|
Activity Number:
|
291
|
|
Type:
|
Invited
|
|
Date/Time:
|
Tuesday, August 5, 2008 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Statistics in Epidemiology
|
| Abstract - #300137 |
|
Title:
|
Vaccine Efficacy Among the Doomed When Monotonicity Is Unreasonable
|
|
Author(s):
|
Bryan E. Shepherd*+
|
|
Companies:
|
Vanderbilt University
|
|
Address:
|
1161 21st Avenue South, S2323 MCN, Nashville, TN, 37232,
|
|
Keywords:
|
principal stratification ; HIV ; causal effects
|
|
Abstract:
|
In vaccine efficacy trials, researchers often want to know the effect of vaccination on outcomes that are only well-defined in those who become infected. However, an analysis comparing post-infection outcomes between those infected in the vaccine and placebo arms could be biased because infection is not randomized. To assess the efficacy of a vaccine on a post-infection outcome, it may be better to focus on the principal stratum of individuals who were doomed to be infected regardless of treatment assignment. To identify who belongs to this stratum, monotonicity is often assumed, that there is no person who would be infected if randomized to vaccine but not if randomized to placebo. In light of recent results from the STEP HIV vaccine trial, such an assumption may be unreasonable. Here I discuss sensitivity analyses for relaxing monotonicity.
|
- 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 2008 program |
