A crossover study is a type of randomized controlled trial in which the participants are allocated to different study arms in which they receive a sequence of two or more treatments. The simplest version of a crossover study involves two study arms — 1 and 2 — each with two treatments — A and B. Half of the study participants are assigned to arm 1 and receive treatment order A/B while the other half are assigned to arm 2 and receive treatment order B/A. After receiving the first treatment, the participants then receive the other treatment after a washout period. This washout period serves the purpose of eliminating any carryover effects from the first treatment.

Crossover trials allow the response of the participants to each treatment to be compared against themselves. This removes the confounding effects of individual variation and makes crossover studies more efficient and precise at estimating treatment effects than parallel group trials, in which the participants are only exposed to one treatment. Additionally, crossover studies often require a smaller sample size than parallel group studies.