Online Program Home
My Program

Abstract Details

Activity Number: 20 - Statistical Considerations for Using Historical Controls in Clinical Trials
Type: Topic Contributed
Date/Time: Sunday, July 29, 2018 : 2:00 PM to 3:50 PM
Sponsor: Biopharmaceutical Section
Abstract #329683
Title: Designing Trials Using Bayesian Methods with Historical Controls
Author(s): Michael Sonksen*
Companies: Eli Lilly and Company
Keywords: Mixture Priors; Power Priors; Clinical Trials

Bayesian methodology is a coherent process for building prior distributions based on historical data. Using this data can increase the power of clinical trials while keeping sample size the same. However, there are a number of approaches to do this with no clear "best" way to incorporate the prior knowledge. For example, simple pooling may follow directly from Bayes Theorem but can lead to upward bias when applied to clinical trials. In this talk we will review a number of useful approaches of using historical controls with example illustrations. In particular, we will examine hierarchical models, mixture priors, and variants of power priors. Practical considerations as well as impact on type I and type II errors will be discussed.

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

Back to the full JSM 2018 program