The split-plot design was useful for evaluating complex, multi-level interventions but there is need for improvement in its design and report

Research output: Contribution to journalReview article

1 Citation (Scopus)
4 Downloads (Pure)

Abstract

OBJECTIVE: To describe the sample size calculation, analysis and reporting of split-plot randomised controlled trials (RCT) in healthcare (trials that use two units of randomisation: one at a cluster-level and one at a level lower than the cluster).

STUDY DESIGN AND SETTING: We carried out a comprehensive search in the EMBASE database from 1946 to 2016. Healthcare trials with a split-plot design in human subjects were included. Three authors screened and assessed the studies and data were extracted on methodology and reporting standards based upon CONSORT.

RESULTS: 18 split-plot studies were included, with authors using nine different designations to describe them. Units of randomisation were unclear in nine abstracts. Explicit rationale for choosing the design was not givenTen studies presented a sample size calculation accounting for clustering; the analyses were coherent with that. Flow of participant diagrams were presented but incomplete in 14 articles.

CONCLUSION: Split-plot designs can be useful complex designs, but challenging to report. Researchers need to clearly describe the rationale, sample size calculation and participant flow. We provide a suggested CONSORT style participant flow diagram to aid reporting. There is need for more research regarding sample size calculation for split-plots.

Original languageEnglish
Pages (from-to)120-125
Number of pages6
JournalJournal of Clinical Epidemiology
Volume96
Early online date4 Nov 2017
DOIs
Publication statusPublished - Apr 2018

Fingerprint

Sample Size
Random Allocation
Delivery of Health Care
Cluster Analysis
Randomized Controlled Trials
Research Personnel
Databases
Research

Keywords

  • Split-plots
  • Systematic review
  • 2x2
  • Factorial
  • Cluster randomised
  • Randomised controlled trial

Cite this

@article{0a5981ea480b41e0a5b8a3b0c9777bf1,
title = "The split-plot design was useful for evaluating complex, multi-level interventions but there is need for improvement in its design and report",
abstract = "OBJECTIVE: To describe the sample size calculation, analysis and reporting of split-plot randomised controlled trials (RCT) in healthcare (trials that use two units of randomisation: one at a cluster-level and one at a level lower than the cluster).STUDY DESIGN AND SETTING: We carried out a comprehensive search in the EMBASE database from 1946 to 2016. Healthcare trials with a split-plot design in human subjects were included. Three authors screened and assessed the studies and data were extracted on methodology and reporting standards based upon CONSORT.RESULTS: 18 split-plot studies were included, with authors using nine different designations to describe them. Units of randomisation were unclear in nine abstracts. Explicit rationale for choosing the design was not givenTen studies presented a sample size calculation accounting for clustering; the analyses were coherent with that. Flow of participant diagrams were presented but incomplete in 14 articles.CONCLUSION: Split-plot designs can be useful complex designs, but challenging to report. Researchers need to clearly describe the rationale, sample size calculation and participant flow. We provide a suggested CONSORT style participant flow diagram to aid reporting. There is need for more research regarding sample size calculation for split-plots.",
keywords = "Split-plots, Systematic review, 2x2, Factorial, Cluster randomised, Randomised controlled trial",
author = "Beatriz Goul{\~a}o and Graeme MacLennan and Craig Ramsay",
note = "Copyright {\circledC} 2017 Elsevier Inc. All rights reserved.",
year = "2018",
month = "4",
doi = "10.1016/j.jclinepi.2017.10.019",
language = "English",
volume = "96",
pages = "120--125",
journal = "Journal of Clinical Epidemiology",
issn = "0895-4356",
publisher = "Elsevier USA",

}

TY - JOUR

T1 - The split-plot design was useful for evaluating complex, multi-level interventions but there is need for improvement in its design and report

AU - Goulão, Beatriz

AU - MacLennan, Graeme

AU - Ramsay, Craig

N1 - Copyright © 2017 Elsevier Inc. All rights reserved.

PY - 2018/4

Y1 - 2018/4

N2 - OBJECTIVE: To describe the sample size calculation, analysis and reporting of split-plot randomised controlled trials (RCT) in healthcare (trials that use two units of randomisation: one at a cluster-level and one at a level lower than the cluster).STUDY DESIGN AND SETTING: We carried out a comprehensive search in the EMBASE database from 1946 to 2016. Healthcare trials with a split-plot design in human subjects were included. Three authors screened and assessed the studies and data were extracted on methodology and reporting standards based upon CONSORT.RESULTS: 18 split-plot studies were included, with authors using nine different designations to describe them. Units of randomisation were unclear in nine abstracts. Explicit rationale for choosing the design was not givenTen studies presented a sample size calculation accounting for clustering; the analyses were coherent with that. Flow of participant diagrams were presented but incomplete in 14 articles.CONCLUSION: Split-plot designs can be useful complex designs, but challenging to report. Researchers need to clearly describe the rationale, sample size calculation and participant flow. We provide a suggested CONSORT style participant flow diagram to aid reporting. There is need for more research regarding sample size calculation for split-plots.

AB - OBJECTIVE: To describe the sample size calculation, analysis and reporting of split-plot randomised controlled trials (RCT) in healthcare (trials that use two units of randomisation: one at a cluster-level and one at a level lower than the cluster).STUDY DESIGN AND SETTING: We carried out a comprehensive search in the EMBASE database from 1946 to 2016. Healthcare trials with a split-plot design in human subjects were included. Three authors screened and assessed the studies and data were extracted on methodology and reporting standards based upon CONSORT.RESULTS: 18 split-plot studies were included, with authors using nine different designations to describe them. Units of randomisation were unclear in nine abstracts. Explicit rationale for choosing the design was not givenTen studies presented a sample size calculation accounting for clustering; the analyses were coherent with that. Flow of participant diagrams were presented but incomplete in 14 articles.CONCLUSION: Split-plot designs can be useful complex designs, but challenging to report. Researchers need to clearly describe the rationale, sample size calculation and participant flow. We provide a suggested CONSORT style participant flow diagram to aid reporting. There is need for more research regarding sample size calculation for split-plots.

KW - Split-plots

KW - Systematic review

KW - 2x2

KW - Factorial

KW - Cluster randomised

KW - Randomised controlled trial

U2 - 10.1016/j.jclinepi.2017.10.019

DO - 10.1016/j.jclinepi.2017.10.019

M3 - Review article

VL - 96

SP - 120

EP - 125

JO - Journal of Clinical Epidemiology

JF - Journal of Clinical Epidemiology

SN - 0895-4356

ER -