Developing and validating subjective and objective risk-assessment measures for predicting mortality after major surgery: An international prospective cohort study
Table 4
Performance metrics for clinician prediction versus SORT and versus a logistic regression model combining clinician and SORT prediction.
Calculations based on the subset of patients in whom clinician judgement alone was used to estimate risk (n = 17,845). The reported AUROC for the combined model is the optimism-corrected value from bootstrapped internal validation.