Choosing the target difference ('effect size') for a randomised controlled trial - DELTA2 guidance protocol

Jonathan A. Cook; Steven A. Julious; William Sones; Joanne C. Rothwell; Craig R. Ramsay; Lisa V. Hampson; Richard Emsley; Stephen J. Walters; Catherine Hewitt; Martin Bland; Dean A. Fergusson; Jesse A. Berlin; Doug Altman; Luke D. Vale

doi:10.1186/s13063-017-1969-5

Choosing the target difference ('effect size') for a randomised controlled trial - DELTA² guidance protocol

Jonathan A. Cook^*
, Steven A. Julious
, William Sones
, Joanne C. Rothwell
, Craig R. Ramsay
, Lisa V. Hampson
, Richard Emsley
, Stephen J. Walters
, Catherine Hewitt
, Martin Bland
, Dean A. Fergusson
, Jesse A. Berlin
, Doug Altman
, Luke D. Vale

^*Corresponding author for this work

Research output: Contribution to journal › Article › peer-review

12 Citations (Scopus)

Abstract

Background: A key step in the design of a randomised controlled trial (RCT) is the estimation of the number of participants needed. By far the most common approach is to specify a target difference and then estimate the corresponding sample size; this sample size is chosen to provide reassurance that the trial will have high statistical power to detect such a difference between the randomised groups (at the planned statistical significance level). The sample size has many implications for the conduct of the study, as well as carrying scientific and ethical aspects to its choice. Despite the critical role of the target difference for the primary outcome in the design of an RCT, the manner in which it is determined has received little attention. This article reports the protocol of the Difference ELicitation in TriAls (DELTA²) project, which will produce guidance on the specification and reporting of the target difference for the primary outcome in a sample size calculation for RCTs. Methods/design: The DELTA² project has five components: systematic literature reviews of recent methodological developments (stage 1) and existing funder guidance (stage 2); a Delphi study (stage 3); a 2-day consensus meeting bringing together researchers, funders and patient representatives, as well as one-off engagement sessions at relevant stakeholder meetings (stage 4); and the preparation and dissemination of a guidance document (stage 5). Discussion: Specification of the target difference for the primary outcome is a key component of the design of an RCT. There is a need for better guidance for researchers and funders regarding specification and reporting of this aspect of trial design. The aim of this project is to produce consensus based guidance for researchers and funders.

Original language	English
Article number	271
Number of pages	7
Journal	Trials
Volume	18
Issue number	1
DOIs	https://doi.org/10.1186/s13063-017-1969-5
Publication status	Published - 12 Jun 2017

Bibliographical note

Funding Information:
ation [EME], Health Technology Assessment [HTA], Research for Patient Benefit Programme [RfPB], Programme Grants for Applied Research [PGfAR], Public Health Research [PHR], Invention for Innovation [i4i], Health Services and Delivery Research); MRC (Developmental Pathway Funding Scheme [DPFS], Wellcome Trust [Health Challenge Innovation Fund], Arthritis Research UK, British Heart Foundation, Cancer Research UK [clinical research, new agent, population research]); NIHR Research Design Service; NIHR Statistics Group; and NHS Health Research Authority (HRA).

Funding Information:
BOS: Bristol Online Surveys; CIHR: Canadian Institutes of Health Research; CRUK: Cancer Research UK; CTU: Clinical Trials Tnit; DELTA: Difference ELicitation in TriAls; DPFS: Developmental Pathway Funding Scheme; EME: Efficacy and Mechanism Evaluation; HTA: Health Technology Assessment; HTMR: Hubs for Trials Methodology Research; i4i: Invention for Innovation; MRC: Medical Research Council; NHMRC: National Health and Medical Research Council; NIH: National Institutes of Health; NIHR: National Institute for Health Research; PCORI: Patient-Centered Outcomes Research Institute; PGfAR: Programme Grants for Applied Research; PHR: Public Health Research; PSI: Statisticians in the Pharmaceutical Industry; RCT: Randomised controlled trial; RDS: Research Design Service; SCT: Society for Clinical Trials; UKCRC: UK Clinical Research Collaboration

