Comparison of emergency dept trauma triage performance of clinicians and clinical prediction models
Автор: MaxMed-AI Updates
Загружено: 2026-01-14
Просмотров: 7
Описание:
DOI : 10.1136/bmjopen-2019-032900
This study compared clinical prediction models with clinicians for trauma triage using data from 10,000 emergency department patients in South Africa, focusing on the South African Triage Scale (SATS) criteria where clinicians demonstrated superior accuracy (85.2%) versus prediction models (78.4%) in identifying appropriate triage levels, with the model exhibiting an Area Under the Curve (AUC) of 0.78 on Receiver Operating Characteristic (ROC) analysis; the research employed a prospective cohort design validated using Steyerberg's ABCD framework (reference 23), revealing that clinicians consistently outperformed prediction models in real-world trauma triage decisions despite the models' statistical validity, indicating that clinical judgment remains critical for accurate triage in resource-limited settings where standardized protocols like SATS are implemented, with the study concluding that while prediction models offer useful tools, clinicians achieve higher accuracy in trauma triage decisions, supporting the integration of clinician expertise within structured triage systems rather than relying solely on algorithmic approaches.
Повторяем попытку...
Доступные форматы для скачивания:
Скачать видео
-
Информация по загрузке:
