Figure 1

New Classification and Regression Tree to Predict Clinical Deterioration. The classification tree consists of four biomarker-based decision rules with five terminal daughter nodes. The tree incorporates four of five PERSEVERE biomarkers: interleukin-8 (IL8), heat shock protein 70 kDa 1B (HSPA1B), chemokine ligand 3 (CCL3), and matrix metalloproteinase-8 (MMP8). Each node denotes the number of subjects in the node, the serum concentration of a given biomarker determining the branch point (pg/mL), and both the total number and accompanying rate of subjects with clinical deterioration or clinical stability. Terminal Nodes 1 and 3 are considered low risk terminal nodes; terminal nodes 2, 4, and 5 are high risk. The AUROC for this model was 0.81.