Close
  Indian J Med Microbiol
 

Figure 2: Comparison of receiver operating characteristic (ROC) curves from two independent studies of the comorbidity-polypharmacy score (CPS). Part (A) shows the comparison of ROC curves for models incorporating CPS (red) versus Charlson Scoring System (blue) for in-hospital mortality area under curve (AUC = 0.75 vs. 0.80, respectively; P = 0.02, modifi ed from Holmes et al., J Am Coll Surg, 2014;219(4):631). Part (B) shows the comparison of ROC curves for in-hospital mortality from a post hoc, previously unpublished, secondary analysis of 5,580 older trauma patients for ISS alone (AUC = 0.79), ISS+CPS (AUC = 0.83), ISS + AGE (AUC = 0.79), and ISS + AGE + CPS (AUC = 0.78, all, P < 0.01). As evident from the above comparisons, ISS in combination with CPS provides the best predictive model for in-hospital mortality among older trauma patients

Figure 2: Comparison of receiver operating characteristic (ROC) curves from two independent studies of the comorbidity-polypharmacy score (CPS). Part (A) shows the comparison of ROC curves for models incorporating CPS (red) versus Charlson Scoring System (blue) for in-hospital mortality area under curve (AUC = 0.75 vs. 0.80, respectively; <i>P</i> = 0.02, modifi ed from Holmes <i>et al</i>., J Am Coll Surg, 2014;219(4):631). Part (B) shows the comparison of ROC curves for in-hospital mortality from a <i>post hoc</i>, previously unpublished, secondary analysis of 5,580 older trauma patients for ISS alone (AUC = 0.79), ISS+CPS (AUC = 0.83), ISS + AGE (AUC = 0.79), and ISS + AGE + CPS (AUC = 0.78, all, <i>P</i> < 0.01). As evident from the above comparisons, ISS in combination with CPS provides the best predictive model for in-hospital mortality among older trauma patients