TADS is a randomized controlled clinical trial that will compare the effectiveness of established treatments---cognitive-behavioral psychotherapy, medication management, and their combination---for adolescents suffering from major depressive disorder (MDD).

The experimental design consists of three treatment stages and a follow-up phase. Stage I (12 weeks) is a four-group randomized comparison of four treatments: antidepressant medication alone (Fluoxetine); psychotherapy alone (CBT); a combination of the medication and psychotherapy (Comb); and a placebo control medication condition (Pbo). Stage II (six weeks) is a treatment consolidation phase in which we ask whether longer treatment in responders and higher intensity treatment in partial responders to their Stage I treatment would be helpful. Non-responders at the end of Stage I will be referred to open community treatment, or for ethical and practical reasons in the case of non-responders to Placebo, to open treatment of their choice with one of the three active study treatments administered by the study team. Responders at the end of Stage I advance to 6 weeks of maintenance treatment in their assigned arm. Partial responders to CBT receive an additional 6 weeks of CBT in their assigned arm; partial responders to Fluoxetine may receive a higher dose for six weeks. Partial responders to the Comb treatment will receive an additional 6 weeks of CBT and may receive a higher dose of medication for six weeks. Stage III (18 weeks) is a treatment maintenance phase for those teenagers who have continued to respond well. Treatment will be continued and progress will be monitored. Stage IV (one year) is an assessment-only follow-up phase to help us understand the long-term benefits of the treatments.

The recruitment strategy is designed to enter into treatment a volunteer clinical sample of 432 teenagers, both males and females, ages 12 to 17, at ten sites. A multiple gating procedure will be used in which patients will be screened, assessed for study eligibility, and if eligible, consented before randomization to one of the four treatment groups. Patients will be selected without regard to race, gender, or ethnicity and it is expected that the sample will match patients seen in general clinical practice. Patients will be recruited from multiple sources including: mental health identified children, i.e., children already coming to a clinic; primary care identified children (pediatric and family physicians); teacher or school identified children (i.e., school refers through the parents or primary caretaker); and families who self-refer.