The Effect of Mental Health App Customization on Depressive Symptoms: A Randomized Controlled Trial (Preprint)

BACKGROUND Mental health apps have shown promise for improving mental health symptoms, including depressive symptoms. However, limited research has been aimed at understanding how specific app features and designs can optimize the therapeutic benefits and adherence of such mental health apps. OBJECTIVE The primary purpose of this study was to investigate the effect of avatar customization on depressive symptoms and adherence to use a novel cognitive behavioral therapy-based mental health app. Furthermore, the secondary aim was to examine whether specific app features, including journaling, mood tracking, and reminders, impact the usability of the mental health app. METHODS College students (N=83) participated in a randomized controlled trial in which participants were randomized to either a customization or no-customization version of the app. Customization involved personalizing a virtual avatar and a virtual travel vehicle and use of one’s name throughout the app. Participants completed a 14-day trial with a novel cognitive behavior therapy-based mental health app called AirHeart. Self-report scores of depressive symptoms, anxiety, and stress were measured at baseline and post-intervention. Post-intervention survey measures also included usability and avatar identification questionnaires. RESULTS AirHeart app use significantly reduced symptoms of depression (P = .006) from baseline to the end of the two-week intervention period for all participants regardless of customization condition. However, no differences in depressive symptoms (P = .165) or adherence (P = .799) were observed between the customization and no-customization conditions. Frequency of journaling, usefulness of mood tracking, and helpfulness of reminders were not associated with change in depressive symptoms or adherence (Ps > .10). Exploratory analyses showed that there were three moderate positive correlations between avatar identification and depressive symptoms (identification: r = -.312, P = .018), (connection: r = -.305, P = .021), and (lack of relatability: r = .338, P = .010). CONCLUSIONS These results indicate that cognitive behavioral therapy mental health apps have the potential to reduce depressive symptoms over a brief intervention period. RCT results demonstrated that customization of app features like avatars does not further reduce depressive symptoms over and above the cognitive behavioral therapy modules. However, further research elucidating the relationship between virtual avatar identification and mental health systems is needed as society becomes increasingly more digitized. These findings have potential implications for improving the optimization of mental health app design. CLINICALTRIAL https://osf.io/7wv9s/