Accepted for/Published in: Journal of Medical Internet Research
Date Submitted: Feb 23, 2026
Date Accepted: Jul 28, 2026
Technological Integration in Triage: A Systematic Review and Meta-analysis of Artificial Intelligence and Machine Learning compared with Conventional Triage, with Narrative Synthesis of Telehealth for Patient Prioritization in Emergency and Acute Care Settings
ABSTRACT
Background:
Triage is critical for emergency department (ED) efficiency and patient safety, yet conventional systems rely on subjective clinician judgment and limited snapshot data. Advances in artificial intelligence (AI), machine learning (ML), and telehealth technologies have introduced new approaches to patient prioritization, but their diagnostic performance and broader system-level implications remain incompletely synthesized
Objective:
To systematically evaluate the diagnostic accuracy of AI-, ML-, and telehealth-enhanced triage systems and to translate performance metrics into projected operational, economic, and equity impacts at the emergency department level.
Methods:
Comparative studies (2010-2025) evaluating technology-enhanced triage were identified via biomedical and engineering databases. Diagnostic Test Accuracy (PRISMA-DTA) protocols guided the review. A random-effects meta-analysis was performed for studies reporting the Area Under the Receiver Operating Characteristic curve (AUC), while scenario-based modelling translated diagnostic performance into projected impacts per 10,000 emergency department visits.
Results:
Twenty-seven studies met inclusion criteria. In adult populations, AI-assisted triage demonstrated superior discrimination for critical care or deterioration, with a pooled AUC of 0.87 (95% CI 0.86–0.89) compared to 0.74–0.80 for conventional triage. Admission prediction models showed similar pooled discrimination (AUC 0.87; 95% CI 0.79-0.93) but substantial heterogeneity. While pediatric studies reported high AUCs (0.85-0.99), data were insufficient for pooling. Narrative synthesis indicated that tele-triage improved front-door operational metrics, whereas patient-facing symptom checkers prioritized sensitivity over specificity. Modelling indicated that AI integration could reduce missed critical illness by 12-38 cases per 10,000 visits, yielding projected net cost savings and proportionally greater benefits for underserved populations.
Conclusions:
AI-assisted triage consistently outperforms conventional systems for adult critical outcomes and offers measurable operational and equity gains. However, evidence remains heterogeneous, and limited pediatric data highlight the need for robust prospective evaluations. Clinical Trial: CRD420251160796
Citation
Request queued. Please wait while the file is being generated. It may take some time.
Copyright
© The authors. All rights reserved. This is a privileged document currently under peer-review/community review (or an accepted/rejected manuscript). Authors have provided JMIR Publications with an exclusive license to publish this preprint on it's website for review and ahead-of-print citation purposes only. While the final peer-reviewed paper may be licensed under a cc-by license on publication, at this stage authors and publisher expressively prohibit redistribution of this draft paper other than for review purposes.