Activity Number:
|
480
- Foundational Aspects of Bayesian Analysis
|
Type:
|
Contributed
|
Date/Time:
|
Thursday, August 6, 2020 : 10:00 AM to 2:00 PM
|
Sponsor:
|
Section on Bayesian Statistical Science
|
Abstract #311130
|
|
Title:
|
Bayesian Sequential Probability Ratio Test on Vaccine Efficacy Trial
|
Author(s):
|
Erina Paul* and Santosh Sutradhar and Devan Mehrotra
|
Companies:
|
Merck & Co., Inc. and Merck & Co., Inc. and Merck
|
Keywords:
|
Bayesian;
Beta-binomial;
Poisson rates;
SPRT;
Vaccine efficacy
|
Abstract:
|
In vaccine efficacy (VE) trails, Frequentist fixed sample designs are very large as it requires recruiting large number of participants due to low incidence of failure rate. In addition, we have to wait for a long time in order to accrue the required number of events for analysis. Therefore, looking to the data at the earlier time point may help to make decision and stop study early for futility. Interim analysis with sequential probability ratio test (SPRT) may be helpful to include many analyses while controlling for both type I error and type II error. We propose a sequential approach for decision-making to demonstrate Bayesian estimation of VE for the two Poisson incidence rates. We illustrate this method using a simulation study based on Bayesian priors for equal and unequal allocations. Through simulations, we demonstrate the Bayesian form of SPRT and compare it with the Frequentist version.
|
Authors who are presenting talks have a * after their name.