Identification of subpopulations with distinct treatment benefit rate using the Bayesian tree

Changyu Shen, Yang Hu, Xiaochun Li, Yadong Wang, Peng-Sheng Chen, Alfred E. Buxton

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

Characterization of a subpopulation by the difference in marginal means of the outcome under the intervention and control may not be sufficient to provide informative guidance for individual decision and public policy making. Specifically, often we are interested in the treatment benefit rate (TBR), that is, the probability of benefitting an intervention in a meaningful way. For binary outcomes, TBR is the proportion that has "unfavorable" outcome under the control and "favorable" outcome under the intervention. Identification of subpopulations with distinct TBR by baseline characteristics will have significant implications in clinical setting where a medical intervention with potential negative health impact is under consideration for a given patient. In addition, these subpopulations with unique TBR set the basis for guidance in implementing the intervention toward a more personalized scheme of treatment. In this article, we propose a Bayesian tree based latent variable model to seek subpopulations with distinct TBR. Our method offers a nonparametric Bayesian framework that accounts for the uncertainty in estimating potential outcomes and allows more exhaustive search of the partitions of the baseline covariates space. The method is evaluated through a simulation study and applied to a randomized clinical trial of implantable cardioverter defibrillators to reduce mortality.

Original languageEnglish (US)
JournalBiometrical Journal
DOIs
StateAccepted/In press - 2016

Fingerprint

Distinct
Guidance
Baseline
Therapeutics
Potential Outcomes
Implantable Defibrillators
Policy Making
Randomized Clinical Trial
Binary Outcomes
Public Policy
Bayesian Nonparametrics
Latent Variable Models
Exhaustive Search
Uncertainty
Mortality
Randomized Controlled Trials
Covariates
Health
Proportion
Partition

Keywords

  • Bayesian analysis
  • Causal inference
  • Heterogeneity of treatment effect
  • Matching
  • Subgroup analysis

ASJC Scopus subject areas

  • Statistics and Probability
  • Medicine(all)
  • Statistics, Probability and Uncertainty

Cite this

Identification of subpopulations with distinct treatment benefit rate using the Bayesian tree. / Shen, Changyu; Hu, Yang; Li, Xiaochun; Wang, Yadong; Chen, Peng-Sheng; Buxton, Alfred E.

In: Biometrical Journal, 2016.

Research output: Contribution to journalArticle

@article{34a1db0881624878b93c46658db8c346,
title = "Identification of subpopulations with distinct treatment benefit rate using the Bayesian tree",
abstract = "Characterization of a subpopulation by the difference in marginal means of the outcome under the intervention and control may not be sufficient to provide informative guidance for individual decision and public policy making. Specifically, often we are interested in the treatment benefit rate (TBR), that is, the probability of benefitting an intervention in a meaningful way. For binary outcomes, TBR is the proportion that has {"}unfavorable{"} outcome under the control and {"}favorable{"} outcome under the intervention. Identification of subpopulations with distinct TBR by baseline characteristics will have significant implications in clinical setting where a medical intervention with potential negative health impact is under consideration for a given patient. In addition, these subpopulations with unique TBR set the basis for guidance in implementing the intervention toward a more personalized scheme of treatment. In this article, we propose a Bayesian tree based latent variable model to seek subpopulations with distinct TBR. Our method offers a nonparametric Bayesian framework that accounts for the uncertainty in estimating potential outcomes and allows more exhaustive search of the partitions of the baseline covariates space. The method is evaluated through a simulation study and applied to a randomized clinical trial of implantable cardioverter defibrillators to reduce mortality.",
keywords = "Bayesian analysis, Causal inference, Heterogeneity of treatment effect, Matching, Subgroup analysis",
author = "Changyu Shen and Yang Hu and Xiaochun Li and Yadong Wang and Peng-Sheng Chen and Buxton, {Alfred E.}",
year = "2016",
doi = "10.1002/bimj.201500180",
language = "English (US)",
journal = "Biometrical Journal",
issn = "0323-3847",
publisher = "Wiley-VCH Verlag",

}

TY - JOUR

T1 - Identification of subpopulations with distinct treatment benefit rate using the Bayesian tree

AU - Shen, Changyu

AU - Hu, Yang

AU - Li, Xiaochun

AU - Wang, Yadong

AU - Chen, Peng-Sheng

AU - Buxton, Alfred E.

PY - 2016

Y1 - 2016

N2 - Characterization of a subpopulation by the difference in marginal means of the outcome under the intervention and control may not be sufficient to provide informative guidance for individual decision and public policy making. Specifically, often we are interested in the treatment benefit rate (TBR), that is, the probability of benefitting an intervention in a meaningful way. For binary outcomes, TBR is the proportion that has "unfavorable" outcome under the control and "favorable" outcome under the intervention. Identification of subpopulations with distinct TBR by baseline characteristics will have significant implications in clinical setting where a medical intervention with potential negative health impact is under consideration for a given patient. In addition, these subpopulations with unique TBR set the basis for guidance in implementing the intervention toward a more personalized scheme of treatment. In this article, we propose a Bayesian tree based latent variable model to seek subpopulations with distinct TBR. Our method offers a nonparametric Bayesian framework that accounts for the uncertainty in estimating potential outcomes and allows more exhaustive search of the partitions of the baseline covariates space. The method is evaluated through a simulation study and applied to a randomized clinical trial of implantable cardioverter defibrillators to reduce mortality.

AB - Characterization of a subpopulation by the difference in marginal means of the outcome under the intervention and control may not be sufficient to provide informative guidance for individual decision and public policy making. Specifically, often we are interested in the treatment benefit rate (TBR), that is, the probability of benefitting an intervention in a meaningful way. For binary outcomes, TBR is the proportion that has "unfavorable" outcome under the control and "favorable" outcome under the intervention. Identification of subpopulations with distinct TBR by baseline characteristics will have significant implications in clinical setting where a medical intervention with potential negative health impact is under consideration for a given patient. In addition, these subpopulations with unique TBR set the basis for guidance in implementing the intervention toward a more personalized scheme of treatment. In this article, we propose a Bayesian tree based latent variable model to seek subpopulations with distinct TBR. Our method offers a nonparametric Bayesian framework that accounts for the uncertainty in estimating potential outcomes and allows more exhaustive search of the partitions of the baseline covariates space. The method is evaluated through a simulation study and applied to a randomized clinical trial of implantable cardioverter defibrillators to reduce mortality.

KW - Bayesian analysis

KW - Causal inference

KW - Heterogeneity of treatment effect

KW - Matching

KW - Subgroup analysis

UR - http://www.scopus.com/inward/record.url?scp=84977111252&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84977111252&partnerID=8YFLogxK

U2 - 10.1002/bimj.201500180

DO - 10.1002/bimj.201500180

M3 - Article

C2 - 27356196

AN - SCOPUS:84977111252

JO - Biometrical Journal

JF - Biometrical Journal

SN - 0323-3847

ER -