Preliminary clinical changes during use of a personalized video-based mobile intervention for community-dwelling individuals with severe mental illness: An exploratory randomized crossover trial.

Severe mental illness (SMI) often limits participation in traditional therapy due to motivational and cognitive deficits. We examined short-term clinical changes during use of a personalized video-based mobile app (MARO)-providing psychoeducation, emotion regulation, cognitive restructuring, and behavioral activation-in community-dwelling individuals with SMI and explored whether the timing of minimal human coaching was associated with these outcomes. This 8-week exploratory randomized crossover trial compared an Immediate-Coaching group (app + coaching → app only) and a Delayed-Coaching group (app only → app + coaching); both groups used MARO throughout the study. Among 40 randomized participants, 27 completed all assessments. Depression (Patient Health Questionnaire-9), anxiety (Generalized Anxiety Disorder-7), and positive automatic thoughts (Automatic Thoughts Questionnaire-Positive) were assessed at baseline and after 4 and 8 weeks. Both groups showed decreases in PHQ-9 and GAD-7 scores over the 8-week intervention period, while ATQ-P scores increased. Overall PHQ-9 scores were lower and ATQ-P scores were higher in the Immediate-Coaching group than in the Delayed-Coaching group. These preliminary findings support further evaluation in larger, adequately powered controlled studies of personalized video-based mobile interventions as a potentially scalable digital-human approach to community-based care for individuals with severe mental illness. Trial Registration: CRIS, KCT0011492.