Title
Good Health: Statistical Challenges in Personalizing Disease Prevention
Speaker
Alice Whittemore, PhD. Department of Health Research and Policy, Stanford University School of Medicine
Abstract
Increasingly, patients and clinicians are basing health care decisions on statistical models that use a person’s covariates to assign him/her a probability of developing a disease or other adverse outcome in a given future time period. Of particular interest is the extent to which such models can be improved by adding information from biological markers. In this talk I will describe some of the statistical opportunities and challenges that arise when evaluating the accuracy and utility of these models using longitudinal cohort data from subjects at risk of the outcome. I will illustrate the issues by considering risk models for the development of breast cancer, ovarian cancer and other chronic diseases.
