The prevention of mood episodes is an important goal of the maintenance treatment of patients with bipolar disorder. The rate of relapse on placebo compared with that on active treatment is an important issue in the design of future clinical trials of maintenance treatment. We examine the range and time course of placebo relapse rates in studies of patients with bipolar I disorder. In addition, we address the potential variables associated with placebo response, strategies to minimize placebo response, the optimum duration of placebo-controlled maintenance trials, possible alternatives to placebo control groups, and the impact of these considerations in maintenance studies of children, adolescents, and older adults with bipolar disorder.