Analysis of mixtures of drugs/chemicals along a fixed-ratio ray without single-chemical data to support an additivity model.

This article presents and illustrates an approach to designing and analyzing studies involving mixtures/combinations of drugs or chemicals along fixed-ratio rays of the drugs or chemicals for generalized linear models. When interest can be restricted to a specific ray, we consider fixed-ratio ray designs to reduce the amount of experimental effort. When a ray design is used, we have shown that the hypothesis of additivity can be rejected when higher order polynomial terms are required in the total dose-response model. Thus, it is important that we have precise parameter estimates for these higher order polynomial terms in the linear predictor. We have developed methodology for finding a D _s -optimal design based on this subset of the terms in the linear predictor. [ABSTRACT FROM AUTHOR]