Fig. 1From: Accounting for expected attrition in the planning of cluster randomized trials for assessing treatment effect heterogeneityHeatmap of the ratio of sample size estimated based on the proposed formula under MCAR to that obtained from the direct inflation method under the follow-up rate of \(\pi =0.6.\), the cluster size of \(m\in \left\{20, 100\right\}\) and the missingness ICC of \(\tau \in \left\{\text{0.05,0.6,1}\right\}\)  Back to article page