Individual participant data meta-analysis to examine interactions between treatment effect and participant-level covariates: Statistical recommendations for conduct and planning.
effect modifier
individual participant data (IPD)
meta-analysis
subgroup effect
treatment-covariate interaction
Journal
Statistics in medicine
ISSN: 1097-0258
Titre abrégé: Stat Med
Pays: England
ID NLM: 8215016
Informations de publication
Date de publication:
10 07 2020
10 07 2020
Historique:
received:
26
09
2019
revised:
07
02
2020
accepted:
08
02
2020
pubmed:
1
5
2020
medline:
22
6
2021
entrez:
1
5
2020
Statut:
ppublish
Résumé
Precision medicine research often searches for treatment-covariate interactions, which refers to when a treatment effect (eg, measured as a mean difference, odds ratio, hazard ratio) changes across values of a participant-level covariate (eg, age, gender, biomarker). Single trials do not usually have sufficient power to detect genuine treatment-covariate interactions, which motivate the sharing of individual participant data (IPD) from multiple trials for meta-analysis. Here, we provide statistical recommendations for conducting and planning an IPD meta-analysis of randomized trials to examine treatment-covariate interactions. For conduct, two-stage and one-stage statistical models are described, and we recommend: (i) interactions should be estimated directly, and not by calculating differences in meta-analysis results for subgroups; (ii) interaction estimates should be based solely on within-study information; (iii) continuous covariates and outcomes should be analyzed on their continuous scale; (iv) nonlinear relationships should be examined for continuous covariates, using a multivariate meta-analysis of the trend (eg, using restricted cubic spline functions); and (v) translation of interactions into clinical practice is nontrivial, requiring individualized treatment effect prediction. For planning, we describe first why the decision to initiate an IPD meta-analysis project should not be based on between-study heterogeneity in the overall treatment effect; and second, how to calculate the power of a potential IPD meta-analysis project in advance of IPD collection, conditional on characteristics (eg, number of participants, standard deviation of covariates) of the trials (potentially) promising their IPD. Real IPD meta-analysis projects are used for illustration throughout.
Identifiants
pubmed: 32350891
doi: 10.1002/sim.8516
pmc: PMC7401032
doi:
Types de publication
Journal Article
Research Support, Non-U.S. Gov't
Langues
eng
Sous-ensembles de citation
IM
Pagination
2115-2137Subventions
Organisme : Department of Health
ID : DRF-2018-11-ST2-077
Pays : United Kingdom
Organisme : Medical Research Council
ID : MC_UU_12023/21
Pays : United Kingdom
Informations de copyright
© 2020 The Authors. Statistics in Medicine published by John Wiley & Sons, Ltd.
Références
BMJ. 2017 Mar 3;356:j573
pubmed: 28258124
Stat Med. 2016 Mar 30;35(7):966-77
pubmed: 26415869
BMJ. 2017 Sep 13;358:j3932
pubmed: 28903924
BMC Med Res Methodol. 2011 Jun 20;11:94
pubmed: 21689447
Stat Med. 2003 Aug 30;22(16):2591-602
pubmed: 12898546
BMJ. 2006 May 6;332(7549):1080
pubmed: 16675816
Res Synth Methods. 2019 Sep;10(3):360-375
pubmed: 30523676
Stat Med. 2011 Sep 10;30(20):2481-98
