|
Activity Number:
|
215
|
|
Type:
|
Invited
|
|
Date/Time:
|
Tuesday, August 5, 2008 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Biopharmaceutical Section
|
| Abstract - #300278 |
|
Title:
|
Safety Analyses: Thoughts on Tackling Multiplicity
|
|
Author(s):
|
Devan V. Mehrotra*+ and Joseph Heyse
|
|
Companies:
|
Merck Research Laboratories and Merck Research Laboratories
|
|
Address:
|
Mailstop: UG1CD-44, North Wales, PA, 19454,
|
|
Keywords:
|
Adverse experiences ; Clinical Trials ; False discovery rate ; Multiplicity ; Safety analyses
|
|
Abstract:
|
Adverse experience (AE) data in randomized clinical trials are routinely evaluated using between-group p-values for every AE encountered within each of several body systems. If the p-values (or confidence intervals) are interpreted without multiplicity considerations, there is a potential for excessive false positive findings that can needlessly cloud the safety profile of the drug or vaccine. We propose a method for addressing multiplicity that achieves an ICH-recommended balance between type I and type II errors. The method involves a two-step application of the Benjamini and Hochberg false discovery rate (FDR) procedure. Data from three moderate to large vaccine trials are used to illustrate the proposed "Double FDR" approach, and a fourth example serves to reinforce the potential consequences of failing to account for multiplicity. (Joint work with the late Professor John W. Tukey.)
|
