A menu-driven facility for sample-size calculations in cluster randomized controlled trials

Abstract

We introduce the Stata menu-driven command clustersampsi, which calculates sample sizes, detectable differences, and power for cluster randomized controlled trials. The command permits continuous, binary, and rate outcomes (with normal approximations) for comparisons of two-sided tests in two equal-sized arms. The command allows for specification of the number of clusters available, or the cluster size, or the average cluster size along with an estimate of the variation of cluster sizes. When the number of clusters available is insufficient to detect the required difference at the prespecified power, clustersampsi will return the minimum number of clusters required under the prespecified design along with the minimum detectable difference and maximum achievable power (both for the prespecified number of clusters). Cluster heterogeneity can be parameterized by using either the intracluster correlation or the coefficient of variation. The command is illustrated via examples. © 2013 StataCorp LP.