Stepped wedge cluster randomized trials (SW-CRTs) are increasingly being used to evaluate interventions in medical, public health, educational, and social science contexts. With the longitudinal and crossover natures of an SW-CRT, complex analysis techniques are often needed, which makes appropriately powering SW-CRTs challenging. In this article, we introduce a newly developed SW-CRT power calculator embedded within the power command in Stata. The power calculator assumes a marginal model (that is, generalized estimating equations) for the primary analysis of SW-CRTs, for which other currently available SW-CRT power calculators may not be suitable. The command accommodates complete cross-sectional and closed-cohort designs and includes multilevel correlation structures appropriate for such designs. We discuss the methods and formulas underlying our SW-CRT calculator and provide illustrative examples of the use of power swgee. We provide suggestions about the choice of parameters in power swgee and conclude by discussing areas of future research that may improve the command.
BaioG.LeechR.2016. SWSamp: Computes sample size for a stepped wedge design, using simulation-based calculations. R package version 0.3. https://rdrr.io/github/giabaio/SWSamp/.
3.
BarkerD.McElduffP.D’EsteC.CampbellM. J.2016. Stepped wedge cluster randomised trials: A review of the statistical methodology used and available. BMC Medical Research Methodology16: 69. https://doi.org/10.1186/s12874-016-0176-5.
4.
BrimblecombeJ.FergusonM.ChatfieldM. D.LiberatoS. C.GuntherA.BallK.MoodieM., et al.2017. Effect of a price discount and consumer education strategy on food and beverage purchases in remote Indigenous Australia: A stepped-wedge randomised controlled trial. Lancet Public Health2: e82–e95. https://doi.org/10.1016/S2468-2667(16)30043-3.
5.
ChenJ.ZhouX.LiF.SpiegelmanD.2022. swdpwr: A SAS macro and an R package for power calculations in stepped wedge cluster randomized trials. Computer Methods and Programs in Biomedicine213: 106522. https://doi.org/10.1016/j.cmpb.2021.106522.
6.
CopasA. J.LewisJ. J.ThompsonJ. A.DaveyC.BaioG.HargreavesJ. R.2015. Designing a stepped wedge trial: Three main designs, carry-over effects and randomisation approaches. Trials16: 352. https://doi.org/10.1186/s13063-015-0842-7.
7.
EldridgeS. M.UkoumunneO. C.CarlinJ. B.2009. The intra-cluster correlation coefficient in cluster randomized trials: A review of definitions. International Statistical Review77: 378–394. https://doi.org/10.1111/j.1751-5823.2009.00092.x.
FordW. P.WestgateP. M.2017. Improved standard error estimator for maintaining the validity of inference in cluster randomized trials with a small number of clusters. Biometrical Journal59: 478–495. https://doi.org/10.1002/bimj.201600182.
10.
FordW. P.WestgateP. M.2020. Maintaining the validity of inference in small-sample stepped wedge cluster randomized trials with binary outcomes when using generalized estimating equations. Statistics in Medicine39: 2779–2792. https://doi.org/10.1002/sim.8575.
11.
GirlingA. J.HemmingK.2016. Statistical efficiency and optimal design for stepped cluster studies under linear mixed effects models. Statistics in Medicine35: 2149–2166. https://doi.org/10.1002/sim.6850.
12.
GraylingM. J.WasonJ. M. S.ManderA. P.2017. Stepped wedge cluster randomized controlled trial designs: A review of reporting quality and design features. Trials18: 33. https://doi.org/10.1186/s13063-017-1783-0.
13.
HargreavesJ. R.CopasA. J.BeardE.OsrinD.LewisJ. J.DaveyC.ThompsonJ. A.BaioG.FieldingK. L.ProstA.2015. Five questions to consider before conducting a stepped wedge trial. Trials16: 350. https://doi.org/10.1186/s13063-015-0841-8.
14.
HemmingK.GirlingA.2014. A menu-driven facility for power and detectabledifference calculations in stepped-wedge cluster-randomized trials. Stata Journal14: 363–380. https://doi.org/10.1177/1536867X1401400208.
15.
HemmingK.HainesT. P.ChiltonP. J.GirlingA. J.LilfordR. J.2015. The stepped wedge cluster randomised trial: Rationale, design, analysis, and reporting. BMJ350: h391. https://doi.org/10.1136/bmj.h391.
16.
HemmingK.KaszaJ.HooperR.ForbesA.TaljaardM.2020. A tutorial on sample size calculation for multiple-period cluster randomized parallel, cross-over and stepped-wedge trials using the shiny CRT calculator. International Journal of Epidemiology49: 979–995. https://doi.org/10.1093/ije/dyz237.
17.
