Background: Social media is firmly embedded in adolescents' everyday lives, with 96% of 15-year-olds across the EU reporting daily use in 2022. Alongside potential benefits for identity formation, social participation and support, social media use has also been associated with risks, including depressive and anxiety symptoms, adverse effects on sleep and body image, and problematic use. However, the evidence largely stems from cross-sectional studies based on self-reported data and shows small to moderate associations. Regulatory measures are increasingly being implemented or proposed internationally, including a planned minimum age of 14 for social media use announced in Austria in March 2026. The aim of this report was to inform health policy decision-making through a systematic synthesis of existing measures, their effectiveness and safety, and their ethical implications.

Methods: Three intervention types for individuals under 19 were assessed: full regulatory bans, partial restrictions (including family-mediated approaches) and educational programmes. A structured manual search identified existing regulations across 23 countries. This was complemented by a systematic review in five databases. Outcomes were selected through scoping the literature, a school survey (n=39) and stakeholder consultation. Study selection, risk of bias assessment and data extraction were carried out independently by two persons and synthesis was performed narratively. Social and moral aspects were analysed using the EUnetHTA Core Model and presented descriptively.

Results: 45 documents covering 38 regulations were identified, predominantly age-based access restrictions (mostly ≤16 or ≤18 years) and limits on platform functions and content. Implementation was primarily carried out through age verification by the platform providers. The review included seven completed and nine ongoing studies. No completed evaluations of full bans were identified. A one-hour daily use limit demonstrated small improvements in mental health, body image and sleep duration. A school-level smartphone ban showed no between-group differences. Parental role modelling and educational programmes were associated with lower problematic use. School-level smartphone bans and media literacy training showed no meaningful effects on mental health. The ethical analysis (42 documents) addressed the balance between rights to protection and participation, as well as the benefits and harms for health and development; concerns included restrictions on autonomy, disadvantages for vulnerable groups, data protection risks, implementation problems, and a shift of responsibility away from platforms.

Discussion: The transferability of the evidence to the Austrian proposal is limited: no studies on age-based bans were identified, and available findings mostly concern partial restrictions with a different mechanism of action, short follow-up periods and self-reported outcomes. Austria's threshold of 14 years also falls below the 15 or 16 years predominantly adopted internationally. The absence of robust evidence does not in itself argue against regulation, but it increases the importance of ethical-normative justifications.

Conclusion: The effectiveness of specific regulatory approaches can currently neither be confirmed nor ruled out. Their potential effectiveness depends on successful implementation, particularly on reliable age assurance. Multi-level strategies, an ethical proportionality assessment, accompanying communication measures and an independent, pre-defined evaluation framework are required.