Case-Control Sampling for Longitudinal Studies

• Activity Number: 559
• Type: Invited
• Date/Time: Wednesday, August 4, 2010 : 2:00 PM to 3:50 PM
• Sponsor: Section on Statistics in Epidemiology
• Author(s): Paul J. Rathouz*+ and Jonathan Schildcrout
• Companies: The University of Chicago and Vanderbilt University
• Keywords: Bias ; case-control sampling ; efficiency ; logistic regression ; outcome dependent sampling

Abstract

We discuss design and analysis of longitudinal studies under case-control sampling, wherein interest lies in the relationship between a longitudinal response and a set of covariates, but response is also related to the case-control sampling variable. We propose modelling based on an ancillary model for subjects' case-control status. In this approach, the analyst posits the population prevalence of being a case, which is then used as an offset term in the ancillary model. Parameter estimates from this model are used to correct sampling bias for the longitudinal response model of interest. We study design and analysis issues impacting study efficiency, namely: choice of sampling variable and the strength of its relationship to the response, sample stratification, and degree of flexibility of the ancillary model.