|
Activity Number:
|
340
- SPEED: SPAAC SESSION III
|
|
Type:
|
Topic-Contributed
|
|
Date/Time:
|
Thursday, August 12, 2021 : 10:00 AM to 11:50 AM
|
|
Sponsor:
|
Section on Statistics in Epidemiology
|
|
Abstract #318088
|
|
|
Title:
|
Exact Inference for Fixed-Effects Meta-Analysis of Proportions
|
|
Author(s):
|
Spencer Lynn Hansen* and Ken Rice
|
|
Companies:
|
Department of Biostatistics, University of Washington and Department of Biostatistics, University of Washington
|
|
Keywords:
|
meta-analysis;
fixed effects;
proportion;
binomial;
exact methods
|
|
Abstract:
|
In meta-analysis of proportions, counts of independent binary successes/failures are aggregated over multiple studies. The method is widely used, for example in adverse event monitoring. When the underlying proportions are homogeneous, the total has a binomial distribution and exact inference is straightforward. While heterogeneous proportions (a “fixed-effects model”) is more realistic, no exact method is yet available and the various known approximate methods may disagree substantially. To address these issues, we present a simple fixed-effects method giving tests and confidence intervals for meta-analysis of proportions that are exact under heterogeneity. They can be implemented in standard software and the underlying parameter of interest is readily interpretable. Data from a recent kidney disease meta-analysis shows how the method’s performance can be assessed in practice.
|
Authors who are presenting talks have a * after their name.
Back to the full JSM 2021 program
|
