Empirical comparison of univariate and multivariate meta-analyses in Cochrane Pregnancy and Childbirth reviews with multiple binary outcomes.


Journal

Research synthesis methods
ISSN: 1759-2887
Titre abrégé: Res Synth Methods
Pays: England
ID NLM: 101543738

Informations de publication

Date de publication:
Sep 2019
Historique:
received: 30 07 2018
revised: 04 04 2019
accepted: 13 04 2019
pubmed: 7 5 2019
medline: 29 4 2020
entrez: 7 5 2019
Statut: ppublish

Résumé

Multivariate meta-analysis (MVMA) jointly synthesizes effects for multiple correlated outcomes. The MVMA model is potentially more difficult and time-consuming to apply than univariate models, so if its use makes little difference to parameter estimates, it could be argued that it is redundant. We assessed the applicability and impact of MVMA in Cochrane Pregnancy and Childbirth (CPCB) systematic reviews. We applied MVMA to CPCB reviews published between 2011 and 2013 with two or more binary outcomes with at least three studies and compared findings with results of univariate meta-analyses. Univariate random effects meta-analysis models were fitted using restricted maximum likelihood estimation (REML). Eighty CPCB reviews were published. MVMA could not be applied in 70 of these reviews. MVMA was not feasible in three of the remaining 10 reviews because the appropriate models failed to converge. Estimates from MVMA agreed with those of univariate analyses in most of the other seven reviews. Statistical significance changed in two reviews: In one, this was due to a very small change in P value; in the other, the MVMA result for one outcome suggested that previous univariate results may be vulnerable to small-study effects and that the certainty of clinical conclusions needs consideration. MVMA methods can be applied only in a minority of reviews of interventions in pregnancy and childbirth and can be difficult to apply because of missing correlations or lack of convergence. Nevertheless, clinical and/or statistical conclusions from MVMA may occasionally differ from those from univariate analyses.

Sections du résumé

BACKGROUND BACKGROUND
Multivariate meta-analysis (MVMA) jointly synthesizes effects for multiple correlated outcomes. The MVMA model is potentially more difficult and time-consuming to apply than univariate models, so if its use makes little difference to parameter estimates, it could be argued that it is redundant.
METHODS METHODS
We assessed the applicability and impact of MVMA in Cochrane Pregnancy and Childbirth (CPCB) systematic reviews. We applied MVMA to CPCB reviews published between 2011 and 2013 with two or more binary outcomes with at least three studies and compared findings with results of univariate meta-analyses. Univariate random effects meta-analysis models were fitted using restricted maximum likelihood estimation (REML).
RESULTS RESULTS
Eighty CPCB reviews were published. MVMA could not be applied in 70 of these reviews. MVMA was not feasible in three of the remaining 10 reviews because the appropriate models failed to converge. Estimates from MVMA agreed with those of univariate analyses in most of the other seven reviews. Statistical significance changed in two reviews: In one, this was due to a very small change in P value; in the other, the MVMA result for one outcome suggested that previous univariate results may be vulnerable to small-study effects and that the certainty of clinical conclusions needs consideration.
CONCLUSIONS CONCLUSIONS
MVMA methods can be applied only in a minority of reviews of interventions in pregnancy and childbirth and can be difficult to apply because of missing correlations or lack of convergence. Nevertheless, clinical and/or statistical conclusions from MVMA may occasionally differ from those from univariate analyses.

Identifiants

pubmed: 31058440
doi: 10.1002/jrsm.1353
pmc: PMC6771837
doi:

Substances chimiques

Serotonin Antagonists 0

Types de publication

Comparative Study Journal Article Meta-Analysis

Langues

eng

Sous-ensembles de citation

IM

Pagination

440-451

Subventions

Organisme : Medical Research Council
ID : MC_UU_12023/21
Pays : United Kingdom
Organisme : Medical Research Council
ID : MR/J013595/1
Pays : United Kingdom
Organisme : NIHR CLAHRC West Midlands+

Informations de copyright

© 2019 The Authors Research Synthesis Methods Published by John Wiley & Sons Ltd.

