Haplotype-Based Statistical Inference for Population-Based Case-Control Studies with Complex Sample Designs

• Activity Number: 349
• Type: Contributed
• Date/Time: Tuesday, July 31, 2012 : 10:30 AM to 12:20 PM
• Sponsor: Section on Survey Research Methods
• Author(s): Daoying Lin*+ and Yan Li
• Companies: The University of Texas at Arlington and The University of Texas at Arlington
• Keywords: Association studies ; Case-control studies ; Haplotype ; Stratified multistage cluster sampling ; Stratified multistage cluster sampling

Abstract

The use of complex sampling for the selection of study subjects is becoming more common in population-based case-control or cross-sectional studies. This is prompted by the usual advantages of conducting complex sampling, such as time and cost efficiency. More importantly, the use of proper complex sample designs can also obtain representative samples from the target population and thus avoid the biased selection of controls and/or cases. However, at least two complications, i.e. differential population weights and intracluster correlation, are induced by the complex sampling. Although most SNP-based association studies with complex sampling account for these complications, many recent haplotype-based genetic association studies with complex sampling tend to ignore them, which can lead to invalid inferences. Therefore there is a need to develop methods for taking into account the complications induced by complex sampling in haplotype-based association studies. In this paper, we propose methods that properly account for complex sample designs. Our methods are evaluated via simulation studies and illustrated using data from a case-control study, the U.S. Kidney Cancer Study.