Funding Information:
We will also review online guidance that has been provided by the relevant UK trial funding schemes run by the National Institute for Health Research (NIHR), including EME, Health Technology Assessment (HTA), the Research for Patient Benefit Programme, Programme Grants for Applied Research (PGfAR), Public Health Research (PHR), Invention for Innovation (i4i), and Health Services and Delivery Research; the Medical Research Council (MRC) Developmental Pathway Funding Scheme (DPFS); the Wellcome Trust (Health Challenge Innovation Fund); and Cancer Research UK (CRUK) (phase III clinical trial, new agent, population research). We will also review any guidance documents relating to sample size specification provided by the NIHR Research Design Service (RDS). Online guidance documents will be reviewed with individual schemed contact to provide clarification where necessary. We will also review guidance provided by leading international funding streams (National Institutes of Health [NIH], Patient-Centered Outcomes Research Institute [PCORI], Canadian Institutes of Health Research [CIHR], National Research Council [NHMRC]).

Publisher Copyright:
© 2017 The Author(s).

Funding

ation [EME], Health Technology Assessment [HTA], Research for Patient Benefit Programme [RfPB], Programme Grants for Applied Research [PGfAR], Public Health Research [PHR], Invention for Innovation [i4i], Health Services and Delivery Research); MRC (Developmental Pathway Funding Scheme [DPFS], Wellcome Trust [Health Challenge Innovation Fund], Arthritis Research UK, British Heart Foundation, Cancer Research UK [clinical research, new agent, population research]); NIHR Research Design Service; NIHR Statistics Group; and NHS Health Research Authority (HRA). BOS: Bristol Online Surveys; CIHR: Canadian Institutes of Health Research; CRUK: Cancer Research UK; CTU: Clinical Trials Tnit; DELTA: Difference ELicitation in TriAls; DPFS: Developmental Pathway Funding Scheme; EME: Efficacy and Mechanism Evaluation; HTA: Health Technology Assessment; HTMR: Hubs for Trials Methodology Research; i4i: Invention for Innovation; MRC: Medical Research Council; NHMRC: National Health and Medical Research Council; NIH: National Institutes of Health; NIHR: National Institute for Health Research; PCORI: Patient-Centered Outcomes Research Institute; PGfAR: Programme Grants for Applied Research; PHR: Public Health Research; PSI: Statisticians in the Pharmaceutical Industry; RCT: Randomised controlled trial; RDS: Research Design Service; SCT: Society for Clinical Trials; UKCRC: UK Clinical Research Collaboration We will also review online guidance that has been provided by the relevant UK trial funding schemes run by the National Institute for Health Research (NIHR), including EME, Health Technology Assessment (HTA), the Research for Patient Benefit Programme, Programme Grants for Applied Research (PGfAR), Public Health Research (PHR), Invention for Innovation (i4i), and Health Services and Delivery Research; the Medical Research Council (MRC) Developmental Pathway Funding Scheme (DPFS); the Wellcome Trust (Health Challenge Innovation Fund); and Cancer Research UK (CRUK) (phase III clinical trial, new agent, population research). We will also review any guidance documents relating to sample size specification provided by the NIHR Research Design Service (RDS). Online guidance documents will be reviewed with individual schemed contact to provide clarification where necessary. We will also review guidance provided by leading international funding streams (National Institutes of Health [NIH], Patient-Centered Outcomes Research Institute [PCORI], Canadian Institutes of Health Research [CIHR], National Research Council [NHMRC]).

