A method and system for performing a clinical trial having a reduced placebo effect is disclosed. The method includes randomizing study participants into three or more treatment groups and performing a first phase of testing on the groups. In a typical embodiment, the first phase of testing includes administering an active treatment to a first group, and administering a placebo to a second group and to a third group. Responders and non-responders are determined for each group. A second phase of testing is then performed. The second phase of testing includes administering the placebo to non-responders in the first group, administering the active treatment to non-responders in the second group, and administering the placebo to non-responders in the third group. The data from the first phase of testing and from the second phase of testing is pooled and analyzed to determine response rates to active treatment and placebo.
