Assessing intervention effects in a school-based nutrition intervention trial

Which analytic model is most powerful?

Jessica B. Janega, David M. Murray, Sherri P. Varnell, Jonathan L. Blitstein, Amanda Birnbaum, Leslie A. Lytle

Research output: Contribution to journalArticleResearchpeer-review

18 Citations (Scopus)

Abstract

This article compares four mixed-model analyses valid for group-randomized trials (GRTs) involving a nested cohort design with a single pretest and posttest. This study makes estimates of intraclass correlations (ICCs) available to investigators planning GRTs addressing dietary outcomes. It also provides formulae demonstrating the potential benefits to the standard error of the intervention effect (σ Δ) from adjustments for both fixed and time-varying covariates and correlations over time. These estimates will allow other researchers using these variables to plan their studies by estimating a priori detectable differences and sample size requirements for any of the four analytic options. These methods are demonstrated using data from the Teens Eating for Energy and Nutrition at School study. Mixed-model analyses of covariance proved to be the most powerful analysis in that data set. The formulae may be applied to any dependent variable in any GRT given corresponding information for those variables on the parameters that define the formulae.

Original languageEnglish
Pages (from-to)756-774
Number of pages19
JournalHealth Education and Behavior
Volume31
Issue number6
DOIs
StatePublished - 1 Dec 2004

Fingerprint

Research Personnel
Sample Size
Eating
Nutrition
Datasets
Planning
Posttests
Cohort
Energy

Keywords

  • Analytic methods
  • Group-randomized trial
  • Power

Cite this

Janega, Jessica B. ; Murray, David M. ; Varnell, Sherri P. ; Blitstein, Jonathan L. ; Birnbaum, Amanda ; Lytle, Leslie A. / Assessing intervention effects in a school-based nutrition intervention trial : Which analytic model is most powerful?. In: Health Education and Behavior. 2004 ; Vol. 31, No. 6. pp. 756-774.
@article{1cb3ef106fb7491c82b7d255150d7cbc,
title = "Assessing intervention effects in a school-based nutrition intervention trial: Which analytic model is most powerful?",
abstract = "This article compares four mixed-model analyses valid for group-randomized trials (GRTs) involving a nested cohort design with a single pretest and posttest. This study makes estimates of intraclass correlations (ICCs) available to investigators planning GRTs addressing dietary outcomes. It also provides formulae demonstrating the potential benefits to the standard error of the intervention effect (σ Δ) from adjustments for both fixed and time-varying covariates and correlations over time. These estimates will allow other researchers using these variables to plan their studies by estimating a priori detectable differences and sample size requirements for any of the four analytic options. These methods are demonstrated using data from the Teens Eating for Energy and Nutrition at School study. Mixed-model analyses of covariance proved to be the most powerful analysis in that data set. The formulae may be applied to any dependent variable in any GRT given corresponding information for those variables on the parameters that define the formulae.",
keywords = "Analytic methods, Group-randomized trial, Power",
author = "Janega, {Jessica B.} and Murray, {David M.} and Varnell, {Sherri P.} and Blitstein, {Jonathan L.} and Amanda Birnbaum and Lytle, {Leslie A.}",
year = "2004",
month = "12",
day = "1",
doi = "10.1177/1090198104263406",
language = "English",
volume = "31",
pages = "756--774",
journal = "Health Education and Behavior",
issn = "1090-1981",
publisher = "SAGE Publications Inc.",
number = "6",

}

Assessing intervention effects in a school-based nutrition intervention trial : Which analytic model is most powerful? / Janega, Jessica B.; Murray, David M.; Varnell, Sherri P.; Blitstein, Jonathan L.; Birnbaum, Amanda; Lytle, Leslie A.

In: Health Education and Behavior, Vol. 31, No. 6, 01.12.2004, p. 756-774.

Research output: Contribution to journalArticleResearchpeer-review

TY - JOUR

T1 - Assessing intervention effects in a school-based nutrition intervention trial

T2 - Which analytic model is most powerful?

AU - Janega, Jessica B.

AU - Murray, David M.

AU - Varnell, Sherri P.

AU - Blitstein, Jonathan L.

AU - Birnbaum, Amanda

AU - Lytle, Leslie A.

PY - 2004/12/1

Y1 - 2004/12/1

N2 - This article compares four mixed-model analyses valid for group-randomized trials (GRTs) involving a nested cohort design with a single pretest and posttest. This study makes estimates of intraclass correlations (ICCs) available to investigators planning GRTs addressing dietary outcomes. It also provides formulae demonstrating the potential benefits to the standard error of the intervention effect (σ Δ) from adjustments for both fixed and time-varying covariates and correlations over time. These estimates will allow other researchers using these variables to plan their studies by estimating a priori detectable differences and sample size requirements for any of the four analytic options. These methods are demonstrated using data from the Teens Eating for Energy and Nutrition at School study. Mixed-model analyses of covariance proved to be the most powerful analysis in that data set. The formulae may be applied to any dependent variable in any GRT given corresponding information for those variables on the parameters that define the formulae.

AB - This article compares four mixed-model analyses valid for group-randomized trials (GRTs) involving a nested cohort design with a single pretest and posttest. This study makes estimates of intraclass correlations (ICCs) available to investigators planning GRTs addressing dietary outcomes. It also provides formulae demonstrating the potential benefits to the standard error of the intervention effect (σ Δ) from adjustments for both fixed and time-varying covariates and correlations over time. These estimates will allow other researchers using these variables to plan their studies by estimating a priori detectable differences and sample size requirements for any of the four analytic options. These methods are demonstrated using data from the Teens Eating for Energy and Nutrition at School study. Mixed-model analyses of covariance proved to be the most powerful analysis in that data set. The formulae may be applied to any dependent variable in any GRT given corresponding information for those variables on the parameters that define the formulae.

KW - Analytic methods

KW - Group-randomized trial

KW - Power

UR - http://www.scopus.com/inward/record.url?scp=9944239802&partnerID=8YFLogxK

U2 - 10.1177/1090198104263406

DO - 10.1177/1090198104263406

M3 - Article

VL - 31

SP - 756

EP - 774

JO - Health Education and Behavior

JF - Health Education and Behavior

SN - 1090-1981

IS - 6

ER -