Abstract Details
Activity Number:
|
173
|
Type:
|
Contributed
|
Date/Time:
|
Monday, August 4, 2014 : 10:30 AM to 12:20 PM
|
Sponsor:
|
Biopharmaceutical Section
|
Abstract #313370
|
|
Title:
|
Meta-Analysis of Rare Events: From Combining Confidence Intervals to Combining Confidence Distributions
|
Author(s):
|
Guang Yang*+ and Dungang Liu and Junyuan Wang and Min-ge Xie
|
Companies:
|
Rutgers University and Yale and BMS and Rutgers University
|
Keywords:
|
clinical trial ;
confidence distribution ;
continuity correction for zero events ;
drug safety ;
exact inference ;
rare events
|
Abstract:
|
In recent years, meta-analysis has been recognized as a promising approach to evaluate drug safety, as a single trial usually observe few or even zero adverse events and thus offer little information. However, conventional meta-analysis approaches may lead to invalid inference, since they often rely on large sample theory and require artificial corrections for zero events. To this end, Tian et al. (2009) developed an exact meta-analysis method through combining confidence intervals, and demonstrated its usefulness in the high-profile safety analysis for the diabetes drug Avandia. The goal of this paper is to show that Tian's approach is a special case of the recently developed meta-analysis framework of combining confidence distributions (Xie et al., 2011). We further show that the confidence distribution combining generalize Tian's method in the following three ways: (1) not relying on ad hoc choice of confidence levels but instead integrate the entire distributional information; (2) using variety of transformation functions to combine distributional information; and (3) improving the efficiency by choosing appropriate study-specific weights and the transformation functions.
|
Authors who are presenting talks have a * after their name.
Back to the full JSM 2014 program
|
2014 JSM Online Program Home
For information, contact jsm@amstat.org or phone (888) 231-3473.
If you have questions about the Professional Development program, please contact the Education Department.
The views expressed here are those of the individual authors and not necessarily those of the JSM sponsors, their officers, or their staff.
Copyright © American Statistical Association.