CONTEXT: Only about 60% of adolescents with depression will show an adequate clinical response to an initial treatment trial with a selective serotonin reuptake inhibitor (SSRI). There are no data to guide clinicians on subsequent treatment strategy. OBJECTIVE: To evaluate the relative efficacy of 4 treatment strategies in adolescents who continued to have depression despite adequate initial treatment with an SSRI. DESIGN, SETTING, AND PARTICIPANTS: Randomized controlled trial of a clinical sample of 334 patients aged 12 to 18 years with a primary diagnosis of major depressive disorder that had not responded to a 2-month initial treatment with an SSRI, conducted at 6 US academic and community clinics from 2000-2006. INTERVENTIONS: Twelve weeks of: (1) switch to a second, different SSRI (paroxetine, citalopram, or fluoxetine, 20-40 mg); (2) switch to a different SSRI plus cognitive behavioral therapy; (3) switch to venlafaxine (150-225 mg); or (4) switch to venlafaxine plus cognitive behavioral therapy. MAIN OUTCOME MEASURES: Clinical Global Impressions-Improvement score of 2 or less (much or very much improved) and a decrease of at least 50% in the Children's Depression Rating Scale-Revised (CDRS-R); and change in CDRS-R over time. RESULTS: Cognitive behavioral therapy plus a switch to either medication regimen showed a higher response rate (54.8%; 95% confidence interval [CI], 47%-62%) than a medication switch alone (40.5%; 95% CI, 33%-48%; P = .009), but there was no difference in response rate between venlafaxine and a second SSRI (48.2%; 95% CI, 41%-56% vs 47.0%; 95% CI, 40%-55%; P = .83). There were no differential treatment effects on change in the CDRS-R, self-rated depressive symptoms, suicidal ideation, or on the rate of harm-related or any other adverse events. There was a greater increase in diastolic blood pressure and pulse and more frequent occurrence of skin problems during venlafaxine than SSRI treatment. CONCLUSIONS: For adolescents with depression not responding to an adequate initial treatment with an SSRI, the combination of cognitive behavioral therapy and a switch to another antidepressant resulted in a higher rate of clinical response than did a medication switch alone. However, a switch to another SSRI was just as efficacious as a switch to venlafaxine and resulted in fewer adverse effects. TRIAL REGISTRATION: clinicaltrials.gov Identifier: NCT00018902.