TESTS FOR TREND IN DEVELOPMENTAL TOXICITY EXPERIMENTS WITH CORRELATEDBINARY DATA

In this article, the operating characteristics of recently proposed te sts for trend in correlated binary data arising in laboratory studies of developmental toxicity are examined using both computer-generated a nd experimental data. Specifically, we consider adjusted Cochran-Armit ge tests based on the Rao-Scott transformation which are of the same g eneral form as that for uncorrelated data. In addition, generalized sc ore tests based on generalized estimating equations allowing for extra -binomial variation in the data are discussed. Specific forms of these statistics demonstrating favorable type I and type II error rates are identified and recommended for use in practice. The application of th ese tests is illustrated using data from studies of developmental toxi city that have been reported in the literature.