Conference Program Home
  My Program

All Times EDT

Abstract Details

Activity Number: 536 - Advanced Bayesian Methods for Modern Clinical Trials
Type: Invited
Date/Time: Thursday, August 11, 2022 : 10:30 AM to 12:20 PM
Sponsor: International Society for Bayesian Analysis (ISBA)
Abstract #320351
Title: Estimating the Design Operating Characteristics in Bayesian Adaptive Clinical Trials
Author(s): Shirin Golchi*
Companies: McGill University
Keywords: Bayesian test statistic; Constrained design; Operating characteristics; Ordinal scale endpoint; Sampling distribution; Trial simulation

Bayesian adaptive designs have gained popularity in all phases of clinical trials in the past few decades. Design of Bayesian adaptive trials, however, requires extensive simulation studies. The required computation may become infeasible in complex modelling frameworks or time sensitive settings. In this talk, I propose a set of methods for efficient estimation and uncertainty quantification for the design operating characteristics of Bayesian adaptive trials. The proposed approach employs spatial modelling techniques to provides estimates of the sampling distribution of the ``test statistic" -typically a posterior or posterior predictive probability statement in Bayesian adaptive designs- throughout the model parameter space. Various design operating characteristics can then be readily obtained as quantiles of this sampling distribution without additional simulations. I consider design of an adaptive clinical trial with the ordinal scale disease progression endpoint analyzed via the proportional odds model to showcase the implementation and performance of the proposed approach.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2022 program