Association Models for Clustered Data with Binary and Continuous Responses

• Activity Number: 275
• Type: Contributed
• Date/Time: Tuesday, August 5, 2008 : 10:30 AM to 12:20 PM
• Sponsor: Biometrics Section
• Author(s): Lanjia Lin*+ and Stuart R. Lipsitz and Dipankar Bandyopadhyay and Debajyoti Sinha
• Companies: Florida State University and Harvard Medical School and Medical University of South Carolina and Florida State University
• Keywords: bivariate discrete and continuous responses ; bridge distribution ; probability integral transformation ; clustered data

Abstract

We consider the study in which the sampling unit is a cluster of subjects and each member of the cluster has mixed bivariate responses, a binary and a continuous outcome. We propose a new bivariate random effects model which induces associations between the binary outcomes in the cluster, the continuous outcomes in the cluster, the binary and continuous outcomes from different subjects in the cluster, as well as the binary and continuous outcomes from the same subject in the cluster. For the ease of interpretation of the regression effects, the marginal model of the binary response probability, integrated over the random effect, preserves the logistic form, and the marginal regression function of the continuous response preserves the linear form. Finally, we illustrate the maximum likelihood estimation of parameters of our model via the analysis of data from a development toxicity study.