|
Activity Number:
|
155
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, August 12, 2002 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Biopharmaceutical Section*
|
| Abstract - #301928 |
|
Title:
|
Problems in Combining Studies for Safety Analyses
|
|
Author(s):
|
Qian Li*+
|
|
Affiliation(s):
|
Food and Drug Administration
|
|
Address:
|
9201 Corporate Blvd, HFD-725, Rockville, MD, 20850, USA
|
|
Keywords:
|
Combining Studies ; Meta-Analysis ; Risks ; Risk Ratios ; Safety Analyses ; Serious Adverse Events
|
|
Abstract:
|
In analyzing serious adverse events in NDA submissions, often studies from phase II to IV are combined together to perform a meta-analysis in a hope to better quantify the adverse events and achieve better power to detect differences between treatment groups. In combining studies, often we see that different dose levels of a studied drug are combined together; so are different comparators, and different patient populations. When studies of different duration are combined, survival analysis will not help if risks of different dose levels were different and risk ratios between treatment groups were not constant over time. Such combining my cause difficulties in interpreting study results, especially for the analyses of serious adverse events. A couple of examples from NDA submissions are used to illustrate the problems.
|
- 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 |
