Effectiveness, Usability, and Satisfaction of a Self-Administered Digital Intervention for Reducing Depression, Anxiety, and Stress in a University Community in the Andean Region of Peru: Randomized Controlled Trial.

Background: University communities, especially in low-resource settings like Peru's Andean region, experience high rates of depression, anxiety, and stress, which harm academic performance and well-being. Traditional mental health services often remain inaccessible, necessitating scalable, self-guided solutions. While digital mental health interventions have shown promise broadly, evidence is scarce for fully self-administered platforms in low-resource university environments. Objective: We evaluated the efficacy of a self-administered digital mental health service in members of a public university to reduce symptoms of depression, anxiety, and perceived stress. Methods: We conducted a double-blind, parallel-group randomized controlled trial with 1:1 allocation to the digital mental health self-care service or a 30-day waiting-list control. We recruited 427 participants (students, teachers, and administrative staff) in May-June 2024, reporting mild to moderate symptoms of depression, anxiety, and stress. Participants were randomized via simple randomization and blinded through automated platform assignment. The intervention comprised 6 sequential 5-day modules grounded in acceptance and commitment therapy, mindfulness, and behavioral activation, delivered via videos, daily text prompts, workbooks, and a responsive chatbot. SMS and WhatsApp (Meta) reminders promoted adherence. Depressive symptoms (Patient Health Questionnaire-9), anxiety symptoms (Generalized Anxiety Disorder-7), and perceived stress (PSS-10) were assessed at baseline and immediately post intervention (day 30). Secondary outcomes in the intervention arm included usability (Computer System Usability Questionnaire), satisfaction (Client Satisfaction Questionnaire-8), and subjective commitment (Twente Engagement with Ehealth Technologies Scales). Analysis of covariance (ANCOVA) adjusted for baseline scores, and multivariate ANCOVA accounted for correlations among outcomes. Effect sizes were quantified using Cohen d and partial epsilon-squared (ε²p). Results: Of 427 randomized, 85 (19.9%) completed all assessments (intervention: n=30; control: n=55). Baseline demographic and clinical characteristics were comparable between groups. Post intervention, the digital mental health self-care service group exhibited significantly greater reductions in mean Patient Health Questionnaire-9 scores (mean difference 2.78; Cohen d=0.64; P=.006), Generalized Anxiety Disorder-7 scores (mean difference 2.13; Cohen d=0.56; P=.015), and PSS-10 scores (mean difference 4.08; Cohen d=0.69; P=.003) than controls. ANCOVA confirmed robust group effects for depression (F₁,₈₂=9.78; P=.002; ε²p=0.31) and anxiety (F₁,₈₂=8.28; P=.005; ε²p=0.32), with a trend toward stress reduction (F₁,₈₂=3.73; P=.057; ε²p=0.46). Multivariate ANCOVA demonstrated a significant multivariate effect (F₄,₁₂=7.23; P=.015). Among intervention completers, 100% scored below the Client Satisfaction Questionnaire-8 satisfaction threshold (<24), 60% rated platform usability as low (Computer System Usability Questionnaire<64); yet, 96.7% reported high subjective commitment (Twente Engagement with Ehealth Technologies Scales ≥18), indicating strong engagement despite interface challenges. Conclusions: A self-administered digital self-care service effectively reduced depression, anxiety, and stress symptoms in a Peruvian university community. High user commitment underscores the platform's relevance, while low satisfaction and usability necessitate interface optimization-streamlined navigation, adaptive personalization, and feedback mechanisms-to enhance user experience and support scalable implementation in low-resource educational settings.