Keywords

Clinically important difference
Effect size
Guidance
Pilot study
Randomised controlled trial
Sample size
Target difference

Access to Document

10.1186/s13063-017-1969-5

Cite this

Cook, J. A., Julious, S. A., Sones, W., Rothwell, J. C., Ramsay, C. R., Hampson, L. V., Emsley, R., Walters, S. J., Hewitt, C., Bland, M., Fergusson, D. A., Berlin, J. A., Altman, D., & Vale, L. D. (2017). Choosing the target difference ('effect size') for a randomised controlled trial - DELTA² guidance protocol. Trials, 18(1), Article 271. https://doi.org/10.1186/s13063-017-1969-5

@article{fb746e1263e9451db071cf51f04601cc,

title = "Choosing the target difference ('effect size') for a randomised controlled trial - DELTA2 guidance protocol",

abstract = "Background: A key step in the design of a randomised controlled trial (RCT) is the estimation of the number of participants needed. By far the most common approach is to specify a target difference and then estimate the corresponding sample size; this sample size is chosen to provide reassurance that the trial will have high statistical power to detect such a difference between the randomised groups (at the planned statistical significance level). The sample size has many implications for the conduct of the study, as well as carrying scientific and ethical aspects to its choice. Despite the critical role of the target difference for the primary outcome in the design of an RCT, the manner in which it is determined has received little attention. This article reports the protocol of the Difference ELicitation in TriAls (DELTA2) project, which will produce guidance on the specification and reporting of the target difference for the primary outcome in a sample size calculation for RCTs. Methods/design: The DELTA2 project has five components: systematic literature reviews of recent methodological developments (stage 1) and existing funder guidance (stage 2); a Delphi study (stage 3); a 2-day consensus meeting bringing together researchers, funders and patient representatives, as well as one-off engagement sessions at relevant stakeholder meetings (stage 4); and the preparation and dissemination of a guidance document (stage 5). Discussion: Specification of the target difference for the primary outcome is a key component of the design of an RCT. There is a need for better guidance for researchers and funders regarding specification and reporting of this aspect of trial design. The aim of this project is to produce consensus based guidance for researchers and funders.",

keywords = "Clinically important difference, Effect size, Guidance, Pilot study, Randomised controlled trial, Sample size, Target difference",

author = "Cook, \{Jonathan A.\} and Julious, \{Steven A.\} and William Sones and Rothwell, \{Joanne C.\} and Ramsay, \{Craig R.\} and Hampson, \{Lisa V.\} and Richard Emsley and Walters, \{Stephen J.\} and Catherine Hewitt and Martin Bland and Fergusson, \{Dean A.\} and Berlin, \{Jesse A.\} and Doug Altman and Vale, \{Luke D.\}",

note = "Funding Information: ation [EME], Health Technology Assessment [HTA], Research for Patient Benefit Programme [RfPB], Programme Grants for Applied Research [PGfAR], Public Health Research [PHR], Invention for Innovation [i4i], Health Services and Delivery Research); MRC (Developmental Pathway Funding Scheme [DPFS], Wellcome Trust [Health Challenge Innovation Fund], Arthritis Research UK, British Heart Foundation, Cancer Research UK [clinical research, new agent, population research]); NIHR Research Design Service; NIHR Statistics Group; and NHS Health Research Authority (HRA). Funding Information: BOS: Bristol Online Surveys; CIHR: Canadian Institutes of Health Research; CRUK: Cancer Research UK; CTU: Clinical Trials Tnit; DELTA: Difference ELicitation in TriAls; DPFS: Developmental Pathway Funding Scheme; EME: Efficacy and Mechanism Evaluation; HTA: Health Technology Assessment; HTMR: Hubs for Trials Methodology Research; i4i: Invention for Innovation; MRC: Medical Research Council; NHMRC: National Health and Medical Research Council; NIH: National Institutes of Health; NIHR: National Institute for Health Research; PCORI: Patient-Centered Outcomes Research Institute; PGfAR: Programme Grants for Applied Research; PHR: Public Health Research; PSI: Statisticians in the Pharmaceutical Industry; RCT: Randomised controlled trial; RDS: Research Design Service; SCT: Society for Clinical Trials; UKCRC: UK Clinical Research Collaboration Funding Information: We will also review online guidance that has been provided by the relevant UK trial funding schemes run by the National Institute for Health Research (NIHR), including EME, Health Technology Assessment (HTA), the Research for Patient Benefit Programme, Programme Grants for Applied Research (PGfAR), Public Health Research (PHR), Invention for Innovation (i4i), and Health Services and Delivery Research; the Medical Research Council (MRC) Developmental Pathway Funding Scheme (DPFS); the Wellcome Trust (Health Challenge Innovation Fund); and Cancer Research UK (CRUK) (phase III clinical trial, new agent, population research). We will also review any guidance documents relating to sample size specification provided by the NIHR Research Design Service (RDS). Online guidance documents will be reviewed with individual schemed contact to provide clarification where necessary. We will also review guidance provided by leading international funding streams (National Institutes of Health [NIH], Patient-Centered Outcomes Research Institute [PCORI], Canadian Institutes of Health Research [CIHR], National Research Council [NHMRC]). Publisher Copyright: {\textcopyright} 2017 The Author(s).",

