Description
In this workshop we will unveil EaSt-2001, an important upgrade of the EaSt-2000 software package for the design and interim monitoring of group-sequential clinical trials. They key features of this software are the great flexibility it provides at the design phase to experiment with a number of alternative designs before choosing the one that is optimal for the study, and the great flexibility it provides at the interim monitoring phase to deviate from the planned timing and number interim looks by utilizing the Lan-DeMets error spending function methodology. Early stopping is permitted both for efficacy and for futility. The basic statistical principles of sequential design and analysis including the generation of stopping boundaries, the preservation of the type-I and type II error, the computation of the error spending function, and sample-size re-estimation will be explained and illustrated with the help of EaSt. Information based monitoring will be described.
|
