Background: Controversy persists about how effectively empirically-supported treatments for major depression work in actual clinical practice as well as how patients choose among them. We examined the acute phase effectiveness of cognitive therapy (CT), interpersonal psychotherapy (IPT), and combined psychotherapy-pharmacotherapy (PHT) in a naturalistic setting, allowing patients their choice of treatment. Methods: The study compared CT (n=63), IPT (n=56), CT-PHT (n=34), and IPT-PHT (n=21) for 174 subjects with major depression in a secondary care mood disorders clinic. Patient preference, rather than randomization, determined treatment selection. The Beck Depression Inventory-II (BDI) was the primary outcome variable. Exclusion criteria were minimal. Results: All treatments were associated with a reduction in depressive symptoms, with a 35% remission rate by week 26. Overall improvement was well within ranges reported in efficacy trials. On average, treatment effects of the different interventions straddled the same range, but moderation analyses revealed that BDI scores dropped faster in the first 16 weeks in patients who received CT alone than patients who received CT and pharmacotherapy, a pattern not found in patients who received IPT (with or without pharmacotherapy). Limitations: Limitations consist of a modest sample size, choice of treatment was made by participants which may have been influenced by many sources, and the absence of a non-active control group. Conclusions: This study supports the effectiveness of empirically-supported antidepressant treatments selected by patients in routine settings, and provides an indication that speed of therapeutic response may vary amongst treatments. (c) 2012 Elsevier B.V. All rights reserved.