|
Activity Number:
|
155
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, August 12, 2002 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Biopharmaceutical Section*
|
| Abstract - #301735 |
|
Title:
|
Using Order Statistics in Clinical Trial Safety Evaluation
|
|
Author(s):
|
A. Gould*+
|
|
Affiliation(s):
|
Merck Research Laboratories
|
|
Address:
|
BL3-2, West Point, Pennsylvania, 19486, USA
|
|
Keywords:
|
adverse events ; correlation
|
|
Abstract:
|
The evaluation of safety in clinical trials differs from the evaluation of efficacy in several important respects. A conclusion about efficacy depends on the result of tests of a few specified hypotheses using specific outcome measures and particular analytic methods, all of which must be specified beforehand. A conclusion about the safety of a treatment may depend on differences between the treatment and control in the incidence of an arbitrary number of adverse events, none of which will have been identified a priori. We describe how simple order statistics for the uniform (0,1) distribution can be used to address the problem posed by the multiple evemts. The method is easy to apply and has reasonable error rate properties even when the events do not occur independently of one another, as might be expected in reality.
|
- 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 2002 program |
