» Articles » PMID: 16109162

A Priori Postulated and Real Power in Cluster Randomized Trials: Mind the Gap

Overview
Publisher Biomed Central
Date 2005 Aug 20
PMID 16109162
Citations 9
Authors
Affiliations
Soon will be listed here.
Abstract

Background: Cluster randomization design is increasingly used for the evaluation of health-care, screening or educational interventions. The intraclass correlation coefficient (ICC) defines the clustering effect and be specified during planning. The aim of this work is to study the influence of the ICC on power in cluster randomized trials.

Methods: Power contour graphs were drawn to illustrate the loss in power induced by an underestimation of the ICC when planning trials. We also derived the maximum achievable power given a specified ICC.

Results: The magnitude of the ICC can have a major impact on power, and with low numbers of clusters, 80% power may not be achievable.

Conclusion: Underestimating the ICC during planning cluster randomized trials can lead to a seriously underpowered trial. Publication of a priori postulated and a posteriori estimated ICCs is necessary for a more objective reading: negative trial results may be the consequence of a loss of power due to a mis-specification of the ICC.

Citing Articles

Cluster-Randomized Studies.

Lorenz E, Kopke S, Pfaff H, Blettner M Dtsch Arztebl Int. 2018; 115(10):163-168.

PMID: 29587960 PMC: 5881078. DOI: 10.3238/arztebl.2018.0163.


Ethical implications of excessive cluster sizes in cluster randomised trials.

Hemming K, Taljaard M, Forbes G, Eldridge S, Weijer C BMJ Qual Saf. 2018; 27(8):664-670.

PMID: 29463768 PMC: 6204928. DOI: 10.1136/bmjqs-2017-007164.


How to design efficient cluster randomised trials.

Hemming K, Eldridge S, Forbes G, Weijer C, Taljaard M BMJ. 2017; 358:j3064.

PMID: 28710062 PMC: 5508848. DOI: 10.1136/bmj.j3064.


The Effects of Skill Training on Social Workers' Professional Competences in Norway: Results of a Cluster-Randomised Study.

Malmberg-Heimonen I, Natland S, Toge A, Hansen H Br J Soc Work. 2016; 46(5):1354-1371.

PMID: 27559232 PMC: 4985729. DOI: 10.1093/bjsw/bcv073.


Sample size calculations for stepped wedge and cluster randomised trials: a unified approach.

Hemming K, Taljaard M J Clin Epidemiol. 2015; 69:137-46.

PMID: 26344808 PMC: 4687983. DOI: 10.1016/j.jclinepi.2015.08.015.


References
1.
Lake S, Kammann E, Klar N, Betensky R . Sample size re-estimation in cluster randomization trials. Stat Med. 2002; 21(10):1337-50. DOI: 10.1002/sim.1121. View

2.
Reading R, Harvey I, McLean M . Cluster randomised trials in maternal and child health: implications for power and sample size. Arch Dis Child. 2000; 82(1):79-83. PMC: 1718186. DOI: 10.1136/adc.82.1.79. View

3.
Donner A, Klar N . Statistical considerations in the design and analysis of community intervention trials. J Clin Epidemiol. 1996; 49(4):435-9. DOI: 10.1016/0895-4356(95)00511-0. View

4.
Smeeth L, Ng E . Intraclass correlation coefficients for cluster randomized trials in primary care: data from the MRC Trial of the Assessment and Management of Older People in the Community. Control Clin Trials. 2002; 23(4):409-21. DOI: 10.1016/s0197-2456(02)00208-8. View

5.
Kerry S, Bland J . Sample size in cluster randomisation. BMJ. 1998; 316(7130):549. PMC: 2665662. DOI: 10.1136/bmj.316.7130.549. View