Table 3

Performance indices for the clinical prediction rule

Performance indexDerivation cohort
(n=9838)
Validation cohort
(n=5091)
Internal validation*
Mean (SD)95% CI
R2 0.0240.0180.025 (0.002)0.024 to 0.025
Brier scores0.0260.0260.026 (0.001)0.025 to 0.026
AUC0.76 (0.73 to 0.79)†0.73 (0.68 to 0.77)†0.76 (0.01)0.71 to 0.80
Discrimination slope0.0330.0290.034 (0.004)0.033 to 0.035
Calibration (p value)0.6050.3110.224 (0.208)0.100 to 0.995
Calibration in the large (p value)<0.001<0.0010.028 (0.005)0.024 to 0.032
Sensitivity (%)70.869.670.7 (2.7)65.5 to 76.1
Specificity (%)67.866.867.8 (0.5)66.9 to 68.8
Accuracy rate (%)67.966.967.8 (0.5)67.0 to 68.8
Positive predictive value (%)5.85.55.8 (0.4)5.0 to 6.7
Negative predictive value (%)98.898.798.8 (0.1)98.6 to 99.1
Positive likelihood ratio2.202.102.19 (0.09)2.02 to 2.37
Negative likelihood ratio0.430.460.43 (0.04)0.35 to 0.51
  • *1000bootstrap samples were generated by using resampling with replacement, and averages of these samples were presented to demonstrate the validity of the prediction rule.

  • †AUC (95% CI).

  • AUC, area under the receiver operating characteristic curve.