Evaluating the performance of propensity score matching based approaches in individual patient data meta-analysis.
Bias
Confounding
IPD-MA
Observational studies
Propensity score matching
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:
23 11 2021
23 11 2021
Historique:
received:
28
09
2020
accepted:
28
10
2021
entrez:
24
11
2021
pubmed:
25
11
2021
medline:
27
1
2022
Statut:
epublish
Résumé
Individual-patient data meta-analysis (IPD-MA) is an increasingly popular approach because of its analytical benefits. IPD-MA of observational studies must overcome the problem of confounding, otherwise biased estimates of treatment effect may be obtained. One approach to reducing confounding bias could be the use of propensity score matching (PSM). IPD-MA can be considered as two-stage clustered data (patients within studies) and propensity score matching can be implemented within studies, across studies, and combining both. This article focuses on implementation of four PSM-based approaches for the analysis of data structure that exploit IPD-MA in two ways: (i) estimation of propensity score model using single-level or random-effects logistic regression; and (ii) matching of propensity scores (PS) across studies, within studies or preferential-within studies. We investigated the performance of these approaches through a simulation study, which considers an IPD-MA that examined the success of different treatments for multidrug-resistant tuberculosis (MDR-TB). The simulation parameters were varied according to three treatment prevalences (according to studies, 50% and 30%), three levels of heterogeneity between studies (low, moderate and high) and three levels of pooled odds ratio (1, 1.5, 3). All approaches showed greater biases at the higher levels of heterogeneity regardless of the choices of treatment prevalences. However, matching of propensity scores using within-study and preferential-within study reported better performance compared to matching across studies when treatment prevalence varied across-studies. For fixed prevalences, a random-effect propensity score model to estimate propensity scores followed by matching of propensity scores across-studies achieved lower biases compared to other PSM-based approaches. Propensity score matching has wide application in health research while only limited literature is available on the implementation of PSM methods in IPD-MA, and until now methodological performance of PSM methods have not been examined. We believe, this work offers an intuition to the applied researcher for the choice of the PSM-based approaches.
Sections du résumé
BACKGROUND
Individual-patient data meta-analysis (IPD-MA) is an increasingly popular approach because of its analytical benefits. IPD-MA of observational studies must overcome the problem of confounding, otherwise biased estimates of treatment effect may be obtained. One approach to reducing confounding bias could be the use of propensity score matching (PSM). IPD-MA can be considered as two-stage clustered data (patients within studies) and propensity score matching can be implemented within studies, across studies, and combining both.
METHODS
This article focuses on implementation of four PSM-based approaches for the analysis of data structure that exploit IPD-MA in two ways: (i) estimation of propensity score model using single-level or random-effects logistic regression; and (ii) matching of propensity scores (PS) across studies, within studies or preferential-within studies. We investigated the performance of these approaches through a simulation study, which considers an IPD-MA that examined the success of different treatments for multidrug-resistant tuberculosis (MDR-TB). The simulation parameters were varied according to three treatment prevalences (according to studies, 50% and 30%), three levels of heterogeneity between studies (low, moderate and high) and three levels of pooled odds ratio (1, 1.5, 3).
RESULTS
All approaches showed greater biases at the higher levels of heterogeneity regardless of the choices of treatment prevalences. However, matching of propensity scores using within-study and preferential-within study reported better performance compared to matching across studies when treatment prevalence varied across-studies. For fixed prevalences, a random-effect propensity score model to estimate propensity scores followed by matching of propensity scores across-studies achieved lower biases compared to other PSM-based approaches.
CONCLUSIONS
Propensity score matching has wide application in health research while only limited literature is available on the implementation of PSM methods in IPD-MA, and until now methodological performance of PSM methods have not been examined. We believe, this work offers an intuition to the applied researcher for the choice of the PSM-based approaches.
Identifiants
pubmed: 34814845
doi: 10.1186/s12874-021-01452-1
pii: 10.1186/s12874-021-01452-1
pmc: PMC8609730
doi:
Types de publication
Journal Article
Meta-Analysis
Langues
eng
Sous-ensembles de citation
IM
Pagination
257Informations de copyright
© 2021. The Author(s).
Références
Stat Med. 2002 Jun 15;21(11):1539-58
pubmed: 12111919
Multivariate Behav Res. 2011 May 31;46(3):514-43
pubmed: 26735885
J Clin Epidemiol. 2013 Sep;66(9):1029-37
pubmed: 23800533
J Am Coll Cardiol. 2017 Jan 24;69(3):345-357
pubmed: 28104076
Stat Med. 2014 Mar 15;33(6):1057-69
pubmed: 24123228
Stat Med. 2002 Jun 15;21(11):1559-73
pubmed: 12111920
Stat Med. 2020 Jul 30;39(17):2308-2323
pubmed: 32297677
Stat Methods Med Res. 2016 Oct;25(5):2214-2237
pubmed: 24463885
Stat Med. 2006 Dec 30;25(24):4279-92
pubmed: 16947139
Res Synth Methods. 2015 Dec;6(4):293-309
pubmed: 26287812
Am J Epidemiol. 2010 Nov 1;172(9):1092-7
pubmed: 20802241
Stat Med. 2016 May 30;35(12):2074-91
pubmed: 26833893
Pharm Stat. 2011 Mar-Apr;10(2):150-61
pubmed: 20925139
Stat Med. 1998 Oct 15;17(19):2265-81
pubmed: 9802183
PLoS One. 2016 Mar 29;11(3):e0151724
pubmed: 27022741
Stat Med. 2007 Feb 20;26(4):734-53
pubmed: 16708349
Stat Med. 2013 Aug 30;32(19):3357-72
pubmed: 23553813
PLoS Med. 2012;9(8):e1001300
pubmed: 22952439
JAMA. 2000 Apr 19;283(15):2008-12
pubmed: 10789670
Med Care. 1993 May;31(5 Suppl):YS54-9
pubmed: 8492586
Multivariate Behav Res. 2011 May;46(3):399-424
pubmed: 21818162
Stat Med. 2014 Sep 10;33(20):3556-75
pubmed: 24771662
J Hypertens Suppl. 1996 Sep;14(2):S9-12; discussion S13
pubmed: 8934372