Online Program Home
My Program

Abstract Details

Activity Number: 525 - Clinical Trial Design- 6
Type: Contributed
Date/Time: Wednesday, August 1, 2018 : 10:30 AM to 12:20 PM
Sponsor: Biopharmaceutical Section
Abstract #328675 Presentation
Title: Estimating Probability of Simultaneous Success with Multiple Endpoints Using Truncated Multi-Variate Correlated Normal Distribution
Author(s): Tianshuang Wu* and Yihua Gu and Ziqian Geng and Saurabh Mukhopadhyay
Companies: AbbVie and AbbVie and AbbVie and AbbVie
Keywords: Sample Size Estimation; Clinical Trial Design; Probability of Success; Multiplicity

Assessments of power and Probability of Success (POS) are the cornerstones in sample size determination of all clinical trials. The methods to estimate the power and POS for a single endpoint have been well established; however, many clinical trials are designed to achieve the success in multiple endpoints simultaneously. The power and POS calculations under the assumption of independence of the endpoints are often used as a simple and conservative proxy. The loss of efficiency with this approach increases as the number of endpoints increases given that the efficacy endpoints for a drug effect are often positively correlated. We propose a joint modeling structure based on a truncated multivariate normal to implement the correlation among the endpoints. This will improve accuracy of estimating probability of simultaneous success (POSS) and thus in turn will improve the efficiency of the trial design.

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

Back to the full JSM 2018 program