CMStatistics 2018: Start Registration
View Submission - CMStatistics
B1368
Title: Design and analysis of pragmatic stepped-wedge clustered randomization trials Authors:  Song Zhang - University of Texas Southwestern Medical Center (United States) [presenting]
Abstract: The stepped-wedge cluster randomized trial design is particularly suitable and has been frequently adopted by pragmatic trials. Under this design, initially all clusters receive control treatment. Subsequently, at pre-defined time points (steps) clusters are randomized to switch to intervention. Outcomes are measured at every step. All clusters receive the intervention at the end of study. We investigate the design and analysis of data arising from stepped-wedge trials under pragmatic situations such as missing data, uneven steps, various types of correlation structures, random cluster sizes, etc. Simulation studies and application examples to real clinical trials are presented.