Patients' preferences in the treatment of depression are important in clinical practice and in research. Antidepressant medication is often prescribed, but adherence is low. This may be caused by patients preferring psychotherapy, which is often not available in primary care. In randomized clinical trials, patients' preferences may affect the external validity. The aim of this article is to study patients' preferences regarding psychotherapy and antidepressant medication and the impact of these preferences on treatment outcome. A systematic review of the literature was performed. The majority of patients preferred psychotherapy in all available studies. Antidepressants were often regarded as addictive and psychotherapy was assumed to solve the cause of depression. Discussing and supporting preferences as part of a quality improvement program of depression care, resulted in more patients receiving the treatment that was most suitable to them. In two patient-preference trials, preferences did not influence treatment outcome. It can be concluded that a substantial percentage of well-informed patients prefer psychotherapy. Patients with strong preferences, mostly for psychotherapy, are likely not to enter antidepressant treatment or randomized clinical trials if their preferences are not supported.