A Bayesian hierarchical model estimating CACE in meta-analysis of randomized clinical trials with noncompliance

Jincheng Zhou, James S Hodges, M. Fareed K. Suri, Haitao Chu

Research output: Contribution to journalArticle

Abstract

Noncompliance to assigned treatment is a common challenge in analysis and interpretation of randomized clinical trials. The complier average causal effect (CACE) approach provides a useful tool for addressing noncompliance, where CACE is defined as the average difference in potential outcomes for the response in the subpopulation of subjects who comply with their assigned treatments. In this article, we present a Bayesian hierarchical model to estimate the CACE in a meta-analysis of randomized clinical trials where compliance may be heterogeneous between studies. Between-study heterogeneity is taken into account with study-specific random effects. The results are illustrated by a re-analysis of a meta-analysis comparing the effect of epidural analgesia in labor versus no or other analgesia in labor on the outcome cesarean section, where noncompliance varied between studies. Finally, we present simulations evaluating the performance of the proposed approach and illustrate the importance of including appropriate random effects and the impact of over- and under-fitting.

Original languageEnglish (US)
JournalBiometrics
DOIs
StatePublished - Jan 1 2019

Fingerprint

Noncompliance
Bayesian Hierarchical Model
Randomized Clinical Trial
Causal Effect
randomized clinical trials
analgesia
meta-analysis
Meta-Analysis
labor
Randomized Controlled Trials
Personnel
Epidural Analgesia
cesarean section
Random Effects
Cesarean Section
compliance
Analgesia
Compliance
Potential Outcomes
Overfitting

Keywords

  • Bayesian hierarchical model
  • CACE
  • causal effect
  • meta-analysis
  • noncompliance
  • randomized trial

Cite this

A Bayesian hierarchical model estimating CACE in meta-analysis of randomized clinical trials with noncompliance. / Zhou, Jincheng; Hodges, James S; Suri, M. Fareed K.; Chu, Haitao.

In: Biometrics, 01.01.2019.

Research output: Contribution to journalArticle

@article{b463cbc9f7be479682a66c86c66edc4c,
title = "A Bayesian hierarchical model estimating CACE in meta-analysis of randomized clinical trials with noncompliance",
abstract = "Noncompliance to assigned treatment is a common challenge in analysis and interpretation of randomized clinical trials. The complier average causal effect (CACE) approach provides a useful tool for addressing noncompliance, where CACE is defined as the average difference in potential outcomes for the response in the subpopulation of subjects who comply with their assigned treatments. In this article, we present a Bayesian hierarchical model to estimate the CACE in a meta-analysis of randomized clinical trials where compliance may be heterogeneous between studies. Between-study heterogeneity is taken into account with study-specific random effects. The results are illustrated by a re-analysis of a meta-analysis comparing the effect of epidural analgesia in labor versus no or other analgesia in labor on the outcome cesarean section, where noncompliance varied between studies. Finally, we present simulations evaluating the performance of the proposed approach and illustrate the importance of including appropriate random effects and the impact of over- and under-fitting.",
keywords = "Bayesian hierarchical model, CACE, causal effect, meta-analysis, noncompliance, randomized trial",
author = "Jincheng Zhou and Hodges, {James S} and Suri, {M. Fareed K.} and Haitao Chu",
year = "2019",
month = "1",
day = "1",
doi = "10.1111/biom.13028",
language = "English (US)",
journal = "Biometrics",
issn = "0006-341X",
publisher = "Wiley-Blackwell",

}

TY - JOUR

T1 - A Bayesian hierarchical model estimating CACE in meta-analysis of randomized clinical trials with noncompliance

AU - Zhou, Jincheng

AU - Hodges, James S

AU - Suri, M. Fareed K.

AU - Chu, Haitao

PY - 2019/1/1

Y1 - 2019/1/1

N2 - Noncompliance to assigned treatment is a common challenge in analysis and interpretation of randomized clinical trials. The complier average causal effect (CACE) approach provides a useful tool for addressing noncompliance, where CACE is defined as the average difference in potential outcomes for the response in the subpopulation of subjects who comply with their assigned treatments. In this article, we present a Bayesian hierarchical model to estimate the CACE in a meta-analysis of randomized clinical trials where compliance may be heterogeneous between studies. Between-study heterogeneity is taken into account with study-specific random effects. The results are illustrated by a re-analysis of a meta-analysis comparing the effect of epidural analgesia in labor versus no or other analgesia in labor on the outcome cesarean section, where noncompliance varied between studies. Finally, we present simulations evaluating the performance of the proposed approach and illustrate the importance of including appropriate random effects and the impact of over- and under-fitting.

AB - Noncompliance to assigned treatment is a common challenge in analysis and interpretation of randomized clinical trials. The complier average causal effect (CACE) approach provides a useful tool for addressing noncompliance, where CACE is defined as the average difference in potential outcomes for the response in the subpopulation of subjects who comply with their assigned treatments. In this article, we present a Bayesian hierarchical model to estimate the CACE in a meta-analysis of randomized clinical trials where compliance may be heterogeneous between studies. Between-study heterogeneity is taken into account with study-specific random effects. The results are illustrated by a re-analysis of a meta-analysis comparing the effect of epidural analgesia in labor versus no or other analgesia in labor on the outcome cesarean section, where noncompliance varied between studies. Finally, we present simulations evaluating the performance of the proposed approach and illustrate the importance of including appropriate random effects and the impact of over- and under-fitting.

KW - Bayesian hierarchical model

KW - CACE

KW - causal effect

KW - meta-analysis

KW - noncompliance

KW - randomized trial

UR - http://www.scopus.com/inward/record.url?scp=85063902413&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85063902413&partnerID=8YFLogxK

U2 - 10.1111/biom.13028

DO - 10.1111/biom.13028

M3 - Article

JO - Biometrics

JF - Biometrics

SN - 0006-341X

ER -