On Estimating the Age-Dependent Population Attributable Fractions from Population-Based Case-Control Data

• Activity Number: 349
• Type: Contributed
• Date/Time: Tuesday, August 6, 2013 : 10:30 AM to 12:20 PM
• Sponsor: Biometrics Section
• Author(s): Wei Zhao*+ and Ying Qing Chen and Li Hsu
• Companies: University of Washington and Fred Hutchinson Cancer Research Center and Fred Hutchinson Cancer Research Center
• Keywords: Population attributable fraction ; Case-control study ; Epidemiologic methods ; Kernel estimator

Abstract

The population attributable fraction (PAF) is commonly used in public health to measure the risk of disease attributable to given risk factors. PAF takes both association and prevalence of risk factors into account thus provides an important link between causality and public health action. Population-based case-control study design has been widely used in epidemiologic studies, and it is of particular interest to evaluate the risk attributable to an exposure. Such excess risks could be highly dependent on age. In this article, we propose an efficient non-parametric estimator of age-dependent PAF from population-based case-control data. Estimation of the standard errors is derived from both asymptotic theory and bootstrapping. Simulations are conducted to evaluate its performance. Genetics and Epidemiology of Colorectal Cancer Consortium (GECCO) study data is used to demonstrate the application of the proposed methodology in real data analysis.