Title: Bayesian multivariate probability of success using historical data with family-wise error rate control
Presenters: Matt Psioda and Ethan Alt, Department of Biostatistics, University of North Carolina
Abstract: Given the cost and duration of phase III and phase IV clinical trials, the development of statistical methods for go/no-go decisions is vital. In this paper, we introduce a Bayesian methodology to compute the probability of success based on the current data of a treatment regimen for the multivariate linear model. Our approach utilizes a Bayesian seemingly unrelated regression model, which allows for multiple endpoints to be modeled jointly even if the covariates between the endpoints are different. Correlations between outcomes are explicitly modeled. This Bayesian joint modeling approach unifies single and multiple testing procedures under a single framework. We develop an approach to multiple testing that asymptotically guarantees exact family-wise error rate control and is more powerful than frequentist approaches to multiplicity. The method effectively yields those of Ibrahim et al. and Chuang-Stein as special cases, and, to our knowledge, is the only method that allows for robust sample size determination for multiple outcomes.
Matt Psioda is an Assistant Professor in the Department of Biostatistics at the University of North Carolina at Chapel Hill and Associate Director of Clinical Trials Research at the Collaborative Studies Coordinating Center. He currently serves as co-investigator for the Data Integration, Algorithm Development and Operations Management Center (DAC) within the NIH Back Pain Consortium (BACPAC), as well as Executive Committee Chair for the Consortium. He collaborates on or oversees biostatistical activities for a number of other clinical trials, including pragmatic comparative effectiveness trials, cluster randomized trials, Bayesian adaptive trials in oncology and pediatric disease settings, and SMART trials. He is a statistical advisor for the Center for Drug Evaluation and Research at the United States Food and Drug Administration (FDA). He also teaches graduate courses in statistical computing and longitudinal data analysis methods. His primary methodological research focus is the design of innovative clinical trials using Bayesian methods but he is also interested in methodological research related to Bayesian computation and estimation of heterogeneous treatment effects.
Ethan Alt is a PhD candidate in the Department of Biostatistics at the University of North Carolina. His dissertation is concerned with developing Bayesian methods for the design and analysis of clinical trials. He is supervised by Drs. Matthew Psioda and Joseph Ibrahim. He holds a bachelor’s degree in economics from Loyola University Chicago and a master’s degree in statistics from the University of Florida.
Join Zoom Meeting
Meeting ID: 220 359 9026
One tap mobile
+16468769923,,2203599026# US (New York)
+16699006833,,2203599026# US (San Jose)
Dial by your location
+1 646 876 9923 US (New York)
+1 669 900 6833 US (San Jose)
+1 253 215 8782 US (Tacoma)
+1 301 715 8592 US (Germantown)
+1 312 626 6799 US (Chicago)
+1 346 248 7799 US (Houston)
855 880 1246 US Toll-free
877 853 5257 US Toll-free
Meeting ID: 220 359 9026
Find your local number: https://diaglobal.zoom.us/u/tdzNsDvQ
Join by Skype for Business