Digital interventions can increase the reach and continuity of physical activity promotion, but evidence remains fragmented across mental health and addictive behaviors. We conducted a comprehensive integrative review supported by structured searches (2015-2026) in biomedical, psychological, multidisciplinary and technology-oriented databases, complemented by backward/forward snowballing. Eligible studies included digital interventions in which physical activity (or sedentary reduction) was a core component and those that reported mental health outcomes (e.g., depression, anxiety, stress, and well-being) and/or addiction-related outcomes (e.g., craving, consumption, lapses/relapse, and treatment retention). We synthesized findings thematically by intervention typology (apps, wearables, hybrid models with human support, and adaptive approaches) and by key active ingredients (goal setting, self-monitoring, feedback, reinforcement, planning, and engagement strategies).

Overall, most studies targeted mental health outcomes and used app-based multicomponent programs, sometimes complemented by wearables, with generally short follow-up and heterogeneous engagement metrics. Evidence in addictions was more context-specific and concentrated in alcohol treatment and opioid agonist therapy settings, supporting feasibility and a plausible role for physical activity as a coping strategy. Advanced personalization frameworks (EMA/EMI/JITAI) provide a clear implementation pathway but are less consistently operationalized when physical activity is the central therapeutic component. This review highlights practical design recommendations and research priorities for scalable, safe, and equity-oriented digital physical activity interventions in mental health promotion and relapse prevention.