Skip to main content

Table 5 Cross-validation of prediction models a

From: The use of the Gail model, body mass index and SNPs to predict breast cancer among women with abnormal (BI-RADS 4) mammograms

 

Total study population ( N  = 464)

 

Tenfold cross-validation

 

AUC

95% CI

P -value b

AUC

95% CI

P -value b

Age, log Gail RR

0.6646

0.5970 to 0.7321

0.839

0.6482

0.5797 to 0.7167

0.159

Age, BMI

0.6845

0.6188 to 0.7501

0.210

0.6583

0.5911 to 0.7255

0.764

Age, log SNP RR

0.6848

0.6195 to 0.7501

0.197

0.6735

0.6077 to 0.7393

0.188

Age, BMI, log SNP RR

0.7007

0.6370 to 0.7645

0.061

0.6753

0.6099 to 0.7407

0.258

Gail factors

0.7144

0.6532 to 0.7755

0.044

0.6522

0.5855 to 0.7188

0.955

Gail factors, BMI

0.7279

0.6705 to 0.7854

0.014

0.6561

0.5919 to 0.7203

0.924

Gail factors, BMI, log SNP RR

0.7377

0.6808 to 0.7946

0.007

0.6727

0.6099 to 0.7356

0.493

  1. aAUC, Area under the receiver operating characteristic curve; BMI, Body mass index; CI, Confidence interval; RR, Relative risk; SNP, Single-nucleotide polymorphism. b P-values derived from DeLong test compared to model with age only.