Artificial Intelligence Technology Enhancing Emergency Department Triage

Vollständiger Abstract

Worum geht es in dieser Arbeit?

Accurate triage is vital in emergency departments, where misidentification delays care, increases morbidity and mortality, and strains resources, particularly amidst overcrowding. Despite advances, gaps persist in understanding the role of artificial intelligence tools across diverse settings and their integration into clinical workflows. The aim of this integrative review was to evaluate artificial intelligence’s potential to enhance triage efficiency and accuracy, guided by the Technology Acceptance Model (TAM) and Systems Thinking Approach. CINAHL, PubMed, Ovid Medline, EBSCOhost, and IEEE Xplore were searched, identifying 283 articles, refined to 20 through exclusions for duplicates, abstracts, and relevance. Retrospective cohort designs predominated, complemented by cross-sectional, prospective, and observational studies spanning global contexts. The key message was that artificial intelligence tools significantly outperformed traditional triage methods like the Emergency Severity Index (ESI). Findings revealed artificial intelligence predicted critical outcomes (eg, mortality, intensive care unit admissions), reduced mistriage rates, and leveraged diverse data (structured and unstructured), though challenges like limited generalizability and data quality remain. These insights advance health care informatics by highlighting the need for enhanced electronic health record systems and training, offering a pathway to optimize ED resource allocation and patient outcomes. Future research should focus on multicenter validations and real-time implementations to fully realize the transformative potential of artificial intelligence in emergency care.