Online Program Home
  My Program

All Times EDT

Abstract Details

Activity Number: 489 - Multiplicity: Methods and Applications
Type: Contributed
Date/Time: Thursday, August 6, 2020 : 10:00 AM to 2:00 PM
Sponsor: Biopharmaceutical Section
Abstract #313340
Title: Assessing the Incidence and Severity of Drug Adverse Events: A Bayesian Hierarchical Cumulative Logit Model
Author(s): Jiawei Duan* and Byron J Gajewski and Jo A. Wick
Companies: Novartis Pharmaceuticals Corporation and KUMC and University of Kansas Medical Center
Keywords: Drug safety; Bayesian mixture model; Multiplicity; Safety signal detection
Abstract:

Detection of safety signals from many types of adverse events (AEs) that are reported in a two-arm clinical trial involves difficult multiplicity problems. A Bayesian hierarchical mixture model proposed by Berry and Berry in 2004 is a good solution to this problem as it borrows information across subgroups and moderates extremes due merely to chance. However, it compares only AE incidence rates, regardless of AE severity. In this article, we propose a three-level Bayesian hierarchical non-proportional odds version of the cumulative logit model. Our model allows for testing the equality of incidence rate and severity for all the AEs simultaneously in a two arm randomized clinical trial while addressing multiplicities. We conduct simulation study to investigate the operating characteristics of the proposed hierarchical model. The simulation results show that the proposed method not only controls for false discovery rate but also performs well in detecting safety signals with respect to AE incidence rate or severity. The proposed method is demonstrated via a simulated dataset from a vaccine trial.


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

Back to the full JSM 2020 program