Sample size determinations for stepped-wedge clinical trials from a three-level data hierarchy perspective

Moonseong Heo, Namhee Kim, Michael L. Rinke, Judith Wylie-Rosett

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

Stepped-wedge (SW) designs have been steadily implemented in a variety of trials. A SW design typically assumes a three-level hierarchical data structure where participants are nested within times or periods which are in turn nested within clusters. Therefore, statistical models for analysis of SW trial data need to consider two correlations, the first and second level correlations. Existing power functions and sample size determination formulas had been derived based on statistical models for two-level data structures. Consequently, the second-level correlation has not been incorporated in conventional power analyses. In this paper, we derived a closed-form explicit power function based on a statistical model for three-level continuous outcome data. The power function is based on a pooled overall estimate of stratified cluster-specific estimates of an intervention effect. The sampling distribution of the pooled estimate is derived by applying a fixed-effect meta-analytic approach. Simulation studies verified that the derived power function is unbiased and can be applicable to varying number of participants per period per cluster. In addition, when data structures are assumed to have two levels, we compare three types of power functions by conducting additional simulation studies under a two-level statistical model. In this case, the power function based on a sampling distribution of a marginal, as opposed to pooled, estimate of the intervention effect performed the best. Extensions of power functions to binary outcomes are also suggested.

Original languageEnglish (US)
Pages (from-to)480-489
Number of pages10
JournalStatistical Methods in Medical Research
Volume27
Issue number2
DOIs
StatePublished - Feb 1 2018

Fingerprint

Sample Size Determination
Power Function
Statistical Models
Wedge
Clinical Trials
Sample Size
Statistical Model
Data Structures
Sampling Distribution
Estimate
Simulation Study
Hierarchical Data
Binary Outcomes
Fixed Effects
Hierarchical Structure
Hierarchy
Closed-form

Keywords

  • design effect
  • effect size
  • sample size
  • statistical power
  • Stepped-wedge design
  • three level data

ASJC Scopus subject areas

  • Epidemiology
  • Statistics and Probability
  • Health Information Management

Cite this

Sample size determinations for stepped-wedge clinical trials from a three-level data hierarchy perspective. / Heo, Moonseong; Kim, Namhee; Rinke, Michael L.; Wylie-Rosett, Judith.

In: Statistical Methods in Medical Research, Vol. 27, No. 2, 01.02.2018, p. 480-489.

Research output: Contribution to journalArticle

@article{592de4f83173404f80754cb085b642de,
title = "Sample size determinations for stepped-wedge clinical trials from a three-level data hierarchy perspective",
abstract = "Stepped-wedge (SW) designs have been steadily implemented in a variety of trials. A SW design typically assumes a three-level hierarchical data structure where participants are nested within times or periods which are in turn nested within clusters. Therefore, statistical models for analysis of SW trial data need to consider two correlations, the first and second level correlations. Existing power functions and sample size determination formulas had been derived based on statistical models for two-level data structures. Consequently, the second-level correlation has not been incorporated in conventional power analyses. In this paper, we derived a closed-form explicit power function based on a statistical model for three-level continuous outcome data. The power function is based on a pooled overall estimate of stratified cluster-specific estimates of an intervention effect. The sampling distribution of the pooled estimate is derived by applying a fixed-effect meta-analytic approach. Simulation studies verified that the derived power function is unbiased and can be applicable to varying number of participants per period per cluster. In addition, when data structures are assumed to have two levels, we compare three types of power functions by conducting additional simulation studies under a two-level statistical model. In this case, the power function based on a sampling distribution of a marginal, as opposed to pooled, estimate of the intervention effect performed the best. Extensions of power functions to binary outcomes are also suggested.",
keywords = "design effect, effect size, sample size, statistical power, Stepped-wedge design, three level data",
author = "Moonseong Heo and Namhee Kim and Rinke, {Michael L.} and Judith Wylie-Rosett",
year = "2018",
month = "2",
day = "1",
doi = "10.1177/0962280216632564",
language = "English (US)",
volume = "27",
pages = "480--489",
journal = "Statistical Methods in Medical Research",
issn = "0962-2802",
publisher = "SAGE Publications Ltd",
number = "2",

}

TY - JOUR

T1 - Sample size determinations for stepped-wedge clinical trials from a three-level data hierarchy perspective

AU - Heo, Moonseong

AU - Kim, Namhee

AU - Rinke, Michael L.

AU - Wylie-Rosett, Judith

PY - 2018/2/1

Y1 - 2018/2/1

N2 - Stepped-wedge (SW) designs have been steadily implemented in a variety of trials. A SW design typically assumes a three-level hierarchical data structure where participants are nested within times or periods which are in turn nested within clusters. Therefore, statistical models for analysis of SW trial data need to consider two correlations, the first and second level correlations. Existing power functions and sample size determination formulas had been derived based on statistical models for two-level data structures. Consequently, the second-level correlation has not been incorporated in conventional power analyses. In this paper, we derived a closed-form explicit power function based on a statistical model for three-level continuous outcome data. The power function is based on a pooled overall estimate of stratified cluster-specific estimates of an intervention effect. The sampling distribution of the pooled estimate is derived by applying a fixed-effect meta-analytic approach. Simulation studies verified that the derived power function is unbiased and can be applicable to varying number of participants per period per cluster. In addition, when data structures are assumed to have two levels, we compare three types of power functions by conducting additional simulation studies under a two-level statistical model. In this case, the power function based on a sampling distribution of a marginal, as opposed to pooled, estimate of the intervention effect performed the best. Extensions of power functions to binary outcomes are also suggested.

AB - Stepped-wedge (SW) designs have been steadily implemented in a variety of trials. A SW design typically assumes a three-level hierarchical data structure where participants are nested within times or periods which are in turn nested within clusters. Therefore, statistical models for analysis of SW trial data need to consider two correlations, the first and second level correlations. Existing power functions and sample size determination formulas had been derived based on statistical models for two-level data structures. Consequently, the second-level correlation has not been incorporated in conventional power analyses. In this paper, we derived a closed-form explicit power function based on a statistical model for three-level continuous outcome data. The power function is based on a pooled overall estimate of stratified cluster-specific estimates of an intervention effect. The sampling distribution of the pooled estimate is derived by applying a fixed-effect meta-analytic approach. Simulation studies verified that the derived power function is unbiased and can be applicable to varying number of participants per period per cluster. In addition, when data structures are assumed to have two levels, we compare three types of power functions by conducting additional simulation studies under a two-level statistical model. In this case, the power function based on a sampling distribution of a marginal, as opposed to pooled, estimate of the intervention effect performed the best. Extensions of power functions to binary outcomes are also suggested.

KW - design effect

KW - effect size

KW - sample size

KW - statistical power

KW - Stepped-wedge design

KW - three level data

UR - http://www.scopus.com/inward/record.url?scp=85041963396&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85041963396&partnerID=8YFLogxK

U2 - 10.1177/0962280216632564

DO - 10.1177/0962280216632564

M3 - Article

C2 - 26988927

AN - SCOPUS:85041963396

VL - 27

SP - 480

EP - 489

JO - Statistical Methods in Medical Research

JF - Statistical Methods in Medical Research

SN - 0962-2802

IS - 2

ER -