Emergency Severity Index: accuracy in risk classification

Abstract

ObjectiveTo verify agreement between estimative of predicted resources using the adapted Emergency Severity Index and the real amount of resources used by patients. To analyze the variables number of years since graduation, years of work experience and years of experience in emergency services especially with accurate anticipation of resources need.MethodsThis retrospective analytical study with a quantitative approach included 538 medical records of patients assisted by 11 triage nurses. Data collected were related to assistances carried out from December 2012 to February 2013.ResultsThere was no significant association between the adequacy of the number of resources used, based on Emergency Severity Index score, number of years since graduation, year of work experience or years of experience in emergency services. Kappa agreement coefficient (0.34) showed that agreement was low between predicted and real used number of resources.ConclusionNurses’ accuracy index to predict resources for patients care from emergency room using the adapted Emergency Severity Index was lower than results reported in the studies in the literature that used the original scale. There was low agreement of diagnostic exams predicted by nurses and those really performed. There was no association among correct prediction of resources needed, number of years since graduation, years of experience in emergency services and years of work experience in the unit where the study was done.

Initial Prehospital Rapid Emergency Medicine Score (REMS) as a Predictor of Patient Outcomes

“…Few standardized objective measures exist to help EMS systems reliably classify patient risk for hospitalization or death using evidence-based criteria. Risk stratification is a valuable operational tool to predict the urgency and resource requirements for individual patients or classes of patients (1). Risk stratification can help inform prioritization of patients when demand exceeds capacity, such as in the case of severe disease outbreak like the COVID-19 pandemic.…”

O uso de protocolos nas unidades de urgência e emergência: uma revisão integrativa

“…Após, os artigos serão apresentados conforme suas características e resultados encontrados na busca dos artigos selecionados conforme o tema do uso de protocolos de urgência e emergência em uma unidade hospitalar, frente às publicações e conhecimentos científicos ao tema proposto. (Silva et al, 2017).…”

Improving ED Emergency Severity Index Acuity Assignment Using Machine Learning and Clinical Natural Language Processing

“…20 Challenges to assigning ESI 1 or ESI 2 (resuscitation or emergent, respectively) and ESI 5 (nonurgent) are common, with specific difficulties reported in differentiating between ESI 2 (unstable) and ESI 3 (stable). 21 Given these challenges, machine learning (ML) approaches have been proposed to aid clinicians in various patient risk assessments. ML has been used to predict inhospital mortality, critical care (admission to an intensive care unit and/or in-hospital death), and hospitalization (direct hospital admission or transfer) in adults [22][23][24][25] and children.…”