and SALT methods (p < 0.001, respectfully). SALT had a significantly lower undertriage rate (9 percent [95%CI 2-15]) than both START (20 percent [95%CI 11-28]) and field triage (37 percent [95%CI 24-52]). There were no significant differences in overtriage rates. CONCLUSIONS: In our study, the SALT triage system was overall more accurate triage method than START at classi-fying patients, specifically in the delayed and immediate categories. In our field exercise, paramedic use of the START methodology yielded a higher rate of undertriage compared to the SALT classification.
OBJECTIVES: We compared Sort, Assess, Lifesaving Intervention, Treatment/Transport (SALT) and Simple Triage and Rapid Treatment (START) triage methodologies to a published reference standard, and evaluated the accuracy of the START method applied by emergency medical services (EMS) personnel in a field simulation. DESIGN: Simulated mass casualty incident (MCI). Paramedics trained in START triage assigned each victim to green (minimal), yellow (delayed), red (immediate), or black (dead) categories. These victim classifications were recorded by investigators and compared to reference standard definitions of each triage category. The victim scenarios were also compared to the a priori classifications as developed by the investigators. SETTING: MCI field simulation. MAIN OUTCOME MEASURE: Comparison of the correlation of START and SALT triage methodologies to reference standard definitions. Another outcome measure was the accuracy of the application of START triage by EMS personnel in the field exercise. RESULTS: The strongest correlation to the reference standard was SALT with an r = 0.860 (p < 0.001) and κ = 0.632 (p < 0.001). START and SALT triage systems agreed 100 percent on both black and green classifications. There were significant correlations between the field triage and both START and SALT methods (p < 0.001, respectfully). SALT had a significantly lower undertriage rate (9 percent [95%CI 2-15]) than both START (20 percent [95%CI 11-28]) and field triage (37 percent [95%CI 24-52]). There were no significant differences in overtriage rates. CONCLUSIONS: In our study, the SALT triage system was overall more accurate triage method than START at classi-fying patients, specifically in the delayed and immediate categories. In our field exercise, paramedic use of the START methodology yielded a higher rate of undertriage compared to the SALT classification.