Presented in poster form at the Eastern Association for the Surgery of Trauma, Fort Meyers, FL, January 16, 2007.
Bayesian Logistic Injury Severity Score: A Method for Predicting Mortality Using International Classification of Disease-9 Codes
Version of Record online: 17 APR 2008
© 2008 by the Society for Academic Emergency Medicine
Academic Emergency Medicine
Volume 15, Issue 5, pages 466–475, May 2008
How to Cite
Burd, R. S., Ouyang, M. and Madigan, D. (2008), Bayesian Logistic Injury Severity Score: A Method for Predicting Mortality Using International Classification of Disease-9 Codes. Academic Emergency Medicine, 15: 466–475. doi: 10.1111/j.1553-2712.2008.00105.x
- Issue online: 17 APR 2008
- Version of Record online: 17 APR 2008
- Received December 6, 2007; revisions received January 25 and February 8, 2008; accepted February 11, 2008.
- International Classification of Diseases;
- trauma severity indices;
- predictive value of tests;
- survival rate
Objectives: Owing to the large number of injury International Classification of Disease-9 revision (ICD-9) codes, it is not feasible to use standard regression methods to estimate the independent risk of death for each injury code. Bayesian logistic regression is a method that can select among a large numbers of predictors without loss of model performance. The purpose of this study was to develop a model for predicting in-hospital trauma deaths based on this method and to compare its performance with the ICD-9–based Injury Severity Score (ICISS).
Methods: The authors used Bayesian logistic regression to train and test models for predicting mortality based on injury ICD-9 codes (2,210 codes) and injury codes with two-way interactions (243,037 codes and interactions) using data from the National Trauma Data Bank (NTDB). They evaluated discrimination using area under the receiver operating curve (AUC) and calibration with the Hosmer-Lemeshow (HL) h-statistic. The authors compared performance of these models with one developed using ICISS.
Results: The discrimination of a model developed using individual ICD-9 codes was similar to that of a model developed using individual codes and their interactions (AUC = 0.888 vs. 0.892). Inclusion of injury interactions, however, improved model calibration (HL h-statistic = 2,737 vs. 1,347). A model based on ICISS had similar discrimination (AUC = .855) but showed worse calibration (HL h-statistic = 45,237) than those based on regression.
Conclusions: A model that incorporates injury interactions had better predictive performance than one based only on individual injuries. A regression approach to predicting injury mortality based on injury ICD-9 codes yields models with better predictive performance than ICISS.