User Character Strengths and Engagement Prediction on a Digital Mental Health Platform for Young People: Longitudinal Observational Study.

Background: Mental ill health is a leading cause of disability worldwide, but access to evidence-based support remains limited. Digital mental health interventions offer a timely and low-cost solution. However, improvements in clinical outcomes are reliant on user engagement, which can be low for digital interventions. User characteristics, including demographics and personality traits, could be used to personalize platforms to promote longer-term engagement and improved outcomes. Objective: This study aims to investigate how character strengths, a set of positive personality traits, influence engagement patterns with moderated online social therapy, a national digital mental health platform offering individualized, evidence-based digital mental health treatment for young people aged 12-25 years. Methods: Data from 6967 young people who enrolled with moderated online social therapy between August 2021 and July 2023 were analyzed. Longitudinal analyses were used to investigate whether scores on 3-character strength dimensions ("social harmony," "positive determination," and "courage and creativity") were associated with (1) an accelerated or decelerated rate of dropout from the platform and (2) patterns of engagement over the first 12 weeks following onboarding. Engagement metrics were time spent on the platform, number of sessions on the platform, use of the embedded social network, and messages with the clinical team. Results: On average, young people used the platform for 72.64 (SD 106.64) days. The 3-character strengths were associated with distinct engagement patterns during this time. Individuals scoring higher on "social harmony" demonstrated an accelerated dropout rate (coefficient=-0.15, 95% CI -0.26 to -0.04; P=.008). Interestingly, higher scores on this character strength were associated with high rates of initial engagement but a more precipitous decline in platform use over the first 12 weeks, in terms of time spent on the platform (β=-.01; SE 0.00; t2748=-5.05; P<.001) and the number of sessions completed (β=-.00; SE 0.00; t2837=-2.26; P=.02). In contrast, higher scores on "positive determination" and "courage and creativity" predicted more modest initial platform use but steadier engagement over time, in terms of time spent on the platform ("positive determination": β=.01; SE 0.00; t2748=4.05; P<.001 and "courage and creativity": β=.01; SE 0.00; t2748=2.66; P=.008). Contrary to our predictions, character strengths did not predict use of the embedded social network or the number of messages sent to the clinical team. Conclusions: Our findings illustrate how character strengths predict distinct engagement trajectories on a digital mental health platform. Specifically, individuals higher on "social harmony" showed high initial engagement that quickly declined, while those higher on "positive determination" and "courage and creativity" demonstrated lower initial engagement but a steadier use of the platform over time. The findings of this study demonstrate an opportunity for digital mental health interventions to be tailored to individual characteristics in a way that would promote greater initial and ongoing engagement.