Conference Program Home
  My Program

All Times EDT

Abstract Details

Activity Number: 385 - SPEED: Statistical Methods and Applications in Medical Research, Risk Analysis, and Marketing Part 1
Type: Contributed
Date/Time: Wednesday, August 10, 2022 : 8:30 AM to 10:20 AM
Sponsor: Biopharmaceutical Section
Abstract #322520
Title: Overlap Weight-Based Adaptive Bayesian Commensurate Prior for Augmenting the Control Arm of a Randomized Controlled Trial
Author(s): Yeonil Kim* and Erina Paul
Companies: Merck & Co., Inc. and Merck & Co., Inc.
Keywords: Historical control; Overlap weight; Commensurate prior
Abstract:

It is of great interest in incorporating historical control for augmenting the control arm in the current clinical trials. Given pre-measured baseline characteristics of historical control are sufficiently similar to those of the current control group, by incorporating historical control, we can expect more precise point estimates, increased power, reduced sample size and cost/duration of the trial. To quantify a degree of borrowing information from historical data, propensity score (PS) with Bayesian methods have been widely used. We propose a method that adopts overlap weighting with commensurate prior. This method first generates overlap-weighted data in which covariates between treatment and control groups can be balanced. As a next step, the commensurate prior can be used to adaptively borrow information from the historical data in the absence of evidence for heterogeneity. The degree of the borrowing is partially determined by the commensurability or similarity of the information between the current and historical data. A simulation study is conducted to evaluate the performance of the proposed method by comparing with other existing methods.


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

Back to the full JSM 2022 program