Application of Discrete False Discovery Rate Controlling Procedures in Clinical Safety Evaluations

• Activity Number: 240 - Topics in Multiplicity and Control of False Discovery Rate
• Type: Contributed
• Date/Time: Monday, July 29, 2019 : 2:00 PM to 3:50 PM
• Sponsor: Biopharmaceutical Section
• Author(s): Li He* and Joe Heyse
• Companies: Merck Research Laboratories and Merck

Abstract

Evaluation of safety is an important part of clinical trials of pharmaceutical and biological products. One important component in the analysis and interpretation of clinical safety data are the multiplicity considerations on the so-called Tier 2 adverse events which represent a large set of adverse events encountered as part of the systematic collection and reporting of safety data. Adverse event associations are typically assessed based on count data using risk differences, risk ratios or odds ratios. In this study, the performances of discrete false discovery rate (FDR) controlling procedures are investigated in the application of clinical safety evaluations. Extensive simulations are performed and a real data application is also included for illustration. One important feature of clinical safety data is that adverse events often have a group structure and can be categorized into various body systems using adverse events dictionaries. Procedures which can incorporate the group structure are also explored.