Activity Number:
|
133
|
Type:
|
Contributed
|
Date/Time:
|
Monday, August 1, 2016 : 8:30 AM to 10:20 AM
|
Sponsor:
|
Biopharmaceutical Section
|
Abstract #319845
|
View Presentation
|
Title:
|
Bayesian Adaptive Design for Trials with Delayed Binary Outcome Using Historical Control Data
|
Author(s):
|
Nusrat Harun* and Mi-Ok Kim and Chunyan Liu
|
Companies:
|
Cincinnati Children's Hospital and Cincinnati Children's Hospital Medical Center and Cincinnati Children's Hospital
|
Keywords:
|
Bayesian adaptive design ;
commensurate priors ;
effective historical sample size ;
delayed outcome
|
Abstract:
|
Adaptive designs assign more patients to one arm based on characteristics of enrolled patients. We propose a Bayesian adaptive design for trials with delayed binary outcomes by utilizing a short-term outcome that is predictive of the primary outcome. We incorporate historical data for the control group as effective historical sample size (EHSS). Randomization ratio is updated as a function of EHSS. More patients are allocated to the new treatment group when the current and historical control group data are commensurate. Commensurate priors centered at the historical parameters are used to specify current parameters. We simulate data from a motivating stroke trial with delayed binary outcome to illustrate the method.
|
Authors who are presenting talks have a * after their name.