Sample size requirement to detect an intervention effect at the end of follow-up in a longitudinal cluster randomized trial

Research output: Contribution to journalArticle

14 Citations (Scopus)

Abstract

It is often anticipated in a longitudinal cluster randomized clinical trial (cluster-RCT) that the course of outcome over time will diverge between intervention arms. In these situations, testing the significance of a local intervention effect at the end of the trial may be more clinically relevant than evaluating overall mean differences between treatment groups. In this paper, we present a closed-form power function for detecting this local intervention effect based on maximum likelihood estimates from a mixed-effects linear regression model for three-level continuous data. Sample size requirements for the number of units at each data level are derived from the power function. The power function and the corresponding sample size requirements are verified by a simulation study. Importantly, it is shown that sample size requirements computed with the proposed power function are smaller than that required when testing group mean difference using data only at the end of trial and ignoring the course of outcome over the entire study period.

Original languageEnglish (US)
Pages (from-to)382-390
Number of pages9
JournalStatistics in Medicine
Volume29
Issue number3
DOIs
StatePublished - Feb 10 2010

Fingerprint

Randomized Trial
Power Function
Sample Size
Linear Models
Requirements
Likelihood Functions
Mixed Effects
Group Testing
Randomized Clinical Trial
Randomized Controlled Trials
Maximum Likelihood Estimate
Linear Regression Model
Diverge
Closed-form
Simulation Study
Entire
Testing
Unit

Keywords

  • Intervention effect size
  • Longitudinal cluster RCT
  • Power function
  • Sample size
  • Three-level data

ASJC Scopus subject areas

  • Epidemiology
  • Statistics and Probability

Cite this

@article{592f4c5fb55f4e2a869a91df2621a359,
title = "Sample size requirement to detect an intervention effect at the end of follow-up in a longitudinal cluster randomized trial",
abstract = "It is often anticipated in a longitudinal cluster randomized clinical trial (cluster-RCT) that the course of outcome over time will diverge between intervention arms. In these situations, testing the significance of a local intervention effect at the end of the trial may be more clinically relevant than evaluating overall mean differences between treatment groups. In this paper, we present a closed-form power function for detecting this local intervention effect based on maximum likelihood estimates from a mixed-effects linear regression model for three-level continuous data. Sample size requirements for the number of units at each data level are derived from the power function. The power function and the corresponding sample size requirements are verified by a simulation study. Importantly, it is shown that sample size requirements computed with the proposed power function are smaller than that required when testing group mean difference using data only at the end of trial and ignoring the course of outcome over the entire study period.",
keywords = "Intervention effect size, Longitudinal cluster RCT, Power function, Sample size, Three-level data",
author = "Moonseong Heo and Yongman Kim and Xue, {Xiaonan (Nan)} and Mimi Kim",
year = "2010",
month = "2",
day = "10",
doi = "10.1002/sim.3806",
language = "English (US)",
volume = "29",
pages = "382--390",
journal = "Statistics in Medicine",
issn = "0277-6715",
publisher = "John Wiley and Sons Ltd",
number = "3",

}

TY - JOUR

T1 - Sample size requirement to detect an intervention effect at the end of follow-up in a longitudinal cluster randomized trial

AU - Heo, Moonseong

AU - Kim, Yongman

AU - Xue, Xiaonan (Nan)

AU - Kim, Mimi

PY - 2010/2/10

Y1 - 2010/2/10

N2 - It is often anticipated in a longitudinal cluster randomized clinical trial (cluster-RCT) that the course of outcome over time will diverge between intervention arms. In these situations, testing the significance of a local intervention effect at the end of the trial may be more clinically relevant than evaluating overall mean differences between treatment groups. In this paper, we present a closed-form power function for detecting this local intervention effect based on maximum likelihood estimates from a mixed-effects linear regression model for three-level continuous data. Sample size requirements for the number of units at each data level are derived from the power function. The power function and the corresponding sample size requirements are verified by a simulation study. Importantly, it is shown that sample size requirements computed with the proposed power function are smaller than that required when testing group mean difference using data only at the end of trial and ignoring the course of outcome over the entire study period.

AB - It is often anticipated in a longitudinal cluster randomized clinical trial (cluster-RCT) that the course of outcome over time will diverge between intervention arms. In these situations, testing the significance of a local intervention effect at the end of the trial may be more clinically relevant than evaluating overall mean differences between treatment groups. In this paper, we present a closed-form power function for detecting this local intervention effect based on maximum likelihood estimates from a mixed-effects linear regression model for three-level continuous data. Sample size requirements for the number of units at each data level are derived from the power function. The power function and the corresponding sample size requirements are verified by a simulation study. Importantly, it is shown that sample size requirements computed with the proposed power function are smaller than that required when testing group mean difference using data only at the end of trial and ignoring the course of outcome over the entire study period.

KW - Intervention effect size

KW - Longitudinal cluster RCT

KW - Power function

KW - Sample size

KW - Three-level data

UR - http://www.scopus.com/inward/record.url?scp=74749097175&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=74749097175&partnerID=8YFLogxK

U2 - 10.1002/sim.3806

DO - 10.1002/sim.3806

M3 - Article

VL - 29

SP - 382

EP - 390

JO - Statistics in Medicine

JF - Statistics in Medicine

SN - 0277-6715

IS - 3

ER -