POINT AND INTERVAL ESTIMATION FOLLOWING A SEQUENTIAL CLINICAL-TRIAL

The usual methodology employed in analysis after a sequential clinical trial is based on orderings of the possible samples resulting from th e design. However, this approach lacks flexibility for use in wider ap plications. In this paper two estimation techniques not based on order ings are considered and modified to obtain improved accuracy. A bias-a djusted maximum likelihood estimate together with a new and general me thod for setting confidence limits are discussed. The realistic scenar io of group sequential monitoring is assumed and methods for exact est imation are given. Accuracy of the methodology after a triangular test and an O'Brien & Fleming test are demonstrated through simulation.