Skip to main content
. 2017 Mar 14;19:29. doi: 10.1186/s13058-017-0820-y

Table 5.

Discrimination performance of final model and sub models in comparison to established risk models

Model AUCa 95% CI LRb
1. Percentage mammographic density, age at mammography, BMI 0.63 0.60–0.65 109.2
2. Model 1 + family history of breast cancer, HRT use 0.64 0.62–0.67 122.0
3. Model 2 + absolute differences for calcifications, masses, density 0.70 0.68–0.72 193.7
4. Model 3 + interaction between percentage density and masses 0.71 0.69–0.73 233.6
Established risk models for comparison
 Tyrer-Cuzickc 0.63 0.60–0.65 88.2
 Gaild 0.56 0.53–0.58 34.9

BMI Body mass index, HRT Hormone replacement therapy, LR Likelihood ratio

aAUC was evaluated for the absolute risks of stated models

bChi-square test of β = 0

cTyrer-Cuzick model included risk factors of age, age at menarche, age at first child, menopause, length, weight, HRT, hyperplasia, atypical hyperplasia, lobular cancer in situ, and first-/second-degree family history of breast cancer. Data coding was done according to the Tyrer-Cuzick protocol

dGail model included risk factors of age, age at menarche, age at first live birth, number of previous breast biopsies, atypical hyperplasia, and first-degree family history of breast cancer. Data coding was done according to the Gail protocol