(PS10-A24) Let's Get Engaged! Discussing User Engagement with a Digital Mental Health App

Background: Since the early 2000s, digital mental health interventions (DMHIs) such as apps have been increasingly widespread with potential to expand and transform mental health care. User engagement is considered a precondition of an app’s effectiveness in improving mental health; however, engagement tends to drop off after the first 2 weeks of app use. When DMHI users have higher engagement it leads to more reliable changes in depressive symptoms. The purpose of this study is (1) to examine the subjective experience engaging with a commercially available app for depression and anxiety and (2) better understand user engagement as related to perceptions of changes in mental health symptoms. 

Method: We recruited 20 participants (M=28.5 yrs, 80% female, 75% White; 15% Hispanic/Latino) who previously participated in a parent randomized controlled trial comparing depression and anxiety outcomes between self-guided use of Sanvello with a waitlist control group. We collected qualitative data from these users (March-October 2022) using individual semi-structured interviews conducted over Zoom teleconferencing software. Interviews focused on user experience and engagement with the app, thoughts about the potential of coaching to support app use, and perceived changes in mental health symptoms related to app use. Responses were double-coded according to principles of thematic analysis using Dedoose software. Deductive coding centered on themes of user engagement related to both app features as well as outside factors (e.g., mental health symptoms, daily life issues).

Results: Qualitative themes emerged regarding (1) user engagement, (2) potential effects of coaching to support app use, and (3) relation of app use with mental health symptoms. Users expressed design features (e.g., aesthetics and app structure) and other features (e.g., cost, glitches) that either facilitated or hindered use of the app. Users reported that interactive aspects of the app, including mood tracking and meditation features, were helpful. Users conveyed mixed opinions on coach support. For some, mental health symptoms (anhedonia, motivation) were barriers to self-guided app use. Most participants perceived the app as helpful in reducing or managing mental health symptoms and no participants reported a negative effect on mental health symptoms. 

Discussion: As shown in other work, user engagement varied widely. A key finding was the influence of general life issues and mental health symptoms on engagement and sustained use of the app. Our findings explored both expected themes of barriers and facilitators to engagement and found perceived improvement in mental health symptoms among most app users. Developing programs which include a range of  therapeutic activities could help facilitate real world behavior change while supporting engagement and fulfilling the desire for more interactive in-app experiences. Limitations of this study included a sample that was primarily young, non-Hispanic White women which restricts generalizability of the findings. Further research may consider these findings and continue to explore variation in engagement among app users.