ABSTRACT
OBJECTIVES: To compare the reliability, validity and resource utilization of the Emergency Severity Index (ESI) and Patient Acuity Category Scale (PACS) triage scales. METHODS: A descriptive, correlational, and cross-sectional design was used. Twenty-seven triage nurses were recruited to test interrater reliability for 20 patient case scenarios. Subsequently, interrater reliability was tested on 300 actual patients. Construct validity was analyzed using patients' hospital dispositions and resources utilized. RESULTS: For patient case scenarios, interrater reliability for both were very good, at 0.87 (95% CI: 0.86-0.88) for ESI and 0.88 (95% CI: 0.87-0.89) for PACS. For actual patients, interrater reliability for both were moderate, at 0.59 (95% CI: 0.50-0.68) for ESI and 0.49 (95% CI: 0.40-0.60) for PACS. Both ESI and PACS demonstrated construct validity with moderate correlations for hospital admissions and the number of resources used. PACS was unable to discriminate between patients that required more or less resources, whereas the ESI can. Patients that required two or more resources had higher rates of admission. CONCLUSIONS: Both triage systems demonstrated moderate interrater reliability and construct validity in triaging actual patients. The ESI has better resource discrimination ability than the PACS and can improve resource management in the ED.
