Cargando…
Bayesian bivariate meta-analysis of correlated effects: Impact of the prior distributions on the between-study correlation, borrowing of strength, and joint inferences
Multivariate random-effects meta-analysis allows the joint synthesis of correlated results from multiple studies, for example, for multiple outcomes or multiple treatment groups. In a Bayesian univariate meta-analysis of one endpoint, the importance of specifying a sensible prior distribution for th...
Autores principales: | , , |
---|---|
Formato: | Online Artículo Texto |
Lenguaje: | English |
Publicado: |
SAGE Publications
2016
|
Materias: | |
Acceso en línea: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5810917/ https://www.ncbi.nlm.nih.gov/pubmed/26988929 http://dx.doi.org/10.1177/0962280216631361 |
_version_ | 1783299788816515072 |
---|---|
author | Burke, Danielle L Bujkiewicz, Sylwia Riley, Richard D |
author_facet | Burke, Danielle L Bujkiewicz, Sylwia Riley, Richard D |
author_sort | Burke, Danielle L |
collection | PubMed |
description | Multivariate random-effects meta-analysis allows the joint synthesis of correlated results from multiple studies, for example, for multiple outcomes or multiple treatment groups. In a Bayesian univariate meta-analysis of one endpoint, the importance of specifying a sensible prior distribution for the between-study variance is well understood. However, in multivariate meta-analysis, there is little guidance about the choice of prior distributions for the variances or, crucially, the between-study correlation, ρ(B); for the latter, researchers often use a Uniform(−1,1) distribution assuming it is vague. In this paper, an extensive simulation study and a real illustrative example is used to examine the impact of various (realistically) vague prior distributions for ρ(B) and the between-study variances within a Bayesian bivariate random-effects meta-analysis of two correlated treatment effects. A range of diverse scenarios are considered, including complete and missing data, to examine the impact of the prior distributions on posterior results (for treatment effect and between-study correlation), amount of borrowing of strength, and joint predictive distributions of treatment effectiveness in new studies. Two key recommendations are identified to improve the robustness of multivariate meta-analysis results. First, the routine use of a Uniform(−1,1) prior distribution for ρ(B) should be avoided, if possible, as it is not necessarily vague. Instead, researchers should identify a sensible prior distribution, for example, by restricting values to be positive or negative as indicated by prior knowledge. Second, it remains critical to use sensible (e.g. empirically based) prior distributions for the between-study variances, as an inappropriate choice can adversely impact the posterior distribution for ρ(B), which may then adversely affect inferences such as joint predictive probabilities. These recommendations are especially important with a small number of studies and missing data. |
format | Online Article Text |
id | pubmed-5810917 |
institution | National Center for Biotechnology Information |
language | English |
publishDate | 2016 |
publisher | SAGE Publications |
record_format | MEDLINE/PubMed |
spelling | pubmed-58109172018-02-20 Bayesian bivariate meta-analysis of correlated effects: Impact of the prior distributions on the between-study correlation, borrowing of strength, and joint inferences Burke, Danielle L Bujkiewicz, Sylwia Riley, Richard D Stat Methods Med Res Articles Multivariate random-effects meta-analysis allows the joint synthesis of correlated results from multiple studies, for example, for multiple outcomes or multiple treatment groups. In a Bayesian univariate meta-analysis of one endpoint, the importance of specifying a sensible prior distribution for the between-study variance is well understood. However, in multivariate meta-analysis, there is little guidance about the choice of prior distributions for the variances or, crucially, the between-study correlation, ρ(B); for the latter, researchers often use a Uniform(−1,1) distribution assuming it is vague. In this paper, an extensive simulation study and a real illustrative example is used to examine the impact of various (realistically) vague prior distributions for ρ(B) and the between-study variances within a Bayesian bivariate random-effects meta-analysis of two correlated treatment effects. A range of diverse scenarios are considered, including complete and missing data, to examine the impact of the prior distributions on posterior results (for treatment effect and between-study correlation), amount of borrowing of strength, and joint predictive distributions of treatment effectiveness in new studies. Two key recommendations are identified to improve the robustness of multivariate meta-analysis results. First, the routine use of a Uniform(−1,1) prior distribution for ρ(B) should be avoided, if possible, as it is not necessarily vague. Instead, researchers should identify a sensible prior distribution, for example, by restricting values to be positive or negative as indicated by prior knowledge. Second, it remains critical to use sensible (e.g. empirically based) prior distributions for the between-study variances, as an inappropriate choice can adversely impact the posterior distribution for ρ(B), which may then adversely affect inferences such as joint predictive probabilities. These recommendations are especially important with a small number of studies and missing data. SAGE Publications 2016-03-17 2018-02 /pmc/articles/PMC5810917/ /pubmed/26988929 http://dx.doi.org/10.1177/0962280216631361 Text en © The Author(s) 2016 http://creativecommons.org/licenses/by-nc/3.0/ This article is distributed under the terms of the Creative Commons Attribution-NonCommercial 3.0 License (http://www.creativecommons.org/licenses/by-nc/3.0/) which permits non-commercial use, reproduction and distribution of the work without further permission provided the original work is attributed as specified on the SAGE and Open Access pages (https://us.sagepub.com/en-us/nam/open-access-at-sage). |
spellingShingle | Articles Burke, Danielle L Bujkiewicz, Sylwia Riley, Richard D Bayesian bivariate meta-analysis of correlated effects: Impact of the prior distributions on the between-study correlation, borrowing of strength, and joint inferences |
title | Bayesian bivariate meta-analysis of correlated effects: Impact of the prior distributions on the between-study correlation, borrowing of strength, and joint inferences |
title_full | Bayesian bivariate meta-analysis of correlated effects: Impact of the prior distributions on the between-study correlation, borrowing of strength, and joint inferences |
title_fullStr | Bayesian bivariate meta-analysis of correlated effects: Impact of the prior distributions on the between-study correlation, borrowing of strength, and joint inferences |
title_full_unstemmed | Bayesian bivariate meta-analysis of correlated effects: Impact of the prior distributions on the between-study correlation, borrowing of strength, and joint inferences |
title_short | Bayesian bivariate meta-analysis of correlated effects: Impact of the prior distributions on the between-study correlation, borrowing of strength, and joint inferences |
title_sort | bayesian bivariate meta-analysis of correlated effects: impact of the prior distributions on the between-study correlation, borrowing of strength, and joint inferences |
topic | Articles |
url | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5810917/ https://www.ncbi.nlm.nih.gov/pubmed/26988929 http://dx.doi.org/10.1177/0962280216631361 |
work_keys_str_mv | AT burkedaniellel bayesianbivariatemetaanalysisofcorrelatedeffectsimpactofthepriordistributionsonthebetweenstudycorrelationborrowingofstrengthandjointinferences AT bujkiewiczsylwia bayesianbivariatemetaanalysisofcorrelatedeffectsimpactofthepriordistributionsonthebetweenstudycorrelationborrowingofstrengthandjointinferences AT rileyrichardd bayesianbivariatemetaanalysisofcorrelatedeffectsimpactofthepriordistributionsonthebetweenstudycorrelationborrowingofstrengthandjointinferences |