|
Activity Number:
|
109
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, July 30, 2007 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Bayesian Statistical Science
|
| Abstract - #308763 |
|
Title:
|
Application of Bayesian Analysis in Review of Adverse Event for Product Quality Issues
|
|
Author(s):
|
Harry Yang*+ and Lanju Zhang and Iksung Cho
|
|
Companies:
|
MedImmune, Inc. and MedImmune, Inc. and MedImmune, Inc.
|
|
Address:
|
One MedImmune Way, Gaithersburg, MD, 20878,
|
|
Keywords:
|
Bayesian analysis ; Adverse event ; Prior ; Monte Carlo simulation
|
|
Abstract:
|
A pharmacovigilance system capable of identifying product lots with quality issues through analyzing post-marketing safety data is very important. Although some researchers have proposed safety data review procedures to signal potentially problematic lots, few of the proposals were based on sound statistical methodologies. We present a Bayesian and frequentist approaches to detecting out-of-trend AEs. The number of AEs is described through a mixture model, taking on binomial and Poisson distributions, with a probability of p and 1 - p, respectively. Assuming parameters p and l (Poisson mean) have Beta and Gamma priors, respectively, the probability for the number of AEs of the current lot under evaluation to exceed its observed value is estimated. This estimate allows us to determine if the product lot associated with the AEs is of quality concerns. A method based on MLE is proposed.
|
- 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 |
