Seminar Series: Yun-Hee Choi, PhD
A Correlated Competing-risks Model with Time-varying Covariates: An application to Hereditary Breast and Ovarian Cancer Families.
Yun-Hee Choi, PhD
Associate Professor
Department of Epidemiology and Biostatistics
Schulich School of Medicine & Dentistry
Western University
Abstract:
Time-to-event data arising from family-based studies is often complex due to various factors such as multiple outcomes, interventions, competing risks, and familial correlation. We aim to estimate the cancer risks over time and evaluate several risk factors associated with cancer occurrence. Hereditary breast and ovarian cancer families suffer from high risks of both breast and ovarian cancers and are recommended to undergo frequent screenings or prophylactic surgery for prevention or early detection. A shared frailty model with time-varying covariates is applied to evaluate the effects of mammographic screening and risk-reducing salpingo oophorectomy on breast cancer risks. The evaluation of these interventions is usually complicated because of their effects changing over time and the presence of correlated competing risks. We propose a competing risks model that accounts for time-varying interventions and their time-dependent effects and provide cause-specific penetrance estimates for breast and ovarian cancers in BRCA1 families. We apply our approach to 498 BRCA1 mutation carrier families recruited through the Breast Cancer Family Registry and illustrate the importance of our approach accounted for both competing risks and time varying effects when estimating cause-specific penetrance of breast cancer in the presence of ovarian cancer and death.
Short Biography:
Dr. Yun-Hee Choi is an Associate Professor of Biostatistics in the Department of Epidemiology and Biostatistics in the Schulich School of Medicine & Dentistry at Western University. Dr. Choi completed her PhD in Statistics at University of Waterloo and her post-doctoral fellowship at Lunenfeld Research Institute in Mt. Sinai Hospital in Toronto. Her research interest lies in developing and evaluating statistical methods for various types of events—including survival and longitudinal outcomes and risk predictions based on correlated data arising from families at high genetic risk.
