Brief use of behavioral activation features predicts benefits of self-help app on depression symptoms: Secondary analysis of a selective prevention trial in young people.

Abstract

Objective To explore which cognitive behavioral therapy (CBT) self-help app usage predicted depression during a selective prevention trial. Method A recent controlled trial (ECoWeB-PREVENT) randomized young people aged 16-22, at increased risk for depression because of elevated worry/rumination, negative appraisals, and/or rejection sensitivity but without past or current history of major depression, to apps that provided self-monitoring, self-monitoring plus CBT self-help, or self-monitoring plus emotional competency self-help. Self-help included coping strategies for moment-by-moment use (Tools) and self-learning/planning exercises (Challenges). On the primary outcome (depression, Patient Health Questionnaire-9 [PHQ-9]) at 3-months follow-up (primary endpoint), only the CBT app outperformed self-monitoring. In this secondary analysis, only data from participants who used the CBT or self-monitoring apps at least once were analyzed to test what app use predicted change in depression from baseline to 3 months. Results Of the original 1,262 participants (79% female), 558 were included (CBT, baseline, n = 273, PHQ-9: M = 7.48, SD = 3.9; 3 months, N = 163, PHQ-9: M = 8.83, SD = 4.92; self-monitoring, baseline, n = 285, PHQ-9: M = 7.45, SD = 4.26; 3 months, N = 183, PHQ-9: M = 7.48, SD = 3.9). Neither total app use, self-monitoring, nor use of Tools predicted change in depression (all ps > .05). Frequency of use of Challenges predicted lower depression symptoms and caseness at 3 months (β = -0.28, 95% CI [-0.53, -0.03], p = .029). Specifically, the use of behavioral activation challenges mediated the effects of the CBT app on depression over 3 months (β = -0.59, 95% CI [-1.13, -0.05], p = .03). Conclusions Brief psychoeducation about behavioral activation principles in an app may protect young people from depression over 3 months, even when only used once. (PsycInfo Database Record (c) 2025 APA, all rights reserved).