Assessing heterogeneity of treatment effect in a clinical trial with the proportional interactions model

Stephanie A. Kovalchik, Ravi Varadhan, Carlos Orval Weiss

Research output: Contribution to journalArticle

Abstract

Understanding how individuals vary in their response to treatment is an important task of clinical research. For standard regression models, a proportional interactions model first described by Follmann and Proschan (1999) offers a powerful approach for identifying effect modification in a randomized clinical trial when multiple variables influence treatment response. In this paper, we present a framework for using the proportional interactions model in the context of a parallel-arm clinical trial with multiple prespecified candidate effect modifiers. To protect against model misspecification, we propose a selection strategy that considers all possible proportional interactions models. We develop a modified Bonferroni correction for multiple testing that accounts for the positive correlation among candidate models. We describe methods for constructing a confidence interval for the proportionality parameter. In simulation studies, we show that our modified Bonferroni adjustment controls familywise error and has greater power to detect proportional interactions compared with multiplcity-corrected subgroup analyses. We demonstrate our methodology by using the Studies of Left Ventricular Dysfunction Treatment trial, a placebo-controlled randomized clinical trial of the efficacy of enalapril to reduce the risk of death or hospitalization in chronic heart failure patients. An R package called anoint is available for implementing the proportional interactions methodology.

Original languageEnglish (US)
Pages (from-to)4906-4923
Number of pages18
JournalStatistics in Medicine
Volume32
Issue number28
DOIs
StatePublished - Dec 10 2013

Fingerprint

Treatment Effects
Clinical Trials
Directly proportional
Randomized Clinical Trial
Bonferroni
Interaction
Enalapril
Left Ventricular Dysfunction
Chronic Heart Failure
Hospitalization
Therapeutics
Randomized Controlled Trials
Heart Failure
Placebos
Model Misspecification
Model
Multiple Testing
Confidence Intervals
Methodology
Confidence interval

Keywords

  • Effect modification
  • Heterogeneity of treatment effect
  • Interaction
  • Risk stratification
  • Subgroup analysis

ASJC Scopus subject areas

  • Epidemiology
  • Statistics and Probability

Cite this

Assessing heterogeneity of treatment effect in a clinical trial with the proportional interactions model. / Kovalchik, Stephanie A.; Varadhan, Ravi; Weiss, Carlos Orval.

In: Statistics in Medicine, Vol. 32, No. 28, 10.12.2013, p. 4906-4923.

Research output: Contribution to journalArticle

@article{01db3fac1aea4c7a8a1fd6677b565a6b,
title = "Assessing heterogeneity of treatment effect in a clinical trial with the proportional interactions model",
abstract = "Understanding how individuals vary in their response to treatment is an important task of clinical research. For standard regression models, a proportional interactions model first described by Follmann and Proschan (1999) offers a powerful approach for identifying effect modification in a randomized clinical trial when multiple variables influence treatment response. In this paper, we present a framework for using the proportional interactions model in the context of a parallel-arm clinical trial with multiple prespecified candidate effect modifiers. To protect against model misspecification, we propose a selection strategy that considers all possible proportional interactions models. We develop a modified Bonferroni correction for multiple testing that accounts for the positive correlation among candidate models. We describe methods for constructing a confidence interval for the proportionality parameter. In simulation studies, we show that our modified Bonferroni adjustment controls familywise error and has greater power to detect proportional interactions compared with multiplcity-corrected subgroup analyses. We demonstrate our methodology by using the Studies of Left Ventricular Dysfunction Treatment trial, a placebo-controlled randomized clinical trial of the efficacy of enalapril to reduce the risk of death or hospitalization in chronic heart failure patients. An R package called anoint is available for implementing the proportional interactions methodology.",
keywords = "Effect modification, Heterogeneity of treatment effect, Interaction, Risk stratification, Subgroup analysis",
author = "Kovalchik, {Stephanie A.} and Ravi Varadhan and Weiss, {Carlos Orval}",
year = "2013",
month = "12",
day = "10",
doi = "10.1002/sim.5881",
language = "English (US)",
volume = "32",
pages = "4906--4923",
journal = "Statistics in Medicine",
issn = "0277-6715",
publisher = "John Wiley and Sons Ltd",
number = "28",

}

TY - JOUR

T1 - Assessing heterogeneity of treatment effect in a clinical trial with the proportional interactions model

AU - Kovalchik, Stephanie A.

AU - Varadhan, Ravi

AU - Weiss, Carlos Orval

PY - 2013/12/10

Y1 - 2013/12/10

N2 - Understanding how individuals vary in their response to treatment is an important task of clinical research. For standard regression models, a proportional interactions model first described by Follmann and Proschan (1999) offers a powerful approach for identifying effect modification in a randomized clinical trial when multiple variables influence treatment response. In this paper, we present a framework for using the proportional interactions model in the context of a parallel-arm clinical trial with multiple prespecified candidate effect modifiers. To protect against model misspecification, we propose a selection strategy that considers all possible proportional interactions models. We develop a modified Bonferroni correction for multiple testing that accounts for the positive correlation among candidate models. We describe methods for constructing a confidence interval for the proportionality parameter. In simulation studies, we show that our modified Bonferroni adjustment controls familywise error and has greater power to detect proportional interactions compared with multiplcity-corrected subgroup analyses. We demonstrate our methodology by using the Studies of Left Ventricular Dysfunction Treatment trial, a placebo-controlled randomized clinical trial of the efficacy of enalapril to reduce the risk of death or hospitalization in chronic heart failure patients. An R package called anoint is available for implementing the proportional interactions methodology.

AB - Understanding how individuals vary in their response to treatment is an important task of clinical research. For standard regression models, a proportional interactions model first described by Follmann and Proschan (1999) offers a powerful approach for identifying effect modification in a randomized clinical trial when multiple variables influence treatment response. In this paper, we present a framework for using the proportional interactions model in the context of a parallel-arm clinical trial with multiple prespecified candidate effect modifiers. To protect against model misspecification, we propose a selection strategy that considers all possible proportional interactions models. We develop a modified Bonferroni correction for multiple testing that accounts for the positive correlation among candidate models. We describe methods for constructing a confidence interval for the proportionality parameter. In simulation studies, we show that our modified Bonferroni adjustment controls familywise error and has greater power to detect proportional interactions compared with multiplcity-corrected subgroup analyses. We demonstrate our methodology by using the Studies of Left Ventricular Dysfunction Treatment trial, a placebo-controlled randomized clinical trial of the efficacy of enalapril to reduce the risk of death or hospitalization in chronic heart failure patients. An R package called anoint is available for implementing the proportional interactions methodology.

KW - Effect modification

KW - Heterogeneity of treatment effect

KW - Interaction

KW - Risk stratification

KW - Subgroup analysis

UR - http://www.scopus.com/inward/record.url?scp=84887172238&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84887172238&partnerID=8YFLogxK

U2 - 10.1002/sim.5881

DO - 10.1002/sim.5881

M3 - Article

VL - 32

SP - 4906

EP - 4923

JO - Statistics in Medicine

JF - Statistics in Medicine

SN - 0277-6715

IS - 28

ER -