Mixture Model for Individual and Combined Data Using Estimating Equations

Abstract

When performing analysis of individual data on the application of a particu lar drug, it is useful to study the within variability. But when two drugs are used in combination, it is of more interest to study any combination effects on the subjects. In this paper we consider a new analytical framework that is a comb ination of the individual and co mbined data analyses, based on an estimating equation approach. The proposed analyses utilize a stochastic model for a two-drug combination and derive the mean and the variance terms based on Ito's calculus. The proposed estimation methods are used to estimate model parameters fro m both individual and co mbined data, and they provide the basis for model free synergy tests. The strength of the fit of the model to the data is examined by statistical measures and the graphical method. Simu lation studies were performed to show the strengths of the proposed approach in estimat ing the model parameters. A synergy test of the model fitted by the individual subjects confirmed that the combination of the isomers under study is synergistic in nature.