Skip to main content

Table 4 Examples of estimates of verification bias corrected AUC when varying the true value of the AUC

From: Statistical methods to correct for verification bias in diagnostic studies are inadequate when there are few false negatives: a simulation study

 

AUC

Fully verified (True value)

0.600

0.700

0.800

0.900

With verification bias correction

    

Replication 1

0.284

0.503

0.688

0.873

Replication 2

0.754

0.838

0.742

0.894

Replication 3

0.567

0.578

0.740

0.873

Replication 4

0.689

0.779

0.883

0.920

Replication 5

0.648

0.755

0.856

0.827