Single time point comparisons in longitudinal randomized controlled trials: Power and bias in the presence of missing data

Erin L. Ashbeck, Melanie L Bell

Research output: Contribution to journalArticle

10 Citations (Scopus)

Abstract

Background: The primary analysis in a longitudinal randomized controlled trial is sometimes a comparison of arms at a single time point. While a two-sample t-test is often used, missing data are common in longitudinal studies and decreases power by reducing sample size. Mixed models for repeated measures (MMRM) can test treatment effects at specific time points, have been shown to give unbiased estimates in certain missing data contexts, and may be more powerful than a two sample t-test. Methods: We conducted a simulation study to compare the performance of a complete-case t-test to a MMRM in terms of power and bias under different missing data mechanisms. Impact of within- and between-person variance, dropout mechanism, and variance-covariance structure were all considered. Results: While both complete-case t-test and MMRM provided unbiased estimation of treatment differences when data were missing completely at random, MMRM yielded an absolute power gain of up to 12 %. The MMRM provided up to 25 % absolute increased power over the t-test when data were missing at random, as well as unbiased estimation. Conclusions: Investigators interested in single time point comparisons should use a MMRM with a contrast to gain power and unbiased estimation of treatment effects instead of a complete-case two sample t-test.

Original languageEnglish (US)
Article number43
JournalBMC Medical Research Methodology
Volume16
Issue number1
DOIs
StatePublished - Apr 12 2016

Fingerprint

Randomized Controlled Trials
Sample Size
Longitudinal Studies
Research Personnel

Keywords

  • Complete-case
  • Longitudinal
  • Mean response profile
  • Missing data
  • Mixed model
  • Power
  • Repeated measures
  • T-test

ASJC Scopus subject areas

  • Health Informatics
  • Epidemiology

Cite this

@article{952e9bcd054240ae88159174489eb136,
title = "Single time point comparisons in longitudinal randomized controlled trials: Power and bias in the presence of missing data",
abstract = "Background: The primary analysis in a longitudinal randomized controlled trial is sometimes a comparison of arms at a single time point. While a two-sample t-test is often used, missing data are common in longitudinal studies and decreases power by reducing sample size. Mixed models for repeated measures (MMRM) can test treatment effects at specific time points, have been shown to give unbiased estimates in certain missing data contexts, and may be more powerful than a two sample t-test. Methods: We conducted a simulation study to compare the performance of a complete-case t-test to a MMRM in terms of power and bias under different missing data mechanisms. Impact of within- and between-person variance, dropout mechanism, and variance-covariance structure were all considered. Results: While both complete-case t-test and MMRM provided unbiased estimation of treatment differences when data were missing completely at random, MMRM yielded an absolute power gain of up to 12 {\%}. The MMRM provided up to 25 {\%} absolute increased power over the t-test when data were missing at random, as well as unbiased estimation. Conclusions: Investigators interested in single time point comparisons should use a MMRM with a contrast to gain power and unbiased estimation of treatment effects instead of a complete-case two sample t-test.",
keywords = "Complete-case, Longitudinal, Mean response profile, Missing data, Mixed model, Power, Repeated measures, T-test",
author = "Ashbeck, {Erin L.} and Bell, {Melanie L}",
year = "2016",
month = "4",
day = "12",
doi = "10.1186/s12874-016-0144-0",
language = "English (US)",
volume = "16",
journal = "BMC Medical Research Methodology",
issn = "1471-2288",
publisher = "BioMed Central",
number = "1",

}

TY - JOUR

T1 - Single time point comparisons in longitudinal randomized controlled trials

T2 - Power and bias in the presence of missing data

AU - Ashbeck, Erin L.

AU - Bell, Melanie L

PY - 2016/4/12

Y1 - 2016/4/12

N2 - Background: The primary analysis in a longitudinal randomized controlled trial is sometimes a comparison of arms at a single time point. While a two-sample t-test is often used, missing data are common in longitudinal studies and decreases power by reducing sample size. Mixed models for repeated measures (MMRM) can test treatment effects at specific time points, have been shown to give unbiased estimates in certain missing data contexts, and may be more powerful than a two sample t-test. Methods: We conducted a simulation study to compare the performance of a complete-case t-test to a MMRM in terms of power and bias under different missing data mechanisms. Impact of within- and between-person variance, dropout mechanism, and variance-covariance structure were all considered. Results: While both complete-case t-test and MMRM provided unbiased estimation of treatment differences when data were missing completely at random, MMRM yielded an absolute power gain of up to 12 %. The MMRM provided up to 25 % absolute increased power over the t-test when data were missing at random, as well as unbiased estimation. Conclusions: Investigators interested in single time point comparisons should use a MMRM with a contrast to gain power and unbiased estimation of treatment effects instead of a complete-case two sample t-test.

AB - Background: The primary analysis in a longitudinal randomized controlled trial is sometimes a comparison of arms at a single time point. While a two-sample t-test is often used, missing data are common in longitudinal studies and decreases power by reducing sample size. Mixed models for repeated measures (MMRM) can test treatment effects at specific time points, have been shown to give unbiased estimates in certain missing data contexts, and may be more powerful than a two sample t-test. Methods: We conducted a simulation study to compare the performance of a complete-case t-test to a MMRM in terms of power and bias under different missing data mechanisms. Impact of within- and between-person variance, dropout mechanism, and variance-covariance structure were all considered. Results: While both complete-case t-test and MMRM provided unbiased estimation of treatment differences when data were missing completely at random, MMRM yielded an absolute power gain of up to 12 %. The MMRM provided up to 25 % absolute increased power over the t-test when data were missing at random, as well as unbiased estimation. Conclusions: Investigators interested in single time point comparisons should use a MMRM with a contrast to gain power and unbiased estimation of treatment effects instead of a complete-case two sample t-test.

KW - Complete-case

KW - Longitudinal

KW - Mean response profile

KW - Missing data

KW - Mixed model

KW - Power

KW - Repeated measures

KW - T-test

UR - http://www.scopus.com/inward/record.url?scp=84963804585&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84963804585&partnerID=8YFLogxK

U2 - 10.1186/s12874-016-0144-0

DO - 10.1186/s12874-016-0144-0

M3 - Article

C2 - 27068578

AN - SCOPUS:84963804585

VL - 16

JO - BMC Medical Research Methodology

JF - BMC Medical Research Methodology

SN - 1471-2288

IS - 1

M1 - 43

ER -