year = "2017",

month = jun,

day = "12",

doi = "10.1186/s13063-017-1969-5",

language = "English",

volume = "18",

journal = "Trials",

issn = "1745-6215",

publisher = "BioMed Central",

number = "1",

}

TY - JOUR

T1 - Choosing the target difference ('effect size') for a randomised controlled trial - DELTA2 guidance protocol

AU - Cook, Jonathan A.

AU - Julious, Steven A.

AU - Sones, William

AU - Rothwell, Joanne C.

AU - Ramsay, Craig R.

AU - Hampson, Lisa V.

AU - Emsley, Richard

AU - Walters, Stephen J.

AU - Hewitt, Catherine

AU - Bland, Martin

AU - Fergusson, Dean A.

AU - Berlin, Jesse A.

AU - Altman, Doug

AU - Vale, Luke D.

N1 - Funding Information: ation [EME], Health Technology Assessment [HTA], Research for Patient Benefit Programme [RfPB], Programme Grants for Applied Research [PGfAR], Public Health Research [PHR], Invention for Innovation [i4i], Health Services and Delivery Research); MRC (Developmental Pathway Funding Scheme [DPFS], Wellcome Trust [Health Challenge Innovation Fund], Arthritis Research UK, British Heart Foundation, Cancer Research UK [clinical research, new agent, population research]); NIHR Research Design Service; NIHR Statistics Group; and NHS Health Research Authority (HRA). Funding Information: BOS: Bristol Online Surveys; CIHR: Canadian Institutes of Health Research; CRUK: Cancer Research UK; CTU: Clinical Trials Tnit; DELTA: Difference ELicitation in TriAls; DPFS: Developmental Pathway Funding Scheme; EME: Efficacy and Mechanism Evaluation; HTA: Health Technology Assessment; HTMR: Hubs for Trials Methodology Research; i4i: Invention for Innovation; MRC: Medical Research Council; NHMRC: National Health and Medical Research Council; NIH: National Institutes of Health; NIHR: National Institute for Health Research; PCORI: Patient-Centered Outcomes Research Institute; PGfAR: Programme Grants for Applied Research; PHR: Public Health Research; PSI: Statisticians in the Pharmaceutical Industry; RCT: Randomised controlled trial; RDS: Research Design Service; SCT: Society for Clinical Trials; UKCRC: UK Clinical Research Collaboration Funding Information: We will also review online guidance that has been provided by the relevant UK trial funding schemes run by the National Institute for Health Research (NIHR), including EME, Health Technology Assessment (HTA), the Research for Patient Benefit Programme, Programme Grants for Applied Research (PGfAR), Public Health Research (PHR), Invention for Innovation (i4i), and Health Services and Delivery Research; the Medical Research Council (MRC) Developmental Pathway Funding Scheme (DPFS); the Wellcome Trust (Health Challenge Innovation Fund); and Cancer Research UK (CRUK) (phase III clinical trial, new agent, population research). We will also review any guidance documents relating to sample size specification provided by the NIHR Research Design Service (RDS). Online guidance documents will be reviewed with individual schemed contact to provide clarification where necessary. We will also review guidance provided by leading international funding streams (National Institutes of Health [NIH], Patient-Centered Outcomes Research Institute [PCORI], Canadian Institutes of Health Research [CIHR], National Research Council [NHMRC]). Publisher Copyright: © 2017 The Author(s).

