Peer-reviewed article
Machine Learning–Based Prediction of Clinical Outcomes for Children During Emergency Department Triage
Tadahiro Goto, Carlos A. Camargo, Mohammad Kamal Faridi, Robert J. Freishtat, Kohei Hasegawa ·
Summary
Written by AI, not yet AICS's view · claude-haiku-4-5-20251001This study uses machine learning models to predict which children need critical care or hospitalization when they arrive at emergency departments. The researchers tested four different machine learning approaches on data from over 52,000 US children and compared them to standard triage methods. Machine learning models performed better at identifying children who would need hospital admission and reduced the number of critically ill children being undertriaged.
Key points
- Machine learning models showed higher discriminative ability than conventional triage for predicting hospitalization outcomes (C statistic 0.80 vs 0.73, P < .001)
- For critical care prediction, machine learning had better discrimination (C statistic 0.85 vs 0.78) though not statistically significant
- Machine learning reduced undertriaging of critically ill children in conventional triage levels 3-5
- Machine learning reduced overtriaging of less ill children who did not require inpatient care
- Deep neural networks, random forests, gradient-boosted decision trees, and lasso regression were all tested
Limitations
This is a prognostic study using historical data from 2007-2015, so it cannot establish whether using these machine learning models in actual practice would improve real-world outcomes for children. The study does not evaluate implementation feasibility or cost-effectiveness.
Abstract
Importance: While machine learning approaches may enhance prediction ability, little is known about their utility in emergency department (ED) triage. Objectives: To examine the performance of machine learning approaches to predict clinical outcomes and disposition in children in the ED and to compare their performance with conventional triage approaches. Design, Setting, and Participants: Prognostic study of ED data from the National Hospital Ambulatory Medical Care Survey from January 1, 2007, through December 31, 2015. A nationally representative sample of 52 037 children aged 18 years or younger who presented to the ED were included. Data analysis was performed in August 2018. Main Outcomes and Measures: The outcomes were critical care (admission to an intensive care unit and/or in-hospital death) and hospitalization (direct hospital admission or transfer). In the training set (70% random sample), using routinely available triage data as predictors (eg, demographic characteristics and vital signs), we derived 4 machine learning-based models: lasso regression, random forest, gradient-boosted decision tree, and deep neural network. In the test set (the remaining 30% of the sample), we measured the models' prediction performance by computing C statistics, prospective prediction results, and decision curves. These machine learning models were built for each outcome and compared with the reference model using the conventional triage classification information. Results: Of 52 037 eligible ED visits by children (median [interquartile range] age, 6 [2-14] years; 24 929 [48.0%] female), 163 (0.3%) had the critical care outcome and 2352 (4.5%) had the hospitalization outcome. For the critical care prediction, all machine learning approaches had higher discriminative ability compared with the reference model, although the difference was not statistically significant (eg, C statistics of 0.85 [95% CI, 0.78-0.92] for the deep neural network vs 0.78 [95% CI, 0.71-0.85] for the reference; P = .16), and lower number of undertriaged critically ill children in the conventional triage levels 3 to 5 (urgent to nonurgent). For the hospitalization prediction, all machine learning approaches had significantly higher discrimination ability (eg, C statistic, 0.80 [95% CI, 0.78-0.81] for the deep neural network vs 0.73 [95% CI, 0.71-0.75] for the reference; P < .001) and fewer overtriaged children who did not require inpatient management in the conventional triage levels 1 to 3 (immediate to urgent). The decision curve analysis demonstrated a greater net benefit of machine learning models over ranges of clinical thresholds. Conclusions and Relevance: Machine learning-based triage had better discrimination ability to predict clinical outcomes and disposition, with reduction in undertriaging critically ill children and overtriaging children who are less ill.