HemmingK.LilfordR.GirlingA. J.2015. Stepped-wedge cluster randomised controlled trials: A generic framework including parallel and multiple-level designs. Statistics in Medicine34: 181–196. https://doi.org/10.1002/sim.6325.
18.
HemmingK.TaljaardM.2020. Reflection on modern methods: When is a stepped-wedge cluster randomized trial a good study design choice?International Journal of Epidemiology49: 1043–1052. https://doi.org/10.1093/ije/dyaa077.
19.
HooperR.TeerenstraS.de HoopE.EldridgeS.2016. Sample size calculation for stepped wedge and other longitudinal cluster randomised trials. Statistics in Medicine35: 4718–4728. https://doi.org/10.1002/sim.7028.
20.
HughesJ. P.GranstonT. S.HeagertyP. J.2015. Current issues in the design and analysis of stepped wedge trials. Contemporary Clinical Trials45: 55–60. https://doi.org/10.1016/j.cct.2015.07.006.
21.
HusseyM. A.HughesJ. P.2007. Design and analysis of stepped wedge cluster randomized trials. Contemporary Clinical Trials28: 182–191. https://doi.org/10.1016/j.cct.2006.05.007.
22.
KaszaJ.ForbesA. B.2019. Inference for the treatment effect in multiple-period cluster randomised trials when random effect correlation structure is misspecified. Statistical Methods in Medical Research28: 3112–3122. https://doi.org/10.1177/0962280218797151.
23.
KaszaJ.HemmingK.HooperR.MatthewsJ. N. S.ForbesA. B.2019. Impact of non-uniform correlation structure on sample size and power in multiple-period cluster randomised trials. Statistical Methods in Medical Research28: 703–716. https://doi.org/10.1177/0962280217734981.
24.
LiF.2020. Design and analysis considerations for cohort stepped wedge cluster randomized trials with a decay correlation structure. Statistics in Medicine39: 438–455. https://doi.org/10.1002/sim.8415.
25.
LiF.ForbesA. B.TurnerE. L.PreisserJ. S.2019. Power and sample size requirements for GEE analyses of cluster randomized crossover trials. Statistics in Medicine38: 636–649. https://doi.org/10.1002/sim.7995.
26.
LiF.HughesJ. P.HemmingK.TaljaardM.MelnickE. R.HeagertyP. J.2020. Mixed-effects models for the design and analysis of stepped wedge cluster randomized trials: An overview. Statistical Methods in Medical Research30: 612–639. https://doi.org/10.1177/0962280220932962.
27.
LiF.TurnerE. L.PreisserJ. S.2018a. Optimal allocation of clusters in cohort stepped wedge designs. Statistics and Probability Letters137: 257–263. https://doi.org/10.1016/j.spl.2018.02.002.
28.
LiF.TurnerE. L.PreisserJ. S.2018b. Sample size determination for GEE analyses of stepped wedge cluster randomized trials. Biometrics74: 1450–1458. https://doi.org/10.1111/biom.12918.
29.
LiF.YuH.RathouzP. J.TurnerE. L.PreisserJ. S.2021. Marginal modeling of cluster-period means and intraclass correlations in stepped wedge designs with binary outcomes. ArXiv Working Paper No. arXiv:2101.00484. https://doi.org/10.48550/arXiv.2101.00484.
MartinJ.GirlingA.NirantharakumarK.RyanR.MarshallT.HemmingK.2016a. Intra-cluster and inter-period correlation coefficients for cross-sectional cluster randomised controlled trials for type-2 diabetes in UK primary care. Trials17: 1–12. https://doi.org/10.1186/s13063-016-1532-9.
32.
MartinJ.TaljaardM.GirlingA.HemmingK.2016b. Systematic review finds major deficiencies in sample size methodology and reporting for stepped-wedge cluster randomised trials. BMJ6: e010166. https://doi.org/10.1136/bmjopen-2015-010166.
33.
MaselkoJ.SikanderS.TurnerE. L.BatesL. M.AhmadI.AtifN.BaranovV., et al.2020. Effectiveness of a peer-delivered, psychosocial intervention on maternal depression and child development at 3 years postnatal: A cluster randomised trial in Pakistan. Lancet Psychiatry7: 775–787. https://doi.org/10.1016/S2215-0366(20)30258-3.
34.
MdegeN. D.ManM. S.Taylor Nee BrownC. A.TorgersonD. J.2011. Systematic review of stepped wedge cluster randomized trials shows that design is particularly used to evaluate interventions during routine implementation. Journal of Clinical Epidemiology64: 936–948. https://doi.org/10.1016/j.jclinepi.2010.12.003.
35.
