A GENERALIZED ESTIMATING EQUATION APPROACH FOR MODELING RANDOM LENGTHBINARY VECTOR DATA

A common measure in clinical trials and epidemiologic studies is the n umber of events such as seizures, hospitalizations, or bouts of diseas e. Frequently, a binary measure of severity for each event is availabl e but is not incorporated in the analysis. This paper proposes methodo logy for jointly modeling the number of events and the vector of corre lated binary severity measures. Our formulation exploits the notion th at a given covariate may affect both outcomes in a similar way. We fun ctionally link the regression parameters for the counts and binary mea ns and discuss a generalized estimating equation (GEE) approach for pa rameter estimation. We discuss conditions under which the proposed joi nt modeling approach provides marked gains in efficiency relative to t he common procedure of simply modeling the counts, and we illustrate t he methodology with epilepsy clinical trial data.