Skip to main content

Table 4 Logistic regression/odds ratio, OCC, calibration (HL), discrimination (ROC) for CASUS, SOFA, SAPSII, APACHEII derivatives

From: A comparative study of four intensive care outcome prediction models in cardiac surgery patients

Derivative of the Scoring model Logistic Regression OCC HL test ROC-Analysis
  O R 95%-CI % χ 1 p-value AUC 95%-CI
Mean-CASUS 2.04 1.85-2.62 98.3 5.75 0.68 0.991 0.987-0.995
Mean-SOFA 2.76 2.43-3.12 97.3 11.33 0.18 0.96 0.94-0.98
Mean-SAPS II 1.26 1.22-1.29 97.2 5.22 0.73 0.982 0.975-0.988
Mean-APACHE II 1.64 1.54-1.75 96.2 3.78 0.88 0.97 0.96-0.97
Max-CASUS 1.60 1.51-1.70 97.8 2.12 0.95 0.98 0.97-0.99
Max-SOFA 1.81 1.69-1.95 95.6 10.54 0.16 0.92 0.90-0.95
Max-SAPS II 1.15 1.13-1.18 95.9 3.75 0.88 0.95 0.94-0.96
Max-APACHE II 1.35 1.30-1.40 95.5 14.38 0.07 0.93 0.91-0.95
  1. 95%-CI: 95%-confidence interval, AUC: Area under ROC curve, HL: Hosmer-Lemeshow, OCC: overall correct classification, OR: Odds ratio for risk of mortality, ROC: receiver operating characteristic.