Abstract:
|
Correlated failure times arise frequently from case-control family studies of chronic diseases. Patterns of the dependence between the failure times of the paired relatives are often used as a first step to untangle the etiology of these diseases. Few approaches have been developed for taking into account the case-control sampling mechanism, and the adaptation of those that were developed for cohort data to case-control family data appears too complicated and computationally intensive to be practical. In this talk, I will present an approach that connects the cross-ratio function and the relative-risk function in a stratified proportional hazards model. This leads to the development of a set of partial likelihood-based estimating equations for the estimation of the dependence parameters. I will also discuss some other practical aspects of case-control family data, e.g, missing or mismeasured covariate problems, and approaches for handling them. Finally, a real data example will be analyzed to illustrate the methods.
|