Analyzing bivariate repeated measures for discrete and continuous outcome variables

A considerable body of literature has arisen over the past 15 years for analyzing univariate repeated measures data. However, it is rare in applied biomedical research for interest to be restricted to a single outcome measure. In this paper, we consider the case of bivariate repeated measures. We apply a generalized estimating equations (GEE) approach to relate each set of repeated measures to important explanatory variables. We then invoke the seemingly unrelated regression paradigm to combine these GEE models into an overall analysis framework. This approach provides a great deal of flexibility in modeling the relationships to fixed and time-dependent covariates for each set of outcome variables. Estimation and hypothesis testing issues are described and the methodology is illustrated with an example.