Electronic Thesis and Dissertation Repository

Degree

Master of Science

Program

Epidemiology and Biostatistics

Collaborative Specialization

Biostatistics

Abstract

Mammographic screening (MS) and prophylactic surgery (PS) can potentially reduce cancer risks in BRCA1/2 mutation families. The evaluation of these interventions is complicated both by the presence of competing events and by their values changing over time. We propose a competing risks model and provide cause-specific penetrance estimates to account for timevarying covariates in the presence of competing events. A shared frailty model is specified to account for familial residual dependence and is estimated to correct for the ascertainment bias induced by the sampling of families through probands via an ascertainment corrected likelihood approach. We evaluate the performance of the estimators for the parameters in the model and the plug-in estimators for the penetrance functions via simulations. We apply our proposed method to BRCA1/2 mutation carrier families recruited through the Breast Cancer Family Registries and evaluate the effects of PS and MS on breast and ovarian cancer risks.

Share

COinS