Document resource
Background Depression is a leading cause of global disability, and subthreshold cases contribute substantially to the burden. Cognitive behavioural therapy (CBT) is effective for depressive symptoms, but its impact on social functioning in subthreshold depression remains unclear.Objective This study evaluated the effects of smartphone-based CBT on social functioning and personal health records among individuals with subthreshold depression and examined whether a personalised and optimised therapy (POT) algorithm enhanced these outcomes.Methods We performed a secondary analysis of the RESiLIENT trial (n=5361). Participants were stratified by baseline Patient Health Questionnaire-9 (PHQ-9) scores (0–4 and 5–14) and randomised to nine CBT app groups or a self-check group. Outcomes included social functioning and health records. A mixed-effects model for repeated measures was used to analyse the outcomes. The POT algorithm selected CBT skills tailored to individual characteristics.Findings Depressive symptoms improved in both PHQ-9 groups compared with those of the self-check group, but gains in social functioning were modest, observed mainly in the PHQ-9 score ≤4 group. No consistent benefits were seen for health behaviours, job resignation, sick leave or healthcare costs. However, in a simulated randomised comparison using 10-fold cross-validated datasets, the POT algorithm had potentially beneficial effects for productivity (standardised mean difference (SMD)=0.16, 95% CI 0.03 to 0.29) and social adjustment (SMD=−0.13, 95% CI −0.26 to 0.00).Conclusions Smartphone CBT effectively reduced depressive symptoms but showed limited impact on social functioning and healthcare costs. Personalised optimisation may modestly improve productivity and adjustment, suggesting its promise for enhancing social outcomes.Clinical implications Personalised tailoring of CBT skills may strengthen the effects of digital interventions on social functioning.Trial registration number UMIN000047124.