The effect of cluster size variability on statistical power in cluster-randomized trials
Stephen A Lauer, Ken P Kleinman, Nicholas G Reich, Stephen A Lauer, Ken P Kleinman, Nicholas G Reich
Abstract
The frequency of cluster-randomized trials (CRTs) in peer-reviewed literature has increased exponentially over the past two decades. CRTs are a valuable tool for studying interventions that cannot be effectively implemented or randomized at the individual level. However, some aspects of the design and analysis of data from CRTs are more complex than those for individually randomized controlled trials. One of the key components to designing a successful CRT is calculating the proper sample size (i.e. number of clusters) needed to attain an acceptable level of statistical power. In order to do this, a researcher must make assumptions about the value of several variables, including a fixed mean cluster size. In practice, cluster size can often vary dramatically. Few studies account for the effect of cluster size variation when assessing the statistical power for a given trial. We conducted a simulation study to investigate how the statistical power of CRTs changes with variable cluster sizes. In general, we observed that increases in cluster size variability lead to a decrease in power.
Trial registration: ClinicalTrials.gov NCT01249625.
Conflict of interest statement
Competing Interests: Co-author Nicholas G Reich is a PLOS ONE Editorial Board member. This does not alter the authors’ adherence to PLOS ONE Editorial policies and criteria.
Figures
References
- Platt R, Takvorian S, Septimus E, Hickok J, Moody J, et al. (2010) Cluster randomized trials in comparative effectiveness research: randomizing hospitals to test methods for prevention of healthcare-associated infections. Medical Care 48: S52 10.1097/MLR.0b013e3181dbebcf
- Campbell MK, Elbourne DR, Altman DG (2004) Consort statement: extension to cluster randomised trials. British Medical Journal 328: 702–708. 10.1136/bmj.328.7441.702
- Campbell MK, Piaggio G, Elbourne DR, et al. (2012) Consort 2010 statement: extension to cluster randomised trials. British Medical Journal 345 10.1136/bmj.e5661
- Donner A, Birkett N, Buck C (1981) Randomization by cluster: sample size requirements and analysis. American Journal of Epidemiology 114: 906–914.
- Kerry SM, Martin Bland J (2001) Unequal cluster sizes for trials in english and welsh general practice: implications for sample size calculations. Statistics in Medicine 20: 377–390. 10.1002/1097-0258(20010215)20:3<377::AID-SIM799>;2-N
- Eldridge SM, Ashby D, Kerry S (2006) Sample size for cluster randomized trials: effect of coefficient of variation of cluster size and analysis method. International Journal of Epidemiology 35: 1292–1300. 10.1093/ije/dyl129
- Campbell MJ, Donner A, Klar N (2007) Developments in cluster randomized trials and statistics in medicine. Statistics in Medicine 26: 2–19. 10.1002/sim.2731
- van Breukelen GJP, Candel MJJM, Berger MPF (2007) Relative efficiency of unequal versus equal cluster sizes in cluster randomized and multicentre trials. Statistics in Medicine 26: 2589–2603. 10.1002/sim.2740
- Candel MJJM, Van Breukelen GJP (2010) Sample size adjustments for varying cluster sizes in cluster randomized trials with binary outcomes analyzed with second-order pql mixed logistic regression. Statistics in Medicine 29: 1488–1501.
- You Z, Williams OD, Aban I, Kabagambe EK, Tiwari HK, et al. (2011) Relative efficiency and sample size for cluster randomized trials with variable cluster sizes. Clinical Trials 8: 27–36. 10.1177/1740774510391492
- Hemming K, Girling A, Sitch A, Marsh J, Lilford R (2011) Sample size calculations for cluster randomised controlled trials with a fixed number of clusters. BMC Medical Research Methodology 11: 102 10.1186/1471-2288-11-102
- R Core Team (2013) R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria. URL . ISBN 3-900051-07-0.
- Reich NG, Myers JA, Obeng D, Milstone AM, Perl TM (2012) Empirical power and sample size calculations for cluster-randomized and cluster randomized crossover studies. PLoS ONE 7: e35564 10.1371/journal.pone.0035564
- Reich NG, Obeng D (2012) clusterPower: Power calculations for cluster-randomized and cluster-randomized crossover trials. URL . R package version 0.5.
- Xie Y (2013) knitr: A general-purpose package for dynamic report generation in r. R package version 15.
- Xie Y (2013) Dynamic Documents with R and knitr. CRC Press.
- Murray DM (1998) Design and Analysis of Group-Randomized Trials. Oxford University Press.
- Using the form of the negative binomial with parameters r and p, where E(X)=r(1−p)p and Var(X)=r(1−p)p2.
Source: PubMed