Browse the glossary:
In a study where one comparison group receives a treatment, for example, a specific type of advice or surgery, and the other comparison group does not receive that treatment, contamination occurs when people allocated to not receiving the treatment receive it, or people allocated to receive the treatment do not receive it.
Sometimes, in cluster randomized studies, clusters (groups) of people are allocated to treatments rather than individuals, to reduce the risk of contamination.
If you feel that this definition hasn't helped you to understand the term, click on our monkey to let us know.