A simulation study to compare different estimation approaches for network meta-analysis and corresponding methods to evaluate the consistency assumption.
Consistency assumption
Indirect comparison
Mixed treatment comparison
Multiple treatments meta-analysis
Network meta-analysis
Simulation study
Journal
BMC medical research methodology
ISSN: 1471-2288
Titre abrégé: BMC Med Res Methodol
Pays: England
ID NLM: 100968545
Informations de publication
Date de publication:
24 02 2020
24 02 2020
Historique:
received:
13
08
2019
accepted:
30
01
2020
entrez:
26
2
2020
pubmed:
26
2
2020
medline:
12
1
2021
Statut:
epublish
Résumé
Network meta-analysis (NMA) is becoming increasingly popular in systematic reviews and health technology assessments. However, there is still ambiguity concerning the properties of the estimation approaches as well as for the methods to evaluate the consistency assumption. We conducted a simulation study for networks with up to 5 interventions. We investigated the properties of different methods and give recommendations for practical application. We evaluated the performance of 3 different models for complex networks as well as corresponding global methods to evaluate the consistency assumption. The models are the frequentist graph-theoretical approach netmeta, the Bayesian mixed treatment comparisons (MTC) consistency model, and the MTC consistency model with stepwise removal of studies contributing to inconsistency identified in a leverage plot. We found that with a high degree of inconsistency none of the evaluated effect estimators produced reliable results, whereas with moderate or no inconsistency the estimator from the MTC consistency model and the netmeta estimator showed acceptable properties. We also saw a dependency on the amount of heterogeneity. Concerning the evaluated methods to evaluate the consistency assumption, none was shown to be suitable. Based on our results we recommend a pragmatic approach for practical application in NMA. The estimator from the netmeta approach or the estimator from the Bayesian MTC consistency model should be preferred. Since none of the methods to evaluate the consistency assumption showed satisfactory results, users should have a strong focus on the similarity as well as the homogeneity assumption.
Sections du résumé
BACKGROUND
Network meta-analysis (NMA) is becoming increasingly popular in systematic reviews and health technology assessments. However, there is still ambiguity concerning the properties of the estimation approaches as well as for the methods to evaluate the consistency assumption.
METHODS
We conducted a simulation study for networks with up to 5 interventions. We investigated the properties of different methods and give recommendations for practical application. We evaluated the performance of 3 different models for complex networks as well as corresponding global methods to evaluate the consistency assumption. The models are the frequentist graph-theoretical approach netmeta, the Bayesian mixed treatment comparisons (MTC) consistency model, and the MTC consistency model with stepwise removal of studies contributing to inconsistency identified in a leverage plot.
RESULTS
We found that with a high degree of inconsistency none of the evaluated effect estimators produced reliable results, whereas with moderate or no inconsistency the estimator from the MTC consistency model and the netmeta estimator showed acceptable properties. We also saw a dependency on the amount of heterogeneity. Concerning the evaluated methods to evaluate the consistency assumption, none was shown to be suitable.
CONCLUSIONS
Based on our results we recommend a pragmatic approach for practical application in NMA. The estimator from the netmeta approach or the estimator from the Bayesian MTC consistency model should be preferred. Since none of the methods to evaluate the consistency assumption showed satisfactory results, users should have a strong focus on the similarity as well as the homogeneity assumption.
Identifiants
pubmed: 32093605
doi: 10.1186/s12874-020-0917-3
pii: 10.1186/s12874-020-0917-3
pmc: PMC7041240
doi:
Substances chimiques
Antidepressive Agents
0
Types de publication
Comparative Study
Journal Article
Research Support, Non-U.S. Gov't
Langues
eng
Sous-ensembles de citation
IM
Pagination
36Références
Stat Med. 2014 Nov 10;33(25):4353-69
pubmed: 24942211
Epidemiol Health. 2019;41:e2019013
pubmed: 30999733
BMC Med. 2014 Jun 05;12:93
pubmed: 24898705
Health Technol Assess. 2005 Jul;9(26):1-134, iii-iv
pubmed: 16014203
J Clin Epidemiol. 1997 Jun;50(6):683-91
pubmed: 9250266
BMC Med Res Methodol. 2014 Feb 18;14:25
pubmed: 24548571
Res Synth Methods. 2012 Dec;3(4):312-24
pubmed: 26053424
BMC Med Res Methodol. 2013 Jan 11;13:2
pubmed: 23311298
Stat Med. 2014 Sep 20;33(21):3639-54
pubmed: 24777711
PLoS One. 2016 Aug 10;11(8):e0160712
pubmed: 27508415
Stat Med. 2016 Mar 15;35(6):819-39
pubmed: 26423209
Control Clin Trials. 1986 Sep;7(3):177-88
pubmed: 3802833
Stat Med. 2009 Nov 10;28(25):3049-67
pubmed: 19630097
Res Synth Methods. 2012 Jun;3(2):80-97
pubmed: 26062083
J Clin Epidemiol. 2014 Feb;67(2):138-43
pubmed: 24090930
BMC Med Res Methodol. 2016 Jul 28;16:87
pubmed: 27465416
Res Synth Methods. 2012 Jun;3(2):111-25
pubmed: 26062085
J Clin Epidemiol. 2017 Feb;82:20-28
pubmed: 27864068
Stat Med. 2017 Mar 15;36(6):917-927
pubmed: 27917493
Stat Med. 2004 Oct 30;23(20):3105-24
pubmed: 15449338
BMJ. 2009 Apr 03;338:b1147
pubmed: 19346285
BMC Med Res Methodol. 2012 Sep 12;12:138
pubmed: 22970794
Res Synth Methods. 2012 Jun;3(2):98-110
pubmed: 26062084
Stat Med. 2010 Mar 30;29(7-8):932-44
pubmed: 20213715
PLoS One. 2011 Jan 21;6(1):e16237
pubmed: 21283698
Stat Med. 2003 Sep 15;22(17):2693-710
pubmed: 12939780
Res Synth Methods. 2012 Jun;3(2):69-70
pubmed: 26062081
Res Synth Methods. 2017 Dec;8(4):416-434
pubmed: 28585257
Res Synth Methods. 2012 Dec;3(4):300-11
pubmed: 26053423
BMC Med Res Methodol. 2014 Sep 19;14:106
pubmed: 25239546