|
Abstract:
|
Identifying the "right" dose is a very critical step in the clinical development process of any medicinal drug. Selecting too high a dose can result in unacceptable toxicity and associated safety problems, while choosing too low a dose leads to smaller chances of showing sufficient effectiveness in confirmatory trials. In this paper we consider the problem of obtaining dose finding designs, efficiently learning about the minimum effective dose (MED), defined as the smallest dose which shows a clinically relevant and a statistically significant effect. The sensitivity of these designs with respect to misspecification in the true parameter and in the underlying dose-response model is investigated. Robust optimal designs are constructed taking into account, simultaneously, a set of potential dose response profiles within classes of models commonly used.
|
