THE IMPACT OF SOCIAL MEDIA ALGORITHMS AND MHEALTH INTERVENTIONS ON THE DETECTION, PROGRESSION, AND RECOVERY OF DEPRESSION: A LITERATURE REVIEW

Abstract

Background: The rapid digitalization of daily life has established the "digital exposome" as a major structural determinant of mental health. While commercial social media algorithms present significant risks for the development of depression, this same digital ecosystem simultaneously offers novel tools for psychiatric screening and therapy. Objectives: This review explores the double-edged nature of digital environments in affective disorders. We aim to clarify how algorithmic exposure drives depressive symptoms, evaluate the potential of artificial intelligence (AI) in early diagnosis, and assess the true clinical value of Mobile Health (mHealth) interventions in psychiatric recovery. Methodology: Guided by PRISMA 2020 standards, we systematically synthesized peer-reviewed literature from major scientific databases published between January 2015 and July 2026. The analysis utilized a narrative synthesis framework to categorize findings into three core areas: detection, progression, and recovery. Results: Longitudinal data reveal that heavy algorithmic exposure actively drives the progression of depression, primarily through passive content consumption, upward social comparison, and severe sleep disruption. On the diagnostic front, AI-driven digital phenotyping shows high accuracy in detecting early depressive markers, though it remains limited by ethical and privacy concerns. For treatment, guided internet-based cognitive behavioral therapy (iCBT) proves highly effective, contrasting sharply with the unregulated and often clinically unvalidated commercial app marketplace. Conclusion: The traditional biopsychosocial model must be expanded to include the digital dimension. To combat the digital depression crisis, clinicians must routinely assess patients' digital habits through a "digital anamnesis," while public health efforts must evolve from ineffective short-term "digital detoxes" toward comprehensive digital well-being education and strict algorithmic regulation.