Machine learning-based triage to identify low-severity patients with a short discharge length of stay in emergency department

Yu-Hsin Chang; Hong-Mo Shih; Jia-En Wu; Fen-Wei Huang; Wei-Kung Chen; Dar-Min Chen; Yu-Ting Chung; Charles C N Wang

doi:10.1186/s12873-022-00632-6

Machine learning-based triage to identify low-severity patients with a short discharge length of stay in emergency department

BMC Emerg Med. 2022 May 20;22(1):88. doi: 10.1186/s12873-022-00632-6.

Authors

Yu-Hsin Chang¹, Hong-Mo Shih¹, Jia-En Wu², Fen-Wei Huang³, Wei-Kung Chen¹, Dar-Min Chen⁴, Yu-Ting Chung⁴, Charles C N Wang⁵

Affiliations

¹ Department of Emergency Medicine, China Medical University Hospital, No. 2, Yude Rd., North Dist., Taichung City, 40447, Taiwan.
² Department of Bioinformatics and Medical Engineering, Asia University, No. 500, Liufeng Rd., Wufeng Dist., Taichung City, 413305, Taiwan.
³ Department of Medical Research, China Medical University Hospital, No. 2, Yude Rd., North Dist., Taichung City, 40447, Taiwan.
⁴ Department of Emergency Medicine, Asia University Hospital, No. 222, Fuxin Rd., Wufeng Dist., Taichung City, 413505, Taiwan.
⁵ Department of Bioinformatics and Medical Engineering, Asia University, No. 500, Liufeng Rd., Wufeng Dist., Taichung City, 413305, Taiwan. cnwang@asia.edu.tw.

Abstract

Background: Overcrowding in emergency departments (ED) is a critical problem worldwide, and streaming can alleviate crowding to improve patient flows. Among triage scales, patients labeled as "triage level 3" or "urgent" generally comprise the majority, but there is no uniform criterion for classifying low-severity patients in this diverse population. Our aim is to establish a machine learning model for prediction of low-severity patients with short discharge length of stay (DLOS) in ED.

Methods: This was a retrospective study in the ED of China Medical University Hospital (CMUH) and Asia University Hospital (AUH) in Taiwan. Adult patients (aged over 20 years) with Taiwan Triage Acuity Scale level 3 were enrolled between 2018 and 2019. We used available information during triage to establish a machine learning model that can predict low-severity patients with short DLOS. To achieve this goal, we trained five models-CatBoost, XGBoost, decision tree, random forest, and logistic regression-by using large ED visit data and examined their performance in internal and external validation.

Results: For internal validation in CMUH, 33,986 patients (75.9%) had a short DLOS (shorter than 4 h), and for external validation in AUH, there were 13,269 (82.7%) patients with short DLOS. The best prediction model was CatBoost in internal validation, and area under the receiver operating cha racteristic curve (AUC) was 0.755 (95% confidence interval (CI): 0.743-0.767). Under the same threshold, XGBoost yielded the best performance, with an AUC value of 0.761 (95% CI: 0.742- 0.765) in external validation.

Conclusions: This is the first study to establish a machine learning model by applying triage information alone for prediction of short DLOS in ED with both internal and external validation. In future work, the models could be developed as an assisting tool in real-time triage to identify low-severity patients as fast track candidates.

Keywords: Decision-making support; Discharge length of stay; Emergency department; Machine learning; Streaming; Triage.

Publication types

Research Support, Non-U.S. Gov't

MeSH terms

Adult
Aged
Emergency Service, Hospital
Humans
Length of Stay
Machine Learning
Patient Discharge*
Retrospective Studies
Triage*