Methods for testing theory and evaluating impact in randomized field trials: Intent-to-treat analyses for integrating the perspectives of person, place, and time

C. Hendricks Brown, Wei Wang, Sheppard G. Kellam, Bengt O. Muthén, Hanno Petras, Peter Toyinbo, Jeanne Poduska, Nicholas Ialongo, Peter A. Wyman, Patricia Chamberlain, Zili Sloboda, David Mackinnon, Amy Windham

Research output: Contribution to journalArticle

118 Citations (Scopus)

Abstract

Randomized field trials provide unique opportunities to examine the effectiveness of an intervention in real world settings and to test and extend both theory of etiology and theory of intervention. These trials are designed not only to test for overall intervention impact but also to examine how impact varies as a function of individual level characteristics, context, and across time. Examination of such variation in impact requires analytical methods that take into account the trial's multiple nested structure and the evolving changes in outcomes over time. The models that we describe here merge multilevel modeling with growth modeling, allowing for variation in impact to be represented through discrete mixtures-growth mixture models-and nonparametric smooth functions-generalized additive mixed models. These methods are part of an emerging class of multilevel growth mixture models, and we illustrate these with models that examine overall impact and variation in impact. In this paper, we define intent-to-treat analyses in group-randomized multilevel field trials and discuss appropriate ways to identify, examine, and test for variation in impact without inflating the Type I error rate. We describe how to make causal inferences more robust to misspecification of covariates in such analyses and how to summarize and present these interactive intervention effects clearly. Practical strategies for reducing model complexity, checking model fit, and handling missing data are discussed using six randomized field trials to show how these methods may be used across trials randomized at different levels.

Original languageEnglish (US)
JournalDrug and Alcohol Dependence
Volume95
Issue numberSUPPL. 1
DOIs
StatePublished - Jun 1 2008

Fingerprint

human being
Testing
Growth
Data handling
Model checking
time
etiology
examination
Group

Keywords

  • Additive models
  • Developmental epidemiology
  • Group-randomized trials
  • Growth models
  • Intent-to-treat analysis
  • Mediation
  • Mixture models
  • Moderation
  • Multilevel models
  • Prevention
  • Random effect models

ASJC Scopus subject areas

  • Medicine(all)
  • Behavioral Neuroscience
  • Toxicology
  • Health(social science)

Cite this

Methods for testing theory and evaluating impact in randomized field trials : Intent-to-treat analyses for integrating the perspectives of person, place, and time. / Brown, C. Hendricks; Wang, Wei; Kellam, Sheppard G.; Muthén, Bengt O.; Petras, Hanno; Toyinbo, Peter; Poduska, Jeanne; Ialongo, Nicholas; Wyman, Peter A.; Chamberlain, Patricia; Sloboda, Zili; Mackinnon, David; Windham, Amy.

In: Drug and Alcohol Dependence, Vol. 95, No. SUPPL. 1, 01.06.2008.

Research output: Contribution to journalArticle

Brown, CH, Wang, W, Kellam, SG, Muthén, BO, Petras, H, Toyinbo, P, Poduska, J, Ialongo, N, Wyman, PA, Chamberlain, P, Sloboda, Z, Mackinnon, D & Windham, A 2008, 'Methods for testing theory and evaluating impact in randomized field trials: Intent-to-treat analyses for integrating the perspectives of person, place, and time', Drug and Alcohol Dependence, vol. 95, no. SUPPL. 1. https://doi.org/10.1016/j.drugalcdep.2007.11.013
Brown, C. Hendricks ; Wang, Wei ; Kellam, Sheppard G. ; Muthén, Bengt O. ; Petras, Hanno ; Toyinbo, Peter ; Poduska, Jeanne ; Ialongo, Nicholas ; Wyman, Peter A. ; Chamberlain, Patricia ; Sloboda, Zili ; Mackinnon, David ; Windham, Amy. / Methods for testing theory and evaluating impact in randomized field trials : Intent-to-treat analyses for integrating the perspectives of person, place, and time. In: Drug and Alcohol Dependence. 2008 ; Vol. 95, No. SUPPL. 1.
@article{407c6c75f1cb4e869377fc431acbda51,
title = "Methods for testing theory and evaluating impact in randomized field trials: Intent-to-treat analyses for integrating the perspectives of person, place, and time",
abstract = "Randomized field trials provide unique opportunities to examine the effectiveness of an intervention in real world settings and to test and extend both theory of etiology and theory of intervention. These trials are designed not only to test for overall intervention impact but also to examine how impact varies as a function of individual level characteristics, context, and across time. Examination of such variation in impact requires analytical methods that take into account the trial's multiple nested structure and the evolving changes in outcomes over time. The models that we describe here merge multilevel modeling with growth modeling, allowing for variation in impact to be represented through discrete mixtures-growth mixture models-and nonparametric smooth functions-generalized additive mixed models. These methods are part of an emerging class of multilevel growth mixture models, and we illustrate these with models that examine overall impact and variation in impact. In this paper, we define intent-to-treat analyses in group-randomized multilevel field trials and discuss appropriate ways to identify, examine, and test for variation in impact without inflating the Type I error rate. We describe how to make causal inferences more robust to misspecification of covariates in such analyses and how to summarize and present these interactive intervention effects clearly. Practical strategies for reducing model complexity, checking model fit, and handling missing data are discussed using six randomized field trials to show how these methods may be used across trials randomized at different levels.",
keywords = "Additive models, Developmental epidemiology, Group-randomized trials, Growth models, Intent-to-treat analysis, Mediation, Mixture models, Moderation, Multilevel models, Prevention, Random effect models",
author = "Brown, {C. Hendricks} and Wei Wang and Kellam, {Sheppard G.} and Muth{\'e}n, {Bengt O.} and Hanno Petras and Peter Toyinbo and Jeanne Poduska and Nicholas Ialongo and Wyman, {Peter A.} and Patricia Chamberlain and Zili Sloboda and David Mackinnon and Amy Windham",
year = "2008",
month = "6",
day = "1",
doi = "10.1016/j.drugalcdep.2007.11.013",
language = "English (US)",
volume = "95",
journal = "Drug and Alcohol Dependence",
issn = "0376-8716",
publisher = "Elsevier Ireland Ltd",
number = "SUPPL. 1",

