Engagement, acceptability, and effectiveness of the Vira digital behavior change platform: A randomized trial comparing self-care and coach-supported versions in young adults at risk for depression and obesity (Preprint)

Abstract

Adolescence and early adulthood are pivotal stages for the onset of mental health disorders and the development of health behaviors. Digital behavioral activation (BA) interventions, with or without coaching support, hold promise for addressing risk factors for both mental and physical health problems by offering scalable approaches to expand access to evidence-based mental health support.

This two-arm pilot study evaluated two versions of a digital behavioral health product, Vira, for their feasibility, acceptability, and preliminary effectiveness to improve mental health in young adults with depressive symptoms and obesity risk factors.

Seventy-three participants recruited from throughout the United States were randomly assigned to use Vira as either a self-guided product (Vira Self-care) or with support from a health coach (Vira + Coaching) for 12 weeks. The Vira smartphone app utilized passive sensing of behavioral data related to mental health (i.e., activity, sleep, mobility, and language patterns) and offered users personalized insights about patterns of behavior associated with their daily mood. Vira targeted positive behavior changes that can both reduce mental health symptoms and directly impact risk factors for obesity and related disorders. Participants completed self-reported outcome measures at baseline and follow-up (12 weeks). All study procedures were completed virtually.

Both versions of Vira showed strong user engagement, acceptability, and evidence of effectiveness in improving mental health and quality of life. Notably, however, users receiving coaching exhibited more sustained engagement with the platform and reported greater reductions in depression (d = 0.45, 95% CI: 0.10, 0.82) and anxiety (d = 0.50; 95% CI: 0.13, 0.86) compared to self-care users. Both interventions also resulted in reduced stress (coaching: d = -1.05, 95% CI: -1.57, -0.50; self-care: d = -0.78, 95% CI: = -1.33, -0.23) and were perceived as useful and easy to use. Coached users also reported reductions in sleep-related impairment (d = -0.51, 95% CI: -1.00, -0.01). Moreover, participants increased their motivation for and confidence in making behavioral changes, with greater improvements in confidence among coached users.

An app-based intervention using passive mobile sensing to track behavior and deliver personalized insights into behavior-mood associations demonstrated feasibility, acceptability, and preliminary effectiveness for reducing depressive symptoms and other mental health problems in young adults. Future directions include 1) optimizing the interventions, such as improving the self-care version by integrating and automating aspects of the coach support; 2) conducting a fully powered trial that includes an active control condition; and 3) testing mediators and moderators of outcome effects.

ClinicalTrials.gov NCT05638516