Inferential methods for adaptive group sequential design in clinical trial

Dorji, Passang (2025) Inferential methods for adaptive group sequential design in clinical trial. Masters thesis, Memorial University of Newfoundland.

[English] PDF - Accepted Version Available under License - The author retains copyright ownership and moral rights in this thesis. Neither the thesis nor substantial extracts from it may be printed or otherwise reproduced without the author's permission. Download (696kB)

Abstract

The group sequential designs of Randomized Clinical Trials collect data in stages and perform interim analysis or modification during the trial for various purposes. The balanced treatment allocation in group sequential designs raises ethical concerns, as it may expose half of the patients to potentially inferior or unsafe treatments. One potential modification from an ethical perspective could be the decision regarding treatment assignment for the subsequent stages. We propose an adaptive group sequential design by using information on treatment effects accumulated during the trial to modify treatment allocation in order to assign fewer patients to the potential inferior treatment arm. However, interim modification of a trial can increase the Type-I error rate. To address this, we propose a weighted t-test and derived formulas for establishing the critical values to control the inflation of Type-I error rates, and we also modify traditional twosample t-test. We conduct simulations to examine the performance of the inferential methods and the adaptive group sequential allocation. Simulation results indicate that the proposed test methods preserve the Type-I error rate and the proposed group sequential design with adaptive treatment assignment has the ethical advantage of reducing the number of patients exposed to inferior treatment.

Actions (login required)

View Item