Variable selection for covariate-adjusted semiparametric inference in randomized clinical trials

Shuai Yuan, Hao Zhang, Marie Davidian

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

Extensive baseline covariate information is routinely collected on participants in randomized clinical trials, and it is well recognized that a proper covariate-adjusted analysis can improve the efficiency of inference on the treatment effect. However, such covariate adjustment has engendered considerable controversy, as post hoc selection of covariates may involve subjectivity and may lead to biased inference, whereas prior specification of the adjustment may exclude important variables from consideration. Accordingly, how to select covariates objectively to gain maximal efficiency is of broad interest. We propose and study the use of modern variable selection methods for this purpose in the context of a semiparametric framework, under which variable selection in modeling the relationship between outcome and covariates is separated from estimation of the treatment effect, circumventing the potential for selection bias associated with standard analysis of covariance methods. We demonstrate that such objective variable selection techniques combined with this framework can identify key variables and lead to unbiased and efficient inference on the treatment effect. A critical issue in finite samples is validity of estimators of uncertainty, such as standard errors and confidence intervals for the treatment effect. We propose an approach to estimation of sampling variation of estimated treatment effect and show its superior performance relative to that of existing methods.

Original languageEnglish (US)
Pages (from-to)3789-3804
Number of pages16
JournalStatistics in Medicine
Volume31
Issue number29
DOIs
StatePublished - Dec 20 2012
Externally publishedYes

Fingerprint

Semiparametric Inference
Randomized Clinical Trial
Variable Selection
Treatment Effects
Covariates
Randomized Controlled Trials
Selection Bias
Uncertainty
Adjustment
Confidence Intervals
Selection of Variables
Analysis of Covariance
Standard error
Biased
Confidence interval
Baseline
Specification
Estimator
Modeling
Demonstrate

Keywords

  • Covariate adjustment
  • False selection rate control
  • Oracle property
  • Semiparametric treatment effect estimation
  • Shrinkage methods
  • Variable selection

ASJC Scopus subject areas

  • Epidemiology
  • Statistics and Probability

Cite this

Variable selection for covariate-adjusted semiparametric inference in randomized clinical trials. / Yuan, Shuai; Zhang, Hao; Davidian, Marie.

In: Statistics in Medicine, Vol. 31, No. 29, 20.12.2012, p. 3789-3804.

Research output: Contribution to journalArticle

@article{07856c0ebb7b4422b0871bba69920ff9,
title = "Variable selection for covariate-adjusted semiparametric inference in randomized clinical trials",
abstract = "Extensive baseline covariate information is routinely collected on participants in randomized clinical trials, and it is well recognized that a proper covariate-adjusted analysis can improve the efficiency of inference on the treatment effect. However, such covariate adjustment has engendered considerable controversy, as post hoc selection of covariates may involve subjectivity and may lead to biased inference, whereas prior specification of the adjustment may exclude important variables from consideration. Accordingly, how to select covariates objectively to gain maximal efficiency is of broad interest. We propose and study the use of modern variable selection methods for this purpose in the context of a semiparametric framework, under which variable selection in modeling the relationship between outcome and covariates is separated from estimation of the treatment effect, circumventing the potential for selection bias associated with standard analysis of covariance methods. We demonstrate that such objective variable selection techniques combined with this framework can identify key variables and lead to unbiased and efficient inference on the treatment effect. A critical issue in finite samples is validity of estimators of uncertainty, such as standard errors and confidence intervals for the treatment effect. We propose an approach to estimation of sampling variation of estimated treatment effect and show its superior performance relative to that of existing methods.",
keywords = "Covariate adjustment, False selection rate control, Oracle property, Semiparametric treatment effect estimation, Shrinkage methods, Variable selection",
author = "Shuai Yuan and Hao Zhang and Marie Davidian",
year = "2012",
month = "12",
day = "20",
doi = "10.1002/sim.5433",
language = "English (US)",
volume = "31",
pages = "3789--3804",
journal = "Statistics in Medicine",
issn = "0277-6715",
publisher = "John Wiley and Sons Ltd",
number = "29",

}

TY - JOUR

T1 - Variable selection for covariate-adjusted semiparametric inference in randomized clinical trials

AU - Yuan, Shuai

AU - Zhang, Hao

AU - Davidian, Marie

PY - 2012/12/20

Y1 - 2012/12/20

N2 - Extensive baseline covariate information is routinely collected on participants in randomized clinical trials, and it is well recognized that a proper covariate-adjusted analysis can improve the efficiency of inference on the treatment effect. However, such covariate adjustment has engendered considerable controversy, as post hoc selection of covariates may involve subjectivity and may lead to biased inference, whereas prior specification of the adjustment may exclude important variables from consideration. Accordingly, how to select covariates objectively to gain maximal efficiency is of broad interest. We propose and study the use of modern variable selection methods for this purpose in the context of a semiparametric framework, under which variable selection in modeling the relationship between outcome and covariates is separated from estimation of the treatment effect, circumventing the potential for selection bias associated with standard analysis of covariance methods. We demonstrate that such objective variable selection techniques combined with this framework can identify key variables and lead to unbiased and efficient inference on the treatment effect. A critical issue in finite samples is validity of estimators of uncertainty, such as standard errors and confidence intervals for the treatment effect. We propose an approach to estimation of sampling variation of estimated treatment effect and show its superior performance relative to that of existing methods.

AB - Extensive baseline covariate information is routinely collected on participants in randomized clinical trials, and it is well recognized that a proper covariate-adjusted analysis can improve the efficiency of inference on the treatment effect. However, such covariate adjustment has engendered considerable controversy, as post hoc selection of covariates may involve subjectivity and may lead to biased inference, whereas prior specification of the adjustment may exclude important variables from consideration. Accordingly, how to select covariates objectively to gain maximal efficiency is of broad interest. We propose and study the use of modern variable selection methods for this purpose in the context of a semiparametric framework, under which variable selection in modeling the relationship between outcome and covariates is separated from estimation of the treatment effect, circumventing the potential for selection bias associated with standard analysis of covariance methods. We demonstrate that such objective variable selection techniques combined with this framework can identify key variables and lead to unbiased and efficient inference on the treatment effect. A critical issue in finite samples is validity of estimators of uncertainty, such as standard errors and confidence intervals for the treatment effect. We propose an approach to estimation of sampling variation of estimated treatment effect and show its superior performance relative to that of existing methods.

KW - Covariate adjustment

KW - False selection rate control

KW - Oracle property

KW - Semiparametric treatment effect estimation

KW - Shrinkage methods

KW - Variable selection

UR - http://www.scopus.com/inward/record.url?scp=84870059607&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84870059607&partnerID=8YFLogxK

U2 - 10.1002/sim.5433

DO - 10.1002/sim.5433

M3 - Article

C2 - 22733628

AN - SCOPUS:84870059607

VL - 31

SP - 3789

EP - 3804

JO - Statistics in Medicine

JF - Statistics in Medicine

SN - 0277-6715

IS - 29

ER -