Table 4 Comparison of diagnostic prediction models.

From: Development and validation of a diagnostic prediction model for children with pertussis

Diagnostic prediction models

Sensitivity

Specificity

S.E

AUC-ROC

P valuea

NRIb

Estimate

95% CI

P valuea

Model A

0.752

0.696

0.024

0.788

    

Model B

0.757

0.667

0.025

0.778

 > 0.05

− 0.036

− 0.124–0.076

0.466

Model C

0.721

0.726

0.025

0.773

 > 0.05

− 0.030

− 0.138–0.081

0.573

Model D

0.622

0.793

0.027

0.753

 > 0.05

− 0.068

− 0.181–0.059

0.260

Model E

0.622

0.689

0.029

0.674

 < 0.01

   
  1. AUC, area under curve; ROC, receiver operating characteristic; NRI, net reclassification improvement.
  2. aCompared with model A.
  3. bB = 1000 repetitions, bootstrap.
Back to article page