Fall Seminar Series  December 3, 2009
University of Minnesota
School of Statistics
College
of Liberal Arts

Extended Case-Control Designs for Cost-Effective Translational Studies

Yongzhao Shao
Department of Statistics
  Iowa State University

Thursday, December 3, 2009
3:30 PM, 115 Ford Hall
Minneapolis, East Bank Campus
Social at 3:00 PM, 300 Ford Hall

 

Abstract

Case-control designs are widely used to assess relative risks of various diseases in epidemiology and many medical studies. In recent clinical and translational studies, however, there are increasing demands for cost-effective study designs for assessing both the relative and absolute risks of disease progression and other adverse events. In this talk, we will discuss some extended case-control designs that combine strengths of the classical case-control and case-cohort study designs. Simple forms of semi-parametric likelihood models are suggested for analyzing the data collected from the study designs. Maximum likelihood estimates of the model parameters can be calculated using existing or freely available software, and are relatively straightforward to interpret by clinical investigators. Numerical demonstrations are also discussed. (This talk is based on join work with Shulian Shang at New York University).