Search for a command to run...
Background Adolescent mental health problems represent a growing public health concern worldwide. Physical activity (PA) is a modifiable protective factor that supports emotional and social wellbeing, yet most research has emphasized the amount or intensity of activity rather than the context in which it occurs. Comparing structured settings (for example, PE, team sport, classroom programs, supervised dance/arts) and unstructured settings (for example, self-directed leisure, community, or nature-based opportunities) may help identify which forms of engagement best promote mental health and guide scalable interventions. Objective To synthesize recent evidence on associations between physical activity contexts and mental health outcomes among adolescents, and to compare structured vs. unstructured approaches. Methods A systematic review with narrative synthesis (without meta-analysis) was conducted following PRISMA 2020. Eligible designs included cross-sectional, cohort, quasi-experimental, and qualitative studies. Risk of bias was appraised using JBI (cross-sectional/cohort), ROBINS-I (non-randomized), and CASP-Qualitative. Narrative synthesis followed SWiM guidance for both quantitative and qualitative evidence, and certainty of evidence was summarized using GRADE and CERQual. Results Across contexts, associations were predominantly favorable for wellbeing/affect, self-concept/competence, and prosocial/connectedness, with more mixed patterns for depression/anxiety and resilience. Effects were moderated by environmental quality, supervision, and autonomy support. Certainty ranged from very low to moderate due to the predominance of observational designs and reliance on self-report measures. Conclusions The available evidence suggests that policy and practice may benefit from combining competence-building, structured opportunities with safe, autonomy-supportive, unstructured, and nature-based options in schools and communities. These implications should be interpreted with caution, given the predominance of observational study designs and the very low to moderate certainty of the evidence. Reporting environmental fidelity and motivation-related constructs may enhance interpretability and support translation into routine mental health promotion services.