Projecting Individualized Absolute Invasive Breast Cancer Risk in African American Women

摘要: Background The Breast Cancer Risk Assessment Tool of the National Institute (NCI) is widely used for counseling and determining eligibility breast cancer prevention trials, although its validity projecting risk in African American women uncertain. We developed a model absolute invasive compared projections with those from Tool. Methods Data 1607 1647 control subjects Women’s Contraceptive Reproductive Experiences (CARE) Study were to compute relative attributable risks that based on age at menarche, number affected mother or sisters, previous benign biopsy examinations. Absolute obtained by combining this information data incidence NCI’s Surveillance, Epidemiology End Results Program national mortality data. Eligibility screening Tamoxifen Raloxifene (STAR) trial determine how new would affect eligibility, independent Health Initiative (WHI) assess well numbers cancers predicted agreed observed cancers. Tables graphs estimating confidence intervals women. Relative family history biopsies estimated CARE population lower than Tool, as was discriminatory accuracy (i.e., concordance). Using STAR trial, we 30.3% have had 5-year least 1.66% use model, only 14.5% it calibrated) WHI, except underestimated Conclusions usually gave higher estimates recommended regarding their cancer.