The effects of triage applying artificial intelligence on triage in the emergency department: A systematic review of prospective studies

Abstract

Background This study aimed to identify the effects of a prospective study applying artificial intelligence-based triage in the clinical field. Methods We conducted a systematic review of prospective studies. The Preferred Reporting Items for Systematic Review and Meta-Analysis (PRISMA) checklist was used to guide the systematic review and reporting. Three researchers independently extracted the data, assessed the study quality, and presented the findings in a descriptive summary. Inconsistencies between the researchers were resolved after discussion. We manually searched for relevant articles through databases, including CINAHL, Cochrane, Embase, PubMed, ProQuest, and two South Korean search engines (KISS and RISS) from March 9 to April 18, 2023. Results Of 1,633 articles, eight met the inclusion criteria for this review. Most studies applied machine learning to triage, and only one study was based on fuzzy logic. Except for one study, all used a 5-level triage classification system, and some developed target-level prediction models. Although the model performance exceeded 70%, the triage prediction accuracy varied from 33.9 to 99.9%. Other outcomes included time reduction, overtriage and undertriage checks, triage risk factors, and outcomes related to patient care and prognosis. Conclusions Triage nurses in the emergency department can use artificial intelligence as a supportive means for patient classification. Ultimately, we hope that it will be a resource that can reduce undertriage and positively affect patient health. Verification of the optimal artificial intelligence algorithm by conducting rigorous interdisciplinary research will be a powerful tool to support triage nurses' decision-making in overcrowded emergency departments. Thus, direct nursing activities will increase and become an important factor in improving the quality of nursing care. Trial registration: We have registered our review in PROSPERO (registration number: CRD***********).