| Description: |
© 2025 The AuthorsBackground: Although cognitive behavioural therapy (CBT) is a well-established intervention for both the prevention and treatment of depression, its potential role in mentally healthy individuals remains underexplored. We aimed to evaluate the effects of smartphone CBT in mentally healthy adults. Methods: This exploratory secondary analysis from the RESiLIENT trial (UMIN000047124), a master protocol trial consisting of four 2 × 2 factorial trials, included 1,425 adults with no or very mild depressive symptoms (PHQ-9 ≤4). Participants were recruited nationwide and randomly assigned via a centralized web-based system to one of 12 six-week programs (n = 117 to 120 each): five single-skill CBT interventions (Behavioural Activation [BA], Cognitive Restructuring [CR], Problem Solving [PS], Assertion Training [AT] and Behaviour therapy for Insomnia [BI]), 4 combinations in which BA was paired with another skill, and three control conditions (Health Information [HI], Self-check [SC], Delayed Treatment [DT]). HI, serving as the primary comparator, delivered general health information in a format similar to active interventions. The primary outcome was change in depressive symptoms (PHQ-9) from baseline to weeks 6 and 26. Secondary outcomes included anxiety (GAD-7), insomnia (ISI), and well-being (SWEMWBS). Findings: The follow-up rate was 96% at week 6 and adherence to the app was 83%. At week 6, all interventions except PS and BA + PS were superior to HI (SMDs −0.62 to −0.22). At week 26, BA, AT and BI or BA + CR appeared to retain moderate efficacy (SMDs −0.30 to −0.27). BA and AT reduced anxiety, and AT and BI improved sleep quality at week 6. In contrast, CR and PS showed limited effects and lower adherence. There was insufficient evidence to show that the interventions improved overall well-being. No serious adverse events were reported. Interpretation: Single-skill smartphone CBT, especially BA and AT, reduced depressive and anxiety symptoms in healthy adults, supporting its use as a scalable ... |