PY - 2017/6/12

Y1 - 2017/6/12

N2 - Background: A key step in the design of a randomised controlled trial (RCT) is the estimation of the number of participants needed. By far the most common approach is to specify a target difference and then estimate the corresponding sample size; this sample size is chosen to provide reassurance that the trial will have high statistical power to detect such a difference between the randomised groups (at the planned statistical significance level). The sample size has many implications for the conduct of the study, as well as carrying scientific and ethical aspects to its choice. Despite the critical role of the target difference for the primary outcome in the design of an RCT, the manner in which it is determined has received little attention. This article reports the protocol of the Difference ELicitation in TriAls (DELTA2) project, which will produce guidance on the specification and reporting of the target difference for the primary outcome in a sample size calculation for RCTs. Methods/design: The DELTA2 project has five components: systematic literature reviews of recent methodological developments (stage 1) and existing funder guidance (stage 2); a Delphi study (stage 3); a 2-day consensus meeting bringing together researchers, funders and patient representatives, as well as one-off engagement sessions at relevant stakeholder meetings (stage 4); and the preparation and dissemination of a guidance document (stage 5). Discussion: Specification of the target difference for the primary outcome is a key component of the design of an RCT. There is a need for better guidance for researchers and funders regarding specification and reporting of this aspect of trial design. The aim of this project is to produce consensus based guidance for researchers and funders.

AB - Background: A key step in the design of a randomised controlled trial (RCT) is the estimation of the number of participants needed. By far the most common approach is to specify a target difference and then estimate the corresponding sample size; this sample size is chosen to provide reassurance that the trial will have high statistical power to detect such a difference between the randomised groups (at the planned statistical significance level). The sample size has many implications for the conduct of the study, as well as carrying scientific and ethical aspects to its choice. Despite the critical role of the target difference for the primary outcome in the design of an RCT, the manner in which it is determined has received little attention. This article reports the protocol of the Difference ELicitation in TriAls (DELTA2) project, which will produce guidance on the specification and reporting of the target difference for the primary outcome in a sample size calculation for RCTs. Methods/design: The DELTA2 project has five components: systematic literature reviews of recent methodological developments (stage 1) and existing funder guidance (stage 2); a Delphi study (stage 3); a 2-day consensus meeting bringing together researchers, funders and patient representatives, as well as one-off engagement sessions at relevant stakeholder meetings (stage 4); and the preparation and dissemination of a guidance document (stage 5). Discussion: Specification of the target difference for the primary outcome is a key component of the design of an RCT. There is a need for better guidance for researchers and funders regarding specification and reporting of this aspect of trial design. The aim of this project is to produce consensus based guidance for researchers and funders.

KW - Clinically important difference

KW - Effect size

KW - Guidance

KW - Pilot study

KW - Randomised controlled trial

KW - Sample size

KW - Target difference

UR - https://www.scopus.com/pages/publications/85020406157

U2 - 10.1186/s13063-017-1969-5

DO - 10.1186/s13063-017-1969-5

M3 - Article

C2 - 28606102

AN - SCOPUS:85020406157

SN - 1745-6215

VL - 18

JO - Trials

JF - Trials

IS - 1

M1 - 271

ER -