withdrawal design
an experimental design in which the treatment or other intervention is removed during one or more periods. A typical withdrawal design consists of three phases: an initial condition for obtaining a baseline, a condition in which the treatment is applied, and another baseline condition in which the treatment has been withdrawn. Often, the baseline condition is represented by the letter A and the treatment condition by the letter B, such that this type of withdrawal design is known as an A-B-A design. A fourth phase of reapplying the intervention may be added, as well as a fifth phase of removing the intervention, to determine whether the effect of the intervention can be reproduced.