Background: Several studies have described real-world prescription patterns of first-line antidepressants for depression but little is known about their fate in terms of duration, intensity and changes. Methods: An inception cohort of new onset non-psychotic depression initiating antidepressant treatment with a new generation antidpressive agent was identified in a large health insurance claims database in Japan between 2009 and 2010. The duration and intensity of first-line antidepressants, the timing and kind of second-line antidepressants and the total duration of antidepressant treatment were examined. Results: We identified 1592 patients. The starting dose and the maximum dose attained with the firstline agent appeared to be largely in line with the guideline recommendations although the latter tended toward the minimum of the recommended range. The continuity of the first-line antidepressant was far below the guideline recommendations, with 28%never returning after the initial prescription and 55%dropping out within 3 months. Of all the first-line antidepressants, 14%were subsequently augmented by another psychotropic agent while 17%were switched to another antidepressant after a median of 3 or 2 months, respectively. The choice of the second-line agents varied extremely widely. The total duration of antidepressant therapy was as short as a median of 4 months, with 68%stopping treatment by 6 months. Limitations: The diagnosis of non-psychotic unipolar depression in the claims database analyses remains approximate. Conclusions: The current guidelines are grossly out of touch with the clinical realities. On the one hand, guidelines need to reflect the real-world practices
on the other hand clinicians should limit their treatment options and allow evidence-based comparative effectiveness research among them so that patients shall no longer be given less effective and more effective treatments without being able to distinguish among them. © 2013 Elsevier B.V. All rights reserved.