Cargando…

Comparison of small-sample standard-error corrections for generalised estimating equations in stepped wedge cluster randomised trials with a binary outcome: A simulation study

Generalised estimating equations with the sandwich standard-error estimator provide a promising method of analysis for stepped wedge cluster randomised trials. However, they have inflated type-one error when used with a small number of clusters, which is common for stepped wedge cluster randomised t...

Descripción completa

Detalles Bibliográficos
Autores principales: Thompson, JA, Hemming, K, Forbes, A, Fielding, K, Hayes, R
Formato: Online Artículo Texto
Lenguaje:English
Publicado: SAGE Publications 2020
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8008420/
https://www.ncbi.nlm.nih.gov/pubmed/32970526
http://dx.doi.org/10.1177/0962280220958735
_version_ 1783672691589382144
author Thompson, JA
Hemming, K
Forbes, A
Fielding, K
Hayes, R
author_facet Thompson, JA
Hemming, K
Forbes, A
Fielding, K
Hayes, R
author_sort Thompson, JA
collection PubMed
description Generalised estimating equations with the sandwich standard-error estimator provide a promising method of analysis for stepped wedge cluster randomised trials. However, they have inflated type-one error when used with a small number of clusters, which is common for stepped wedge cluster randomised trials. We present a large simulation study of binary outcomes comparing bias-corrected standard errors from Fay and Graubard; Mancl and DeRouen; Kauermann and Carroll; Morel, Bokossa, and Neerchal; and Mackinnon and White with an independent and exchangeable working correlation matrix. We constructed 95% confidence intervals using a t-distribution with degrees of freedom including clusters minus parameters (DF(C-P)), cluster periods minus parameters, and estimators from Fay and Graubard (DF(FG)), and Pan and Wall. Fay and Graubard and an approximation to Kauermann and Carroll (with simpler matrix inversion) were unbiased in a wide range of scenarios with an independent working correlation matrix and more than 12 clusters. They gave confidence intervals with close to 95% coverage with DF(FG) with 12 or more clusters, and DF(C-P) with 18 or more clusters. Both standard errors were conservative with fewer clusters. With an exchangeable working correlation matrix, approximated Kauermann and Carroll and Fay and Graubard had a small degree of under-coverage.
format Online
Article
Text
id pubmed-8008420
institution National Center for Biotechnology Information
language English
publishDate 2020
publisher SAGE Publications
record_format MEDLINE/PubMed
spelling pubmed-80084202021-04-08 Comparison of small-sample standard-error corrections for generalised estimating equations in stepped wedge cluster randomised trials with a binary outcome: A simulation study Thompson, JA Hemming, K Forbes, A Fielding, K Hayes, R Stat Methods Med Res Articles Generalised estimating equations with the sandwich standard-error estimator provide a promising method of analysis for stepped wedge cluster randomised trials. However, they have inflated type-one error when used with a small number of clusters, which is common for stepped wedge cluster randomised trials. We present a large simulation study of binary outcomes comparing bias-corrected standard errors from Fay and Graubard; Mancl and DeRouen; Kauermann and Carroll; Morel, Bokossa, and Neerchal; and Mackinnon and White with an independent and exchangeable working correlation matrix. We constructed 95% confidence intervals using a t-distribution with degrees of freedom including clusters minus parameters (DF(C-P)), cluster periods minus parameters, and estimators from Fay and Graubard (DF(FG)), and Pan and Wall. Fay and Graubard and an approximation to Kauermann and Carroll (with simpler matrix inversion) were unbiased in a wide range of scenarios with an independent working correlation matrix and more than 12 clusters. They gave confidence intervals with close to 95% coverage with DF(FG) with 12 or more clusters, and DF(C-P) with 18 or more clusters. Both standard errors were conservative with fewer clusters. With an exchangeable working correlation matrix, approximated Kauermann and Carroll and Fay and Graubard had a small degree of under-coverage. SAGE Publications 2020-09-24 2021-02 /pmc/articles/PMC8008420/ /pubmed/32970526 http://dx.doi.org/10.1177/0962280220958735 Text en © The Author(s) 2020 https://creativecommons.org/licenses/by/4.0/ Creative Commons CC BY: This article is distributed under the terms of the Creative Commons Attribution 4.0 License (https://creativecommons.org/licenses/by/4.0/) which permits any use, reproduction and distribution of the work without further permission provided the original work is attributed as specified on the SAGE and Open Access pages (https://us.sagepub.com/en-us/nam/open-access-at-sage).
spellingShingle Articles
Thompson, JA
Hemming, K
Forbes, A
Fielding, K
Hayes, R
Comparison of small-sample standard-error corrections for generalised estimating equations in stepped wedge cluster randomised trials with a binary outcome: A simulation study
title Comparison of small-sample standard-error corrections for generalised estimating equations in stepped wedge cluster randomised trials with a binary outcome: A simulation study
title_full Comparison of small-sample standard-error corrections for generalised estimating equations in stepped wedge cluster randomised trials with a binary outcome: A simulation study
title_fullStr Comparison of small-sample standard-error corrections for generalised estimating equations in stepped wedge cluster randomised trials with a binary outcome: A simulation study
title_full_unstemmed Comparison of small-sample standard-error corrections for generalised estimating equations in stepped wedge cluster randomised trials with a binary outcome: A simulation study
title_short Comparison of small-sample standard-error corrections for generalised estimating equations in stepped wedge cluster randomised trials with a binary outcome: A simulation study
title_sort comparison of small-sample standard-error corrections for generalised estimating equations in stepped wedge cluster randomised trials with a binary outcome: a simulation study
topic Articles
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8008420/
https://www.ncbi.nlm.nih.gov/pubmed/32970526
http://dx.doi.org/10.1177/0962280220958735
work_keys_str_mv AT thompsonja comparisonofsmallsamplestandarderrorcorrectionsforgeneralisedestimatingequationsinsteppedwedgeclusterrandomisedtrialswithabinaryoutcomeasimulationstudy
AT hemmingk comparisonofsmallsamplestandarderrorcorrectionsforgeneralisedestimatingequationsinsteppedwedgeclusterrandomisedtrialswithabinaryoutcomeasimulationstudy
AT forbesa comparisonofsmallsamplestandarderrorcorrectionsforgeneralisedestimatingequationsinsteppedwedgeclusterrandomisedtrialswithabinaryoutcomeasimulationstudy
AT fieldingk comparisonofsmallsamplestandarderrorcorrectionsforgeneralisedestimatingequationsinsteppedwedgeclusterrandomisedtrialswithabinaryoutcomeasimulationstudy
AT hayesr comparisonofsmallsamplestandarderrorcorrectionsforgeneralisedestimatingequationsinsteppedwedgeclusterrandomisedtrialswithabinaryoutcomeasimulationstudy