|
Activity Number:
|
119
|
|
Type:
|
Invited
|
|
Date/Time:
|
Monday, July 30, 2007 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Biopharmaceutical Section
|
| Abstract - #307870 |
|
Title:
|
A Triage Approach to Evaluating Safety Incorporating Frequentist and Bayesian Methods
|
|
Author(s):
|
A. Lawrence Gould*+
|
|
Companies:
|
Merck & Co., Inc.
|
|
Address:
|
UG1D-88, West Point, PA, 19486,
|
|
Keywords:
|
Adverse Event ; Multiplicity ; Order Statistics ; Planning
|
|
Abstract:
|
The analysis of safety data is as important as the analysis of efficacy data. The most appropriate analysis strategy for an adverse event depends on whether it was identified a priori, as not identified a priori but not 'rare', and as not identified a priori and 'rare.' Multiplicity affects the interpretation of analyses because many adverse events often will be reported. Hypothesis testing will be of limited help for adverse events not identified a priori because the many hypotheses are generated by the data. Quantifying the degree of risk will be more helpful, especially using Bayesian methods. This presentation describes some general considerations in planning for safety evaluation, presents some ways to summarize data using confidence or credible intervals, describes a Bayesian approach to interpreting the outcomes, and suggests a simple graphical way to address multiplicity.
|
- The address information is for the authors that have a + after their name.
- Authors who are presenting talks have a * after their name.
Back to the full JSM 2007 program |
