JSM 2012 Home

JSM 2012 Online Program

The views expressed here are those of the individual authors and not necessarily those of the JSM sponsors, their officers, or their staff.

Online Program Home

Abstract Details

Activity Number: 513
Type: Contributed
Date/Time: Wednesday, August 1, 2012 : 10:30 AM to 12:20 PM
Sponsor: Biopharmaceutical Section
Abstract - #306785
Title: The Case for Bayesian Predictive Probabilities for Interim Monitoring of Clinical Trials
Author(s): Benjamin Saville*+ and Jason Connor and Gregory Ayers and JoAnn Alvarez
Companies: Vanderbilt University School of Medicine and Berry Consultants and Vanderbilt University School of Medicine and Vanderbilt University School of Medicine
Address: 1161 21st Avenue South, Nashville, TN, 37232-2158, United States
Keywords: predictive probability ; posterior probability ; interim monitoring ; Bayesian ; adaptive clinical trials
Abstract:

We explore the advantages of a fully Bayesian predictive probability approach versus a traditional Bayesian approach using posterior probabilities for the interim monitoring of clinical trials. Compared to posterior probabilities, predictive probabilities dramatically increase computing time required to design trials with known operating characteristics via simulations. However, some argue that the cost of the computational burden is overwhelmed by the benefits obtained from using predictive probabilities. In this manuscript, we provide evidence that the predictive probability approach is more closely aligned with the clinical decision process than the posterior probability approach. It is more straightforward to derive stopping rules based on predictive probabilities and we show examples of poorly chosen rules based on posterior probabilities. We explore the relationship between predictive probabilities and posterior probabilities as a function of the amount of remaining data to collect in the trial, and argue that the predictive probability approach is a superior strategy for clinical trial design.


The address information is for the authors that have a + after their name.
Authors who are presenting talks have a * after their name.

Back to the full JSM 2012 program




2012 JSM Online Program Home

For information, contact jsm@amstat.org or phone (888) 231-3473.

If you have questions about the Continuing Education program, please contact the Education Department.