Enhancing Paramedic Decision-Making Through Artificial Intelligence-Based Triage Systems In Prehospital Emergency Care

Abstract

Background:
Paramedics operate in complex, high-pressure environments where rapid and accurate triage decisions are essential to patient survival. Traditional triage systems often rely on subjective clinical judgment, which can lead to inconsistency and delayed intervention. Advances in artificial intelligence (AI) have introduced data-driven triage tools designed to augment paramedic decision-making and improve patient outcomes in prehospital emergency care (PEC).

Objective:
This systematic review aimed to evaluate the effectiveness of AI-based triage systems in enhancing paramedic decision-making, improving triage accuracy, and optimizing response efficiency in prehospital settings.

Methods:
A systematic search was conducted across PubMed, Scopus, Web of Science, IEEE Xplore, and ScienceDirect for studies published between 2020 and 2025. Following PRISMA 2020 guidelines, 24 studies met the inclusion criteria. Data were extracted on AI model types, study design, triage outcomes, and clinical effectiveness. Quality appraisal was performed using the Newcastle–Ottawa Scale and the Joanna Briggs Institute checklist.

Results:
AI-based triage tools consistently outperformed conventional triage systems, demonstrating higher predictive accuracy (AUC range: 0.82–0.93) and reduced under-triage rates. Integrating AI decision support improved paramedic confidence, reduced cognitive load, and enhanced communication with receiving hospitals. Moreover, system-level outcomes such as response time reduction (8–21 seconds) and optimized resource allocation were reported. However, challenges remain regarding algorithm transparency, data interoperability, and ethical oversight.

Conclusion:
AI-based triage systems represent a paradigm shift in prehospital emergency medicine. When appropriately implemented, they enhance clinical decision-making, strengthen patient safety, and support data-informed EMS management. Nevertheless, responsible integration requires ethical frameworks, explainable AI models, and structured training for paramedics to ensure human-centered and trustworthy application.