|
Activity Number:
|
522
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Thursday, August 7, 2008 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Biopharmaceutical Section
|
| Abstract - #301638 |
|
Title:
|
Modeling Correlated Binary Safety Data With Partially Exchangeable Models
|
|
Author(s):
|
Xiaodong Wang*+ and Hanxiang Peng
|
|
Companies:
|
sanofi-aventis and University of Mississippi
|
|
Address:
|
200 Crossing Blvd, Bridgewater, NJ, 08807,
|
|
Keywords:
|
Binomial ; Complete monotonicity ; Correlated binary data ; Markov chain ; Partial exchangeability
|
|
Abstract:
|
In many clinical studies, safety endpoints such as adverse events are binary data, which are often correlated within each patient. One of the important issues in the drug safety analysis is the assessment of medication safety in drug development. An appropriate statistical model is critical in estimating the probability of a safety event. In this paper, we propose a procedure to analyze correlated safety binary data using parsimonious partial exchangeable models resulting from rectangular completely monotonic functions.
|
