Patients' preferences for the treatment of anxiety and depressive disorders: a systematic review of discrete choice experiments

Introduction: Matching available mental health services to patients' preferences, as well as is possible, may increase patient satisfaction and help increase adherence to certain treatments. This study systematically reviewed discrete-choice experiments (DCEs) on patients' preferences for treatment of depressive and anxiety disorders and assessed the relative importance of outcome, process and cost attributes to improve the current and future treatment situations. Methods: A systematic literature review using PubMed, EMBASE and PsychInfo was conducted to retrieve all relevant DCEs published up to 15 April 2019, eliciting patient preferences for treatment of depressive and anxiety disorders. Data were extracted using an extraction sheet, and attributes were classified into outcome, process and cost attributes. The relative importance of each attribute category was then assessed, and studies were evaluated according to their reporting quality, using validated checklists. Results: A total of 11 studies were identified for qualitative analysis. All studies received an aggregate score of 4 on the five-point PREFS checklist (Purpose, Respondents, Explanation, Findings and Significance). Most attributes were outcome related (52%), followed by process (42%) and cost (6%) attributes. Comparing the attribute categories and summing up the relative importance weights for each category within the studies, process attributes were ranked as most important, followed by cost and outcome attributes. Conclusions: In this systematic review, heterogeneous results were observed regarding the inclusion and framing of different attributes across studies. Overall, patients considered process and cost attributes to be more important than outcome attributes. Outcomes and process are important for patients, and thus clinicians should be particularly aware of this and take patients' preferences into account, although the attribute importance may depend on chosen attributes and related levels.