Validity and reliability of the emergency severity index in a Spanish hospital

Abstract

Saturation in hospital emergency departments is one of the main safety problems for the patient, which can generate negative consequences for their health. In response to this issue, triage systems are developed to organize the flow of patients in order to allow the most urgent ones to be treated first. The Emergency Severity Index (ESI) is the most used triage system in the USA and it has been implemented in the General Hospital of La Palma since 2010. The objective of this study is the validation of the ESI adapted to our hospital through the study of its degree of reliability, as well as the criterion validity. The sample consisted of 240 randomly selected cases, with proportional representation of emergencies attended in 2015 and their fraction of urgent ones (Levels 1 and 2). Criterion validity was estimated by sensitivity, specificity, and predictive result values. For reliability, the degree of agreement among the nurses was studied by means of the adapted kappa index kc2. Criterion validity showed a sensitivity of 89% (85–93%) and a specificity of 97% (94–99%), with a positive predictive value of 68% (62–74%) and a negative predictive value of 99% (98–100%) for the discrimination of urgent cases. The reliability analysis showed a kc2 = 0.94 (0.84–0.99) index, a very good agreement according to Landis-Koch criteria. The results of our study have shown adequate validity and reliability in the adaptation and implementation of an ESI triage system suited to the specific conditions of a hospital emergency service in Spain.