Background\udThe current methodology for sample size calculation for steppedwedge\udcluster randomised trials (SW-CRTs) is based on the assumption\udof the clusters being of equal size. However, as is often the case\udin CRTs, the clusters in SW-CRTs are likely to vary in size which in\udCRTs of other designs leads to a reduction in power. The effect of an\udimbalance in cluster sizes on SW-CRTs was not known, nor what an\udappropriate adjustment to the sample size should be.\udTrials 2016, Volume 17 Suppl 1 Page 4 of 6\udMethods\udWe proposed three adjusted design effects (DEs) for use in the calculation\udof the sample size for SW-CRTs with varying degrees of imbalance\udin cluster size, based on those suggested for use in CRTs with\udunequal cluster sizes. A simulation study was conducted which investigated\udthe effect of unequal cluster sizes on the power of SW-CRTs,\udwhen the sample size was calculated using both the standard\udmethod and the three proposed adjusted DEs.\udResults\udAn imbalance in cluster size was not found to significantly affect the\udpower of a SW-CRT, and the proposed adjusted DEs generally resulted\udin trials that were severely over-powered.\udConclusions\udWe recommend that the standard method of sample size calculation\udfor SW-CRTs be used when any imbalance in cluster size is expected\udto be small. When there is likely to be a large imbalance in cluster\udsize it is recommended that simulations be used to determine if additional\udclusters are needed.
展开▼
