A Machine Learning Trauma Triage Model for Critical Care Transport
An erratum to this article was published (JAMA Netw Open, 1 Jul 2025 ↗)
Authors' conclusion
In this cohort study of critically ill patients with trauma in the prehospital setting, an ML-based triage model using physiological features provided accurate predictions of lifesaving intervention delivery to single patients.
Study context
- Population
- Critically ill patients with trauma transported by a large critical care air transport system serving Pennsylvania and surrounding states, 2018–202115,088 two-minute epochs
- Participants
- 2,809 patients
- Outcome
- Administration of a lifesaving intervention (LSI) within a 2-minute epoch during patient care
Methodological detailsmachine learning prediction model, ensemble machine learning (ML) approach
- Machine learning prediction model
- Ensemble ML approach
Main finding
As reported in the abstract
Prediction of LSI in the next 2-minute epoch
- AUC
- 0.810 (95% CI 0.782-0.842)
- Sensitivity
- 0.268 (95% CI 0.193-0.357)
- PPV
- 0.301 (95% CI 0.228-0.356)
- Positive likelihood ratio
- 6.793 (95% CI 4.887-8.795)
- Specificity
- 0.960 (95% CI 0.947-0.972)
- NPV
- 0.953 (95% CI 0.943-0.964)
- Negative likelihood ratio
- 0.763 (95% CI 0.680-0.837)
Observational data: associations, not effects.
Further findings
Lifesaving intervention (LSI) subcategories, features up to 15 min before LSI, first LSI only, mechanism of injury
- Performance equivalent or better
Original publication
A Machine Learning Trauma Triage Model for Critical Care Transport. JAMA Netw Open. Published 2 June 2025. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →