A varying-coefficient model for the evaluation of time-varying concomitant intervention effects in longitudinal studies

Colin O. Wu, Xin Tian, Heejung Bang

Research output: Contribution to journalArticle

5 Citations (Scopus)

Abstract

Concomitant interventions are often introduced during a longitudinal clinical trial to patients who respond undesirably to the pre-specified treatments. In addition to the main objective of evaluating the pre-specified treatment effects, an important secondary objective in such a trial is to evaluate whether a concomitant intervention could change a patient's response over time. Because the initiation of a concomitant intervention may depend on the patient's general trend of pre-intervention outcomes, regression approaches that treat the presence of the intervention as a time-dependent covariate may lead to biased estimates for the intervention effects. Borrowing the techniques of Follmann and Wu (Biometrics 1995; 51:151-168) for modeling informative missing data, we propose a varying-coefficient mixed-effects model to evaluate the patient's longitudinal outcome trends before and after the patient's starting time of the intervention. By allowing the random coefficients to be correlated with the patient's starting time of the intervention, our model leads to less biased estimates of the intervention effects. Nonparametric estimation and inferences of the coefficient curves and intervention effects are developed using B-splines. Our methods are demonstrated through a longitudinal clinical trial in depression and heart disease and a simulation study.

Original languageEnglish (US)
Pages (from-to)3042-3056
Number of pages15
JournalStatistics in Medicine
Volume27
Issue number16
DOIs
StatePublished - Jul 20 2008
Externally publishedYes

Fingerprint

Varying Coefficient Model
Longitudinal Study
Longitudinal Studies
Time-varying
Evaluation
Clinical Trials
Biased
Heart Diseases
Time-dependent Covariates
Nonparametric Inference
Mixed Effects Model
Varying Coefficients
Random Coefficients
Evaluate
Treatment Effects
Nonparametric Estimation
B-spline
Missing Data
Biometrics
Estimate

Keywords

  • Change-point models
  • Concomitant intervention
  • Longitudinal study
  • Polynomial splines
  • Shared-parameter model
  • Varying-coefficient model

ASJC Scopus subject areas

  • Epidemiology

Cite this

A varying-coefficient model for the evaluation of time-varying concomitant intervention effects in longitudinal studies. / Wu, Colin O.; Tian, Xin; Bang, Heejung.

In: Statistics in Medicine, Vol. 27, No. 16, 20.07.2008, p. 3042-3056.

Research output: Contribution to journalArticle

@article{0c4ee36fb1dc4eda81680708927b2ee8,
title = "A varying-coefficient model for the evaluation of time-varying concomitant intervention effects in longitudinal studies",
abstract = "Concomitant interventions are often introduced during a longitudinal clinical trial to patients who respond undesirably to the pre-specified treatments. In addition to the main objective of evaluating the pre-specified treatment effects, an important secondary objective in such a trial is to evaluate whether a concomitant intervention could change a patient's response over time. Because the initiation of a concomitant intervention may depend on the patient's general trend of pre-intervention outcomes, regression approaches that treat the presence of the intervention as a time-dependent covariate may lead to biased estimates for the intervention effects. Borrowing the techniques of Follmann and Wu (Biometrics 1995; 51:151-168) for modeling informative missing data, we propose a varying-coefficient mixed-effects model to evaluate the patient's longitudinal outcome trends before and after the patient's starting time of the intervention. By allowing the random coefficients to be correlated with the patient's starting time of the intervention, our model leads to less biased estimates of the intervention effects. Nonparametric estimation and inferences of the coefficient curves and intervention effects are developed using B-splines. Our methods are demonstrated through a longitudinal clinical trial in depression and heart disease and a simulation study.",
keywords = "Change-point models, Concomitant intervention, Longitudinal study, Polynomial splines, Shared-parameter model, Varying-coefficient model",
author = "Wu, {Colin O.} and Xin Tian and Heejung Bang",
year = "2008",
month = "7",
day = "20",
doi = "10.1002/sim.3262",
language = "English (US)",
volume = "27",
pages = "3042--3056",
journal = "Statistics in Medicine",
issn = "0277-6715",
publisher = "John Wiley and Sons Ltd",
number = "16",

}

TY - JOUR

T1 - A varying-coefficient model for the evaluation of time-varying concomitant intervention effects in longitudinal studies

AU - Wu, Colin O.

AU - Tian, Xin

AU - Bang, Heejung

PY - 2008/7/20

Y1 - 2008/7/20

N2 - Concomitant interventions are often introduced during a longitudinal clinical trial to patients who respond undesirably to the pre-specified treatments. In addition to the main objective of evaluating the pre-specified treatment effects, an important secondary objective in such a trial is to evaluate whether a concomitant intervention could change a patient's response over time. Because the initiation of a concomitant intervention may depend on the patient's general trend of pre-intervention outcomes, regression approaches that treat the presence of the intervention as a time-dependent covariate may lead to biased estimates for the intervention effects. Borrowing the techniques of Follmann and Wu (Biometrics 1995; 51:151-168) for modeling informative missing data, we propose a varying-coefficient mixed-effects model to evaluate the patient's longitudinal outcome trends before and after the patient's starting time of the intervention. By allowing the random coefficients to be correlated with the patient's starting time of the intervention, our model leads to less biased estimates of the intervention effects. Nonparametric estimation and inferences of the coefficient curves and intervention effects are developed using B-splines. Our methods are demonstrated through a longitudinal clinical trial in depression and heart disease and a simulation study.

AB - Concomitant interventions are often introduced during a longitudinal clinical trial to patients who respond undesirably to the pre-specified treatments. In addition to the main objective of evaluating the pre-specified treatment effects, an important secondary objective in such a trial is to evaluate whether a concomitant intervention could change a patient's response over time. Because the initiation of a concomitant intervention may depend on the patient's general trend of pre-intervention outcomes, regression approaches that treat the presence of the intervention as a time-dependent covariate may lead to biased estimates for the intervention effects. Borrowing the techniques of Follmann and Wu (Biometrics 1995; 51:151-168) for modeling informative missing data, we propose a varying-coefficient mixed-effects model to evaluate the patient's longitudinal outcome trends before and after the patient's starting time of the intervention. By allowing the random coefficients to be correlated with the patient's starting time of the intervention, our model leads to less biased estimates of the intervention effects. Nonparametric estimation and inferences of the coefficient curves and intervention effects are developed using B-splines. Our methods are demonstrated through a longitudinal clinical trial in depression and heart disease and a simulation study.

KW - Change-point models

KW - Concomitant intervention

KW - Longitudinal study

KW - Polynomial splines

KW - Shared-parameter model

KW - Varying-coefficient model

UR - http://www.scopus.com/inward/record.url?scp=48049123902&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=48049123902&partnerID=8YFLogxK

U2 - 10.1002/sim.3262

DO - 10.1002/sim.3262

M3 - Article

VL - 27

SP - 3042

EP - 3056

JO - Statistics in Medicine

JF - Statistics in Medicine

SN - 0277-6715

IS - 16

ER -