Adaptive Random Assignment

The primary objective of a comparative trial is to provide a precise and valid treatment comparison. Another objective may be to minimize exposure to the inferior treatment, the identity of which may be revealed only during the course of the study. Unrestricted randomization is not generally used because it is susceptible to chronological bias, and this would interfere with the first objective, the valid treatment comparison. Yet, too many restrictions lead to a predictable allocation sequence, which can also compromise validity. It can be a challenge to find the right balance of restrictions on the randomization, and sometimes a covariate-adaptive design is used. When the second objective is considered as well, then the solution must necessarily be a response-adaptive design that evolves depending on the responses of those enrolled. Keywords: covariate-adaptive randomization; response-adaptive randomization