Accurate confidence intervals for risk difference in meta-analysis with rare events.
Binary outcome
Confidence interval
Importance sampling
Meta-analysis
Rare events
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:
30 04 2020
30 04 2020
Historique:
received:
26
09
2019
accepted:
17
03
2020
entrez:
1
5
2020
pubmed:
1
5
2020
medline:
22
6
2021
Statut:
epublish
Résumé
Meta-analysis provides a useful statistical tool to effectively estimate treatment effect from multiple studies. When the outcome is binary and it is rare (e.g., safety data in clinical trials), the traditionally used methods may have unsatisfactory performance. We propose using importance sampling to compute confidence intervals for risk difference in meta-analysis with rare events. The proposed intervals are not exact, but they often have the coverage probabilities close to the nominal level. We compare the proposed accurate intervals with the existing intervals from the fixed- or random-effects models and the interval by Tian et al. (2009). We conduct extensive simulation studies to compare them with regards to coverage probability and average length, when data are simulated under the homogeneity or heterogeneity assumption of study effects. The proposed accurate interval based on the random-effects model for sample space ordering generally has satisfactory performance under the heterogeneity assumption, while the traditionally used interval based on the fixed-effects model works well when the studies are homogeneous.
Sections du résumé
BACKGROUND
Meta-analysis provides a useful statistical tool to effectively estimate treatment effect from multiple studies. When the outcome is binary and it is rare (e.g., safety data in clinical trials), the traditionally used methods may have unsatisfactory performance.
METHODS
We propose using importance sampling to compute confidence intervals for risk difference in meta-analysis with rare events. The proposed intervals are not exact, but they often have the coverage probabilities close to the nominal level. We compare the proposed accurate intervals with the existing intervals from the fixed- or random-effects models and the interval by Tian et al. (2009).
RESULTS
We conduct extensive simulation studies to compare them with regards to coverage probability and average length, when data are simulated under the homogeneity or heterogeneity assumption of study effects.
CONCLUSIONS
The proposed accurate interval based on the random-effects model for sample space ordering generally has satisfactory performance under the heterogeneity assumption, while the traditionally used interval based on the fixed-effects model works well when the studies are homogeneous.
Identifiants
pubmed: 32349702
doi: 10.1186/s12874-020-00954-8
pii: 10.1186/s12874-020-00954-8
pmc: PMC7191692
doi:
Types de publication
Journal Article
Research Support, N.I.H., Extramural
Research Support, Non-U.S. Gov't
Langues
eng
Sous-ensembles de citation
IM
Pagination
98Subventions
Organisme : NIGMS NIH HHS
ID : P20 GM109025
Pays : United States
Références
J Stat Comput Simul. 2018;88(6):1200-1210
pubmed: 31080301
BMC Med Res Methodol. 2018 Jul 4;18(1):70
pubmed: 29973146
Pharm Stat. 2012 Jul-Aug;11(4):287-94
pubmed: 22408050
Comput Stat Data Anal. 2013 Jan;57(1):246-261
pubmed: 27030785
Alzheimers Dement (N Y). 2018 Jun 14;4:366-371
pubmed: 30175231
BMC Med Res Methodol. 2016 Aug 02;16:90
pubmed: 27485595
J Natl Cancer Inst. 1959 Apr;22(4):719-48
pubmed: 13655060
Control Clin Trials. 1986 Sep;7(3):177-88
pubmed: 3802833
Biometrics. 2016 Dec;72(4):1378-1386
pubmed: 26909752
J Biopharm Stat. 2013;23(5):1081-90
pubmed: 23957517
Stat Methods Med Res. 2009 Aug;18(4):421-32
pubmed: 18562399
BMC Med Res Methodol. 2014 Dec 19;14:135
pubmed: 25524443
PLoS One. 2017 Dec 20;12(12):e0188709
pubmed: 29261690
Stat Med. 2013 Sep 10;32(20):3415-23
pubmed: 23553466
Stat Med. 2010 Sep 10;29(20):2078-89
pubmed: 20623822
J Am Stat Assoc. 2014 Oct;109(508):1450-1465
pubmed: 25620825
PLoS One. 2014 May 16;9(5):e97386
pubmed: 24837970
Stat Med. 2016 Apr 15;35(8):1257-66
pubmed: 26526165
Statistics (Ber). 2018;52(5):1086-1095
pubmed: 30906095
Stat Med. 2015 Mar 30;34(7):1097-116
pubmed: 25446971
Stat Med. 2009 Feb 28;28(5):721-38
pubmed: 19072749
Evid Based Ment Health. 2018 May;21(2):72-76
pubmed: 29650528
Stat Med. 2017 Sep 20;36(21):3437-3438
pubmed: 28776726
Stat Methods Med Res. 2018 Apr;27(4):1045-1055
pubmed: 27389669
Stat Med. 2004 May 15;23(9):1351-75
pubmed: 15116347
Biostatistics. 2009 Apr;10(2):275-81
pubmed: 18922759