Fig. 6
From: The impact of iterative removal of low-information cluster-period cells from a stepped wedge design
Top-left: Design schematic of a 10-period complete stepped wedge design. Top-right: Design schematic of a 10-period incomplete stepped wedge design when \(51.1\%\) of cluster-period cells have been removed. Bottom-left: Information content of the cells in a complete stepped wedge design with 50 subjects measured in each cluster-period, assuming a discrete-time decay model with \(\rho =0.05\) and \(r=0.95\), with \(90.18\%\) power to detect a standardised effect size of 0.2. Bottom-right: Information content of the cells in the incomplete stepped wedge design depicted in the top-right panel, for the same configuration as the complete design when \(51.11\%\) cluster-period cells removed. The corresponding power is \(88.35\%\) and \(6.02\%\) precision is lost