Crossover Design

In randomized trials, a crossover design is one in which each subject receives each treatment, in succession. For example, subject 1 first receives treatment A, then treatment B, then treatment C. Subject 2 might receive treatment B, then treatment A, then treatment C. A crossover design has the advantage of eliminating individual subject differences from the overall treatment effect, thus enhancing statistical power. On the other hand, it is important in a crossover study that the underlying condition (say, a disease) not change over time, and that the effects of one treatment disappear before the next is applied. See also Parallel design.