Design and validation of a clinical simulation method for teaching nurses to use the Emergency Severity Index for triage

Objective. To design and validate a clinical simulation method for training nurses to use the Emergency Severity Index (ESI) for triage. Methods. Descriptive, cross-sectional study carried out in the emergency department of Hospital Clinico Universitario Virgen de la Arrixaca in Murcia, Spain. The educational intervention had 3 phases focused on theory, skills, and application. The participants, who were nurses with at least a year's experience in the emergency department, numbered 55, 43, and 32 in each phase, respectively. They undertook training in ESI for triage (phase 1), participated in simulations with patient-actors (phase 2), and applied their acquired skills (phase 3). We used Cohen's K and the intraclass correlation coefficient (ICC) to compare the nurses' triage assignments to gold-standard triage assignments. Results. We had data for 1100, 799, and 410 observations of performance in each phase, respectively. Agreement statistics between the nurses' performance and the gold standard were 0.68, 0.67, and 0.68 (Cohen's K) and 0.870, 0.836, and 0.811 (ICC) in each phase, respectively. These results reflect good agreement. Conclusions. This study supports the relevance of clinical simulations for training nurses to implement structured triage, and in particular to use the ESI. We can state that this study opens the door to further research into whether the reliability of triage can be improved by using this approach, specifically whether it can facilitate better interobserver agreement.