MhurchuC. N.GortonD.TurleyM.JiangY.MichieJ.MaddisonR.HattieJ.2013. Effects of a free school breakfast programme on children’s attendance, academic achievement and short-term hunger: Results from a stepped-wedge, cluster randomised controlled trial. Journal of Epidemiology and Community Health67: 257–264. https://doi.org/10.1136/jech-2012-201540.
36.
MhurchuC. N.TurleyM.GortonD.JiangY.MichieJ.MaddisonR.HattieJ.2010. Effects of a free school breakfast programme on school attendance, achievement, psychosocial function, and nutrition: A stepped wedge cluster randomised trial. BMC Public Health10: 738. https://doi.org/10.1186/1471-2458-10-738.
37.
MurrayD. M.1998. Design and Analysis of Group-Randomized Trials. New York: Oxford University Press.
38.
ParkerD. R.EvangelouE.EatonC. B.2005. Intraclass correlation coefficients for cluster randomized trials in primary care: The cholesterol education and research trial (CEART). Contemporary Clinical Trials26: 260–267. https://doi.org/10.1016/j.cct.2005.01.002.
39.
PreisserJ. S.LuB.QaqishB. F.2008. Finite sample adjustments in estimating equations and covariance estimators for intracluster correlations. Statistics in Medicine27: 5764–5785. https://doi.org/10.1002/sim.3390.
40.
PreisserJ. S.YoungM. L.ZaccaroD. J.WolfsonM.2003. An integrated population-averaged approach to the design, analysis and sample size determination of cluster-unit trials. Statistics in Medicine22: 1235–1254. https://doi.org/10.1002/sim.1379.
41.
QaqishB. F.2003. A family of multivariate binary distributions for simulating correlated binary variables with specified marginal means and correlations. Biometrika90: 455–463. https://doi.org/10.1093/biomet/90.2.455.
RothJ.Sant’AnnaP. H. C.2021. Efficient estimation for staggered rollout designs. ArXiv Working Paper No. arXiv:2102.01291. https://doi.org/10.48550/arXiv.2102.01291.
44.
ScottJ. M.deCampA.JuraskaM.FayM. P.GilbertP. B.2017. Finite-sample corrected generalized estimating equation of population average treatment effects in stepped wedge cluster randomized trials. Statistical Methods in Medical Research26: 583–597. https://doi.org/10.1177/0962280214552092.
45.
SvetkeyL. P.BatchB. C.LinP. H.IntilleS. S.CorsinoL.TysonC. C.BosworthH. B., et al.2015. Cell phone intervention for you (CITY): A randomized, controlled trial of behavioral weight loss intervention for young adults using mobile technology. Obesity23: 2133–2141. https://doi.org/10.1002/oby.21226.
46.
TaljaardM.TeerenstraS.IversN. M.FergussonD. A.2016. Substantial risks associated with few clusters in cluster randomized and stepped wedge designs. Clinical Trials13: 459–463. https://doi.org/10.1177/1740774516634316.
47.
ThompsonJ. A.HemmingK.ForbesA.FieldingK.HayesR.2020. Comparison of small-sample standard-error corrections for generalised estimating equations in stepped wedge cluster randomised trials with a binary outcome: A simulation study. Statistical Methods in Medical Research30: 425–439. https://doi.org/10.1177/0962280220958735.
48.
TianZ.PreisserJ.EssermanD.TurnerE.RathouzP.LiF.2022. Impact of unequal cluster sizes for GEE analyses of stepped wedge cluster randomized trials with binary outcomes. Biometrical64: 419–439. https://doi.org/10.1002/bimj.202100112.
49.
TurnerE. L.LiF.GallisJ. A.PragueM.MurrayD. M.2017a. Review of recent methodological developments in group-randomized trials: Part 1—Design. American Journal of Public Health107: 907–915. https://www.doi.org/10.2105/AJPH.2017.303706.
50.
TurnerE. L.PragueM.GallisJ. A.LiF.MurrayD. M.2017b. Review of recent methodological developments in group-randomized trials: Part 2—Analysis. American Journal of Public Health107: 1078–1086. https://www.doi.org/10.2105/AJPH.2017.303707.
51.
VoldalE. C.HakhuN. R.XiaF.HeagertyP. J.HughesJ. P.2020. swCRTdesign: An R package for stepped wedge trial design and analysis. Computer Methods and Programs in Biomedicine196: 105514. https://doi.org/10.1016/j.cmpb.2020.105514.
ZhouX.LiaoX.KunzL. M.NormandS.-L. T.WangM.SpiegelmanD.2020. A maximum likelihood approach to power calculations for stepped wedge designs of binary outcomes. Biostatistics21: 102–121. https://doi.org/10.1093/biostatistics/kxy031.
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
