Modelling covariance structure in ascending dose studies of isolated tissues and organs

This paper describes the analysis of two pharmacology assays: the guinea pig papillary muscle assay (an example of an isolated tissue assay) and an assay looking at pressure changes in isolated rat lungs. Both assays use an ascending dose design to minimize carryover effects. This is often necessary in these studies, due to the limited life span of the tissues. Various mixed models, with different covariance structures, are fitted to find the most appropriate model. These are then compared to two other possible methods of analysis: paired t‐tests and two‐way analysis of variance. For both assays, the mixed model was found to be the best approach. These examples illustrate the importance of modelling covariance structure correctly in any ascending dose study, whether in isolated organs/tissues, in animals or phase I volunteers. Copyright © 2003 John Wiley & Sons, Ltd.