Abstract:
|
The sample size re-estimation (SSR) adaptive design draws considerable attention because of their flexibility in modifying the sample size based on unblinded interim analysis results. The efficiency of SSR designs has been discussed comparing to the traditional designs (non-adaptive) and group sequential (GS) design. Some researchers raised the concerns that the SSR designs are inefficient and one can improve them uniformly by using the standard GS designs. In this talk, we will propose a SSR design based on expected net present value and compare the proposed SSR design with the traditional design, standard SSR, as well as GS designs with the consideration of the overall income, cost and profit in a simple drug development setting.
|