Cargando…

A comparison of seven random‐effects models for meta‐analyses that estimate the summary odds ratio

Comparative trials that report binary outcome data are commonly pooled in systematic reviews and meta‐analyses. This type of data can be presented as a series of 2‐by‐2 tables. The pooled odds ratio is often presented as the outcome of primary interest in the resulting meta‐analysis. We examine the...

Descripción completa

Detalles Bibliográficos
Autores principales: Jackson, Dan, Law, Martin, Stijnen, Theo, Viechtbauer, Wolfgang, White, Ian R.
Formato: Online Artículo Texto
Lenguaje:English
Publicado: John Wiley and Sons Inc. 2018
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5841569/
https://www.ncbi.nlm.nih.gov/pubmed/29315733
http://dx.doi.org/10.1002/sim.7588
_version_ 1783304771429466112
author Jackson, Dan
Law, Martin
Stijnen, Theo
Viechtbauer, Wolfgang
White, Ian R.
author_facet Jackson, Dan
Law, Martin
Stijnen, Theo
Viechtbauer, Wolfgang
White, Ian R.
author_sort Jackson, Dan
collection PubMed
description Comparative trials that report binary outcome data are commonly pooled in systematic reviews and meta‐analyses. This type of data can be presented as a series of 2‐by‐2 tables. The pooled odds ratio is often presented as the outcome of primary interest in the resulting meta‐analysis. We examine the use of 7 models for random‐effects meta‐analyses that have been proposed for this purpose. The first of these models is the conventional one that uses normal within‐study approximations and a 2‐stage approach. The other models are generalised linear mixed models that perform the analysis in 1 stage and have the potential to provide more accurate inference. We explore the implications of using these 7 models in the context of a Cochrane Review, and we also perform a simulation study. We conclude that generalised linear mixed models can result in better statistical inference than the conventional 2‐stage approach but also that this type of model presents issues and difficulties. These challenges include more demanding numerical methods and determining the best way to model study specific baseline risks. One possible approach for analysts is to specify a primary model prior to performing the systematic review but also to present the results using other models in a sensitivity analysis. Only one of the models that we investigate is found to perform poorly so that any of the other models could be considered for either the primary or the sensitivity analysis.
format Online
Article
Text
id pubmed-5841569
institution National Center for Biotechnology Information
language English
publishDate 2018
publisher John Wiley and Sons Inc.
record_format MEDLINE/PubMed
spelling pubmed-58415692018-03-30 A comparison of seven random‐effects models for meta‐analyses that estimate the summary odds ratio Jackson, Dan Law, Martin Stijnen, Theo Viechtbauer, Wolfgang White, Ian R. Stat Med Research Articles Comparative trials that report binary outcome data are commonly pooled in systematic reviews and meta‐analyses. This type of data can be presented as a series of 2‐by‐2 tables. The pooled odds ratio is often presented as the outcome of primary interest in the resulting meta‐analysis. We examine the use of 7 models for random‐effects meta‐analyses that have been proposed for this purpose. The first of these models is the conventional one that uses normal within‐study approximations and a 2‐stage approach. The other models are generalised linear mixed models that perform the analysis in 1 stage and have the potential to provide more accurate inference. We explore the implications of using these 7 models in the context of a Cochrane Review, and we also perform a simulation study. We conclude that generalised linear mixed models can result in better statistical inference than the conventional 2‐stage approach but also that this type of model presents issues and difficulties. These challenges include more demanding numerical methods and determining the best way to model study specific baseline risks. One possible approach for analysts is to specify a primary model prior to performing the systematic review but also to present the results using other models in a sensitivity analysis. Only one of the models that we investigate is found to perform poorly so that any of the other models could be considered for either the primary or the sensitivity analysis. John Wiley and Sons Inc. 2018-01-08 2018-03-30 /pmc/articles/PMC5841569/ /pubmed/29315733 http://dx.doi.org/10.1002/sim.7588 Text en © 2018 The Authors. Statistics in Medicine Published by John Wiley & Sons Ltd. This is an open access article under the terms of the Creative Commons Attribution (http://creativecommons.org/licenses/by/4.0/) License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.
spellingShingle Research Articles
Jackson, Dan
Law, Martin
Stijnen, Theo
Viechtbauer, Wolfgang
White, Ian R.
A comparison of seven random‐effects models for meta‐analyses that estimate the summary odds ratio
title A comparison of seven random‐effects models for meta‐analyses that estimate the summary odds ratio
title_full A comparison of seven random‐effects models for meta‐analyses that estimate the summary odds ratio
title_fullStr A comparison of seven random‐effects models for meta‐analyses that estimate the summary odds ratio
title_full_unstemmed A comparison of seven random‐effects models for meta‐analyses that estimate the summary odds ratio
title_short A comparison of seven random‐effects models for meta‐analyses that estimate the summary odds ratio
title_sort comparison of seven random‐effects models for meta‐analyses that estimate the summary odds ratio
topic Research Articles
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5841569/
https://www.ncbi.nlm.nih.gov/pubmed/29315733
http://dx.doi.org/10.1002/sim.7588
work_keys_str_mv AT jacksondan acomparisonofsevenrandomeffectsmodelsformetaanalysesthatestimatethesummaryoddsratio
AT lawmartin acomparisonofsevenrandomeffectsmodelsformetaanalysesthatestimatethesummaryoddsratio
AT stijnentheo acomparisonofsevenrandomeffectsmodelsformetaanalysesthatestimatethesummaryoddsratio
AT viechtbauerwolfgang acomparisonofsevenrandomeffectsmodelsformetaanalysesthatestimatethesummaryoddsratio
AT whiteianr acomparisonofsevenrandomeffectsmodelsformetaanalysesthatestimatethesummaryoddsratio
AT jacksondan comparisonofsevenrandomeffectsmodelsformetaanalysesthatestimatethesummaryoddsratio
AT lawmartin comparisonofsevenrandomeffectsmodelsformetaanalysesthatestimatethesummaryoddsratio
AT stijnentheo comparisonofsevenrandomeffectsmodelsformetaanalysesthatestimatethesummaryoddsratio
AT viechtbauerwolfgang comparisonofsevenrandomeffectsmodelsformetaanalysesthatestimatethesummaryoddsratio
AT whiteianr comparisonofsevenrandomeffectsmodelsformetaanalysesthatestimatethesummaryoddsratio