pubmed: 21268052
Stat Med. 2011 Dec 10;30(28):3341-60
pubmed: 21953493
BMJ Open. 2016 Sep 08;6(9):e011148
pubmed: 27609843
Stat Med. 2014 Sep 28;33(22):3844-58
pubmed: 24789760
JAMA. 2007 Sep 12;298(10):1209-12
pubmed: 17848656
Stat Methods Med Res. 2013 Jun;22(3):324-45
pubmed: 22491174
Biom J. 2012 May;54(3):370-84
pubmed: 22685003
Stat Med. 2019 Feb 10;38(3):326-338
pubmed: 30284314
BMJ. 2004 Oct 23;329(7472):966-8
pubmed: 15499115
J Clin Oncol. 2008 Mar 20;26(9):1397-9
pubmed: 18349388
PLoS One. 2013 Apr 09;8(4):e60650
pubmed: 23585842
Nature. 2018 Nov;563(7733):619-621
pubmed: 30482931
Stat Med. 2008 May 20;27(11):1870-93
pubmed: 18069721
J Clin Epidemiol. 2004 Mar;57(3):229-36
pubmed: 15066682
Res Synth Methods. 2015 Jun;6(2):157-74
pubmed: 26099484
Int J Epidemiol. 2019 Apr 1;48(2):596-608
pubmed: 30445577
Stat Med. 2001 Aug 15;20(15):2219-41
pubmed: 11468761
Stat Med. 2018 Apr 30;37(9):1550-1561
pubmed: 29383818
J Clin Epidemiol. 2019 Oct;114:72-83
pubmed: 31195109
Environ Health Perspect. 1993 Dec;101 Suppl 4:59-66
pubmed: 8206043
BMJ. 2018 Dec 10;363:k4245
pubmed: 30530757
J Clin Epidemiol. 2007 Oct;60(10):1002-9
pubmed: 17884593
Hypertension. 2005 May;45(5):907-13
pubmed: 15837826
BMJ. 2003 Jan 25;326(7382):219
pubmed: 12543843
Syst Rev. 2014 May 20;3:46
pubmed: 24887172
Stat Med. 2012 Dec 20;31(29):3821-39
pubmed: 22807043
Stat Med. 2017 Jan 15;36(1):136-196
pubmed: 27488683
Int J Technol Assess Health Care. 2008 Summer;24(3):358-61
pubmed: 18601805
J Natl Cancer Inst. 1994 Jun 1;86(11):829-35
pubmed: 8182763
Stat Med. 2017 Feb 28;36(5):772-789
pubmed: 27910122
PLoS One. 2012;7(10):e46042
pubmed: 23056232
PLoS Med. 2015 Oct 13;12(10):e1001886
pubmed: 26461078
Stat Methods Med Res. 1997 Jun;6(2):167-83
pubmed: 9261914
BMJ. 2013 Feb 05;346:e5793
pubmed: 23386361
PLoS Med. 2015 Jul 21;12(7):e1001855
pubmed: 26196287
Int J Epidemiol. 2016 Dec 1;45(6):2075-2088
pubmed: 27375287
Stat Med. 2004 Aug 30;23(16):2509-25
pubmed: 15287081
J Clin Epidemiol. 2001 Aug;54(8):774-81
pubmed: 11470385
J R Stat Soc Ser A Stat Soc. 2009 Jan;172(1):137-159
pubmed: 19381330
N Engl J Med. 2007 Jul 5;357(1):39-51
pubmed: 17611206
Stat Med. 2007 Jul 10;26(15):2982-99
pubmed: 17195960
Stat Med. 2017 Feb 28;36(5):855-875
pubmed: 27747915
Res Synth Methods. 2019 Dec;10(4):515-527
pubmed: 31111673
J Clin Epidemiol. 2015 Apr;68(4):470-4
pubmed: 25662008
J Clin Epidemiol. 2011 Sep;64(9):949-67
pubmed: 21411280
BMJ. 2011 Feb 10;342:d549
pubmed: 21310794
BMJ Open. 2014 Jan 27;4(1):e004188
pubmed: 24468721
Stat Med. 2020 Jul 10;39(15):2115-2137
pubmed: 32350891
Stat Med. 2001 Dec 30;20(24):3875-89
pubmed: 11782040
Ann Intern Med. 2011 May 17;154(10):680-3
pubmed: 21576536
Biometrics. 1946 Dec;2(6):110-4
pubmed: 20287815
BMC Med Res Methodol. 2019 Sep 2;19(1):183
pubmed: 31477023
Stat Med. 2006 Jan 15;25(1):127-41
pubmed: 16217841
BMC Med Res Methodol. 2018 May 18;18(1):41
pubmed: 29776399
BMJ. 2010 Mar 30;340:c117
pubmed: 20354011
Stat Med. 2010 Dec 20;29(29):3046-67
pubmed: 20827667
JAMA Intern Med. 2017 Apr 1;177(4):554-560
pubmed: 28192563
Lancet. 2005 Jan 22-28;365(9456):341-6
pubmed: 15664231
Stat Med. 2018 Dec 20;37(29):4404-4420
pubmed: 30101507
Stat Med. 2004 Aug 30;23(16):2567-86
pubmed: 15287085
Res Synth Methods. 2010 Jan;1(1):2-19
pubmed: 26056090