|
Activity Number:
|
26
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Sunday, August 2, 2009 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Biopharmaceutical Section
|
| Abstract - #305227 |
|
Title:
|
Development of a Statistical Methodology to Identify Appropriate Historical Controls for Rare Disease Therapies Using Information Generated from the Peer-Reviewed Scientific Literature
|
|
Author(s):
|
Jessica (Jeongsook) L. Kim*+ and Jean (Chinying) Wang and Charles Maplethorpe
|
|
Companies:
|
FDA and FDA and FDA
|
|
Address:
|
1401 Rockville Pike, Rockville, MD, 20852,
|
|
Keywords:
|
meta-analysis ; historical control ; Bayesian method ; a prior information
|
|
Abstract:
|
In designing studies such as either a placebo control or a historical control has been used for a comparator in generating the study hypothesis, finding a relevant historical control may be a challenge due to innovations in medical treatments or changes in medical practice over time, and due to differences in the specific target patient populations. Current applications of meta-analysis in estimating historical control outcomes (e.g., Hepatitis B Immune Globulin Intravenous (Human) products for liver transplantation outcomes) may not adequately address these issues. Therefore, an improved methodology for identifying appropriate historical controls may play a crucial role in determining a product's efficacy. An alternative Bayesian method will be applied to identify appropriate historical controls utilizing information from literature reviews as a prior information.
|
