Publication Abstract Display
Type: Poster
Title: Power considerations for crossover and parallel design studies on effects of Cannabis on driving in healthy adults.
Authors: Umlauf A, Wilsey B, Marcotte T, Vaida F
Date: 03-29-2017
Abstract:Conclusions: In this setting, subjects serve as their own control as each visit consists of a baseline measurement and follow-up measurements, with the outcome under each experimental condition being the change from baseline. Thus, compared to the PA design, the crossover design has no clear advantages, but important practical disadvantage, such as learning effects, implementation of wash-over period, and potential participants drop out. Within PA design, MEM analysis has a big power advantage over the SCS analysis. The power of the study depends on the within-subject variation, but not on the between-subject variation in the outcome.

return to publications listing