A Comparison of Sample-Size Calculations for Cluster-Randomized Crossover Trials with a Binary Outcome

• Activity Number: 611
• Type: Contributed
• Date/Time: Wednesday, August 12, 2015 : 2:00 PM to 3:50 PM
• Sponsor: Section on Medical Devices and Diagnostics
• Author(s): Rui Zhuang* and Erin Case and Siobhan Brown and Susanne May
• Companies: University of Washington and University of Washington and University of Washington and University of Washington
• Keywords: cluster randomized crossover trial ; binary outcome ; sample size calculation

Abstract

We compared the closed form equation and simulation methods currently available for calculating the sample size for a cluster randomized crossover trial with a binary outcome from Giraudeau, et. al. (2008, Statistics in Medicine, 27, p5578) and Reich, et. al. (2012, PLoS ONE, 7(4), p.E35564), respectively. We explored how cluster size, period effect, and intracluster correlation (ICC) impact the sample size. The methods were compared in hypothetical study scenarios developed from the Resuscitation Outcomes Consortium (ROC) setting, and compared to a simplified sample size calculation used there (the "T-BOSS" method). The simulation and closed form equation were fairly close in the setting of ROC, but diverged in some other scenarios we explored. We recommend when each method would be appropriate when planning a cluster randomized crossover trial with a binary outcome.