Author ORCID Identifier
https://orcid.org/0009-0003-2304-7573
Defense Date
2026
Document Type
Dissertation
Degree Name
Doctor of Philosophy
Department
Biostatistics
First Advisor
Robert Perera
Second Advisor
Roy Sabo
Third Advisor
Nolan Wages
Fourth Advisor
Autumn Lanoye
Fifth Advisor
Alexander Krist
Abstract
Trial designs that improve efficiency in phase III randomized clinical trials (RCTs) are of interest to the Food and Drug Administration, pharmaceutical companies, and researchers alike. One class of designs that aims to improve the efficiency of RCTs while accounting for ethical considerations is integrated multiple adaptive designs (IMADs). IMADs incorporate both ethical allocation and efficiency through a multiple objective function, which simultaneously optimizes response-adaptive randomization and sample size minimization. IMADs have been proposed for trials with continuous and binary outcomes at a single site with two treatment arms. As a result, IMADs are limited in their application. To address this, three extensions of continuous IMADs are proposed. The first specifies two pipelines, each of which incorporates more than two arms into the study. One pipeline is based on a global comparison, while the other is based on pairwise comparisons of interest. The second examines how the inclusion of classic early stopping rules for superiority or futility affects the design as differences deviate from a clinically informative difference. The third has two designs that consider different sites within its design, one which treats sites as fixed effects, the other that treats them as random. Each of these extensions, with varying degrees of success, addressed the problems they were designed to address while maintaining the desired properties of IMADs. Details of each extension, the strengths and limitations of the results, and an R package with the relevant functions are presented.
Rights
© The Author
Is Part Of
VCU University Archives
Is Part Of
VCU Theses and Dissertations
Date of Submission
8-5-2026