Predictive distributions for between‐study heterogeneity and simple methods for their application in Bayesian meta‐analysis

Rebecca M. Turner; Dan Jackson; Yinghui Wei; Simon G. Thompson; Julian P. T. Higgins

doi:10.1002/sim.6381

journal article Open Access Dec 05, 2014

Predictive distributions for between‐study heterogeneity and simple methods for their application in Bayesian meta‐analysis

Rebecca M. Turner

Dan Jackson

Yinghui Wei

Simon G. Thompson

Julian P. T. Higgins

Statistics in Medicine Vol. 34 No. 6 pp. 984-998 · Wiley

View at Publisher Save 10.1002/sim.6381

Abstract

Numerous meta‐analyses in healthcare research combine results from only a small number of studies, for which the variance representing between‐study heterogeneity is estimated imprecisely. A Bayesian approach to estimation allows external evidence on the expected magnitude of heterogeneity to be incorporated.The aim of this paper is to provide tools that improve the accessibility of Bayesian meta‐analysis. We present two methods for implementing Bayesian meta‐analysis, using numerical integration and importance sampling techniques. Based on 14 886 binary outcome meta‐analyses in the Cochrane Database of Systematic Reviews, we derive a novel set of predictive distributions for the degree of heterogeneity expected in 80 settings depending on the outcomes assessed and comparisons made. These can be used as prior distributions for heterogeneity in future meta‐analyses.The two methods are implemented in R, for which code is provided. Both methods produce equivalent results to standard but more complex Markov chain Monte Carlo approaches. The priors are derived as log‐normal distributions for the between‐study variance, applicable to meta‐analyses of binary outcomes on the log odds‐ratio scale. The methods are applied to two example meta‐analyses, incorporating the relevant predictive distributions as prior distributions for between‐study heterogeneity.We have provided resources to facilitate Bayesian meta‐analysis, in a form accessible to applied researchers, which allow relevant prior information on the degree of heterogeneity to be incorporated. © 2014 The Authors. Statistics in Medicine published by John Wiley & Sons Ltd.

Topics

No keywords indexed for this article. Browse by subject →

References

23

[1]

10.1186/1471-2288-11-160

[2]

How does the DerSimonian and Laird procedure for random effects meta-analysis compare with its more efficient but harder to compute counterparts?

Dan Jackson, Jack Bowden, Rose Baker

Journal of Statistical Planning and Inference 10.1016/j.jspi.2009.09.017

[3]

10.1002/(sici)1097-0258(19961230)15:24<2733::aid-sim562>3.0.co;2-0

[4]

10.1002/sim.4780142408

[5]

A Re-Evaluation of Random-Effects Meta-Analysis

Julian P. T. Higgins, Simon G. Thompson, David J. Spiegelhalter

Journal of the Royal Statistical Society Series A:... 10.1111/j.1467-985x.2008.00552.x

[6]

10.1177/0272989x05282643

[7]

10.1093/ije/dys041

[8]

10.1023/a:1008929526011

[9]

R Core Team (2013)

[10]

Meta-analysis in clinical trials

Rebecca DerSimonian, Nan Laird

Controlled Clinical Trials 10.1016/0197-2456(86)90046-2

[11]

10.2307/1390675

[12]

10.1002/9780470316726

[13]

10.2307/1913710

[14]

10.1002/sim.4326

[15]

Bayesian Measures of Model Complexity and Fit

David J. Spiegelhalter, Nicola G. Best, Bradley P. Carlin et al.

Journal of the Royal Statistical Society Series B:... 10.1111/1467-9868.00353

[16]

Cosmi B "Ticlopidine versus oral anticoagulation for coronary stenting" Cochrane Database of Systematic Reviews

[17]

Viechtbauer W "Confidence intervals for the amount of heterogeneity in meta‐analysis" Statistics in Medicine 10.1002/sim.2514

[18]

Gates S "Auricular acupuncture for cocaine dependence" Cochrane Database of Systematic Reviews

[19]

10.1016/j.jclinepi.2007.03.016

[20]

Quantifying heterogeneity in a meta‐analysis

Julian P. T. Higgins, Simon G. Thompson

Statistics in Medicine 10.1002/sim.1186

[21]

10.1093/biostatistics/kxp032

[22]

10.1002/sim.5745

[23]

O'Hagan A (2004)

Metrics

297

Citations

23

References

Details

Published: Dec 05, 2014
Vol/Issue: 34(6)
Pages: 984-998
License: View

Authors

R

Rebecca M. Turner

MRC Biostatistics Unit Cambridge Institute of Public Health Cambridge U.K.

D

Dan Jackson

MRC Biostatistics Unit Cambridge Institute of Public Health Cambridge U.K.

Y

Yinghui Wei

School of Computing and Mathematics University of Plymouth Plymouth U.K.

S

Simon G. Thompson

Department of Public Health and Primary Care University of Cambridge Cambridge U.K.

J

Julian P. T. Higgins

School of Social and Community Medicine University of Bristol Bristol U.K.; Centre for Reviews and Dissemination University of York York U.K.

Funding

Medical Research Council Award: U105260558

Cite This Article

Rebecca M. Turner, Dan Jackson, Yinghui Wei, et al. (2014). Predictive distributions for between‐study heterogeneity and simple methods for their application in Bayesian meta‐analysis. Statistics in Medicine, 34(6), 984-998. https://doi.org/10.1002/sim.6381

Predictive distributions for between‐study heterogeneity and simple methods for their application in Bayesian meta‐analysis

You May Also Like