Prediction of Outcome after Severe and Moderate Head Injury: An Application of Classification and Regression Tree (CART) Technique

Vineet Kumar Kamal, All India Institute of Medical Sciences
Ravindra Mohan Pandey, All India Institute of Medical Sciences
Deepak Agrawal, All India Institute of Medical Sciences

For developing and validating a prognostic model for in-hospital mortality and unfavourable outcome at 6-months in moderate and severe head injury patients, a CART technique was employed in the analysis of a tertiary care trauma database (n=1466 patients) by using 24 prognostic indicators. For in-hospital mortality, there were 7 terminal nodes and the area under curve w as 0.83 and 0.82 for learning and test data sample respectively. The overall classification predictive accuracy was 82% for learning data sample and 79% for test data sample. For 6-months outcome, there were 4 terminal nodes and the AUC was 0.82 and 0.79 for learning and test data sample respectively. The overall classification predictive accuracy was 79% for learning data sample and 76% for test data sample. Methodologically, CART is quite different from the more commonly used statistical methods with the primary benefit of illustrating the important prognostic variables as related to outcome.

See extended abstract

 Presented in Session P2. Data and Methods/Applied Demography/ Spatial Demography/ Demography of Crime