A0664
Title: A caution regarding sample size calculations for stepped-wedge designs with binary outcomes
Authors: Justine Shults - University of Pennsylvania (United States) [presenting]
Abstract: Stepped-wedge designs are widely used when it is difficult to implement traditional parallel designs. The purpose is to discuss and demonstrate some of the challenges faced in the design phase because the primary outcomes for both studies are dichotomous. In particular, accounting for potential temporal trends is more challenging for dichotomous outcomes when doing sample size calculations than it is when the primary outcome is continuous.
