A simple method for inference on an overall effect in meta-analysis

The random effects approach in meta‐analysis due to DerSimonian and Laird is well established and used pervasively. It has been established by Brockwell and Gordon that this method, when used for confidence intervals, leads to coverage probabilities lower than the nominal value. A number of alternat...

Full description

Saved in:
Bibliographic Details
Published in:Statistics in medicine 2007-11, Vol.26 (25), p.4531-4543
Main Authors: Brockwell, Sarah E., Gordon, Ian R.
Format: Article
Language:English
Subjects:
Computer Simulation
Confidence Intervals
coverage properties
Data Interpretation, Statistical
Humans
Magnesium - administration & dosage
Magnesium - therapeutic use
Medical research
Meta-analysis
Meta-Analysis as Topic
Myocardial Infarction - drug therapy
Myocardial Infarction - mortality
Probability
random effects
Statistics
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The random effects approach in meta‐analysis due to DerSimonian and Laird is well established and used pervasively. It has been established by Brockwell and Gordon that this method, when used for confidence intervals, leads to coverage probabilities lower than the nominal value. A number of alternatives have been proposed, but these either have the defect of iterative and complicated calculation, or deficient coverage. In this paper we propose a new approach, which is simple to use, and has coverage probabilities better than the alternatives, based on extensive simulation. We call this approach the ‘quantile approximation’ method. Copyright © 2007 John Wiley & Sons, Ltd.
ISSN:0277-6715
1097-0258
DOI:10.1002/sim.2883