Références

Control Clin Trials. 1986 Sep;7(3):177-88
pubmed: 3802833
Stat Med. 2005 Jul 30;24(14):2241-54
pubmed: 15887296
J R Stat Soc Ser C Appl Stat. 2018 Nov;67(5):1177-1205
pubmed: 30344346
BMJ. 2011 Feb 10;342:d549
pubmed: 21310794
Stat Med. 2007 Jan 15;26(1):78-97
pubmed: 16526010
Stat Med. 2002 Feb 28;21(4):589-624
pubmed: 11836738
Stat Med. 2011 Sep 10;30(20):2481-98
pubmed: 21268052
Biometrics. 2018 Jun;74(2):548-556
pubmed: 28806485
Cochrane Database Syst Rev. 2012 Apr 18;(4):CD008991
pubmed: 22513970
Biostatistics. 2008 Jan;9(1):172-86
pubmed: 17626226
Stat Med. 2016 Dec 20;35(29):5495-5511
pubmed: 27577523
Res Synth Methods. 2015 Jun;6(2):157-74
pubmed: 26099484
J Clin Epidemiol. 2015 May;68(5):542-50
pubmed: 25537265
Cochrane Database Syst Rev. 2012 Nov 14;11:CD009356
pubmed: 23152275
Res Synth Methods. 2019 Sep;10(3):440-451
pubmed: 31058440
Cochrane Database Syst Rev. 2012 Sep 12;(9):CD007579
pubmed: 22972112
Stat Methods Med Res. 2017 Dec;26(6):2853-2868
pubmed: 26546254
Cochrane Database Syst Rev. 2012 Jan 18;1:CD009107
pubmed: 22258999
Stat Med. 1998 Nov 30;17(22):2537-50
pubmed: 9839346
J R Stat Soc Ser A Stat Soc. 2009 Jan;172(1):137-159
pubmed: 19381330
Stat Med. 2013 Mar 30;32(7):1191-205
pubmed: 23208849
Stat Med. 2012 Sep 10;31(20):2179-95
pubmed: 22532016
Clin Trials. 2012 Oct;9(5):610-20
pubmed: 22872546
Cochrane Database Syst Rev. 2012 Feb 15;(2):CD005122
pubmed: 22336808
Stat Med. 2008 Feb 28;27(5):670-86
pubmed: 17492826
J Clin Epidemiol. 2011 Jun;64(6):608-18
pubmed: 21109399
Stat Med. 2003 Jul 30;22(14):2309-33
pubmed: 12854095
Stat Med. 2014 Apr 30;33(9):1441-59
pubmed: 24285290
Cochrane Database Syst Rev. 2011 Oct 05;(10):CD007062
pubmed: 21975760
Stat Med. 2008 Sep 20;27(21):4279-300
pubmed: 18416445
Cochrane Database Syst Rev. 2012 Sep 12;(9):CD009351
pubmed: 22972140

Auteurs

Malcolm J Price (MJ)

Institute of Applied Health Research, University of Birmingham, Birmingham, UK.
NIHR Birmingham Biomedical Research Centre, University Hospitals Birmingham NHS Foundation Trust and University of Birmingham, Birmingham, UK.

Helen A Blake (HA)

Department of Medical Statistics, London School of Hygiene and Tropical Medicine, London, UK.
Department of Health Services Research and Policy, London School of Hygiene and Tropical Medicine, London, UK.

Sara Kenyon (S)

Institute of Applied Health Research, University of Birmingham, Birmingham, UK.

Ian R White (IR)

MRC Clinical Trials Unit, University College London, London, UK.

Dan Jackson (D)

Statistical Innovation Group, AstraZeneca, Cambridge, UK.

Jamie J Kirkham (JJ)

Department of Biostatistics, University of Liverpool, Liverpool, UK.

James P Neilson (JP)

Cochrane Pregnancy & Childbirth Group, Centre for Women's Health Research, University of Liverpool, Liverpool, UK.

Jonathan J Deeks (JJ)

Institute of Applied Health Research, University of Birmingham, Birmingham, UK.
NIHR Birmingham Biomedical Research Centre, University Hospitals Birmingham NHS Foundation Trust and University of Birmingham, Birmingham, UK.

Richard D Riley (RD)

Centre for Prognosis Research, Research Institute for Primary Care & Health Sciences, Keele University, UK.

Articles similaires

[Redispensing of expensive oral anticancer medicines: a practical application].

Lisanne N van Merendonk, Kübra Akgöl, Bastiaan Nuijen
1.00
Humans Antineoplastic Agents Administration, Oral Drug Costs Counterfeit Drugs

Smoking Cessation and Incident Cardiovascular Disease.

Jun Hwan Cho, Seung Yong Shin, Hoseob Kim et al.
1.00
Humans Male Smoking Cessation Cardiovascular Diseases Female
Humans United States Aged Cross-Sectional Studies Medicare Part C
1.00
Humans Yoga Low Back Pain Female Male

Classifications MeSH