Pseudolikelihood modeling of multivariate outcomes in developmental toxicology

The primary goal of this article is to determine benchmark doses based on t he ethylene glycol study, which comprises data from a developmental toxicit y study in mice. Because the data involve a vector of malformation indicato rs, a flexible model for multivariate clustered data is required. An expone ntial family model is considered and pseudolikelihood-based inferential too ls are proposed, hence avoiding excessive computational requirements.