Fig. 1

Depicts the resulting classification and regression tree for predicting in-hospital mortality. The decision tree contains four predictors: time to appropriate antimicrobial therapy, APACHE II score, lactate, and age. Terminal nodes containing predictions for new observations include 1, 5, and 7 (predict death) and 4 and 8 (predict alive). To obtain a prediction, one starts at the top of the tree and follows the arrow corresponding to data for the new observation until a terminal node is reached