Interim analyses are often performed in clinical trials to allow early termination of the study as soon as the investigators can be confident of the results. Such interim analyses better address the ethical and efficiency concerns inherent in a large scale clinical trial. The goal of this research is to broaden the spectrum of studies in which interim analyses can be used and to facilitate the use of such designs. Specific areas of research include: 1) development of methods to adjust survival curves for bias introduced by interim analyses; 2) investigation of methods to allow more flexible determination of number and timing of analyses; 3) specification of group sequential test designs appropriate for analysis of variance; 4) development of estimation methods for use in sequential analysis of variance; 5) investigation into the robustness of group sequential methods to departures from standard assumptions; and 6) development of computer software to integrate new and existing group sequential methods into a statistical analysis package.
| Kittelson, J M; Emerson, S S (1999) A unifying family of group sequential test designs. Biometrics 55:874-82 |
| Emerson, S S; Kittelson, J M (1997) A computationally simpler algorithm for the UMVUE of a normal mean following a group sequential trial. Biometrics 53:365-9 |