}

TY - JOUR

T1 - Methods for testing theory and evaluating impact in randomized field trials

T2 - Intent-to-treat analyses for integrating the perspectives of person, place, and time

AU - Brown, C. Hendricks

AU - Wang, Wei

AU - Kellam, Sheppard G.

AU - Muthén, Bengt O.

AU - Petras, Hanno

AU - Toyinbo, Peter

AU - Poduska, Jeanne

AU - Ialongo, Nicholas

AU - Wyman, Peter A.

AU - Chamberlain, Patricia

AU - Sloboda, Zili

AU - Mackinnon, David

AU - Windham, Amy

PY - 2008/6/1

Y1 - 2008/6/1

N2 - Randomized field trials provide unique opportunities to examine the effectiveness of an intervention in real world settings and to test and extend both theory of etiology and theory of intervention. These trials are designed not only to test for overall intervention impact but also to examine how impact varies as a function of individual level characteristics, context, and across time. Examination of such variation in impact requires analytical methods that take into account the trial's multiple nested structure and the evolving changes in outcomes over time. The models that we describe here merge multilevel modeling with growth modeling, allowing for variation in impact to be represented through discrete mixtures-growth mixture models-and nonparametric smooth functions-generalized additive mixed models. These methods are part of an emerging class of multilevel growth mixture models, and we illustrate these with models that examine overall impact and variation in impact. In this paper, we define intent-to-treat analyses in group-randomized multilevel field trials and discuss appropriate ways to identify, examine, and test for variation in impact without inflating the Type I error rate. We describe how to make causal inferences more robust to misspecification of covariates in such analyses and how to summarize and present these interactive intervention effects clearly. Practical strategies for reducing model complexity, checking model fit, and handling missing data are discussed using six randomized field trials to show how these methods may be used across trials randomized at different levels.

AB - Randomized field trials provide unique opportunities to examine the effectiveness of an intervention in real world settings and to test and extend both theory of etiology and theory of intervention. These trials are designed not only to test for overall intervention impact but also to examine how impact varies as a function of individual level characteristics, context, and across time. Examination of such variation in impact requires analytical methods that take into account the trial's multiple nested structure and the evolving changes in outcomes over time. The models that we describe here merge multilevel modeling with growth modeling, allowing for variation in impact to be represented through discrete mixtures-growth mixture models-and nonparametric smooth functions-generalized additive mixed models. These methods are part of an emerging class of multilevel growth mixture models, and we illustrate these with models that examine overall impact and variation in impact. In this paper, we define intent-to-treat analyses in group-randomized multilevel field trials and discuss appropriate ways to identify, examine, and test for variation in impact without inflating the Type I error rate. We describe how to make causal inferences more robust to misspecification of covariates in such analyses and how to summarize and present these interactive intervention effects clearly. Practical strategies for reducing model complexity, checking model fit, and handling missing data are discussed using six randomized field trials to show how these methods may be used across trials randomized at different levels.

KW - Additive models

KW - Developmental epidemiology

KW - Group-randomized trials

KW - Growth models

KW - Intent-to-treat analysis

KW - Mediation

KW - Mixture models

KW - Moderation

KW - Multilevel models

KW - Prevention

KW - Random effect models

UR - http://www.scopus.com/inward/record.url?scp=42049112459&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=42049112459&partnerID=8YFLogxK

U2 - 10.1016/j.drugalcdep.2007.11.013

DO - 10.1016/j.drugalcdep.2007.11.013

M3 - Article

C2 - 18215473

AN - SCOPUS:42049112459

VL - 95

JO - Drug and Alcohol Dependence

JF - Drug and Alcohol Dependence

SN - 0376-8716

IS - SUPPL. 1

ER -