Citation

Tran Z, Zhang W, Verma A, Cook A, Kim D, Burruss S, Ramezani R, Benharash P. J. Trauma Acute Care Surg. 2021; ePub(ePub): ePub.

Copyright

(Copyright © 2021, Lippincott Williams and Wilkins)

DOI

10.1097/TA.0000000000003416

PMID

unavailable

Abstract

BACKGROUND: Existing mortality prediction models have attempted to quantify injury burden following trauma-related admissions with the most notable being the Injury Severity Score (ISS). Although easy to calculate, it requires additional administrative coding. International Classification of Diseases (ICD)-based models such as the Trauma Mortality Prediction Model (TMPM-ICD10) circumvent these limitations, but they utilize linear modeling which may not adequately capture the intricate relationships of injuries on mortality. Using ICD-10 coding and machine learning algorithms, the present study utilized the National Trauma Data Bank (NTDB) to develop mortality prediction models whose performance was compared to logistic regression, ISS and TMPM-ICD10.

METHODS: The 2015-2017 NTDB was used to identify adults following trauma-related admissions. Of 8,021 ICD-10 codes, injuries were categorized into 1,495 unique variables. The primary outcome was in-hospital mortality. eXtreme Gradient Boosting (XGBoost), a machine learning technique that uses iterations of decision trees, was used to develop mortality models. Model discrimination was compared to logistic regression, ISS and TMPM-ICD10 using receiver operating characteristic (ROC) and probabilistic accuracy with calibration curves.

RESULTS: Of 1,611,063 patients, 54,870 (3.41%) experienced in-hospital mortality. Compared to those who survived, those who died more frequently suffered from penetrating trauma and had a greater number of injuries. The XGBoost model exhibited superior ROC (0.863 (95% CI: 0.862-0.864) compared to logistic regression (0.845 (95% CI: 0.844-0.846)), ISS (0.828 (95% CI: 0.827-0.829)) and TMPM-ICD10 (0.861 (95% CI: 0.860-0.862)) (all p<0.001). Importantly, the ML model also had significantly improved calibration compared to other methodologies (XGBoost: coefficient of determination (R2) = 0.993, logistic regression: R2 = 0.981, ISS: R2 = 0.649, TMPM-ICD10: R2 = 0.830).

CONCLUSIONS: Machine learning models using XGBoost demonstrated superior performance and calibration compared to logistic regression, ISS and TMPM-ICD10. Such approaches in quantifying injury severity may improve its utility in mortality prognostication, quality improvement and trauma research. LEVEL OF EVIDENCE: Level III.

Language: en