Background <p>Evidence-based interventions to reduce adolescents’ exposure to tobacco and air pollution are rarely implemented in low- and middle-income countries, often because they lack contextual fit. The ‘FRESHAIR4Life’ project aimed to optimise implementation of contextually adapted, evidence-based interventions in Greece, Kyrgyzstan, Pakistan, Romania, and Uganda. Selecting appropriate interventions required deliberate, inclusive, and context-sensitive decision-making. This paper outlines the co-creation approach for selecting and adapting interventions, developing implementation strategies and addressing the challenges that arose.</p> Methods <p>Each country followed a four-step process. First, findings from local situational analyses were consolidated to identify barriers and facilitators across individual, community, and system levels and to map targets for change using the Theoretical Domains Framework. This was followed by a workshop with country-teams to identify key risk factors, target behaviours, relevant populations and settings, and to shortlist potential interventions from the FRESHAIR4Life palette alongside implementation strategies using option grids and ranking. A subsequent stakeholder workshop applied a Socio Technical Allocation of Resources informed approach to elicit perspectives on the interventions’ value using qualitative (Conversation Café style discussions) and quantitative methods (ratings of need, importance, likely reach, feasibility, and local adaptations required, fixed budget allocation exercises or ranking of options). Finally, selection meetings with country teams finalised intervention choices and implementation strategies, informed by stakeholder input.</p> Results <p>Countries chose to target the risk factors that were considered most prevalent and the biggest threat to adolescent health in their settings. Of the four countries addressing tobacco use, Pakistan, Greece, and Romania selected ‘implementation intentions (If-Then approach)’ for prevention, while Kyrgyzstan selected ‘very brief advice plus behavioural counselling (VBA+)’ for cessation. All three countries addressing air pollution, Pakistan, Uganda and Kyrgyzstan selected social media campaigns. Key challenges included: (1) establishing a sound basis for selection decisions, which was overcome by adopting a systematic, transparent and inclusive decision process (2) ethical and practical considerations for addressing tobacco/nicotine use within school-based prevention, tackled by adapting intervention to include offer of cessation support (3) limited evidence-based interventions targeting ambient air pollution, addressed by flexibly moving beyond the palette (4) the evolving nature of implementation strategies and (5) embedding interventions into routine practice, both addressed through flexibility and adaptation.</p> Conclusion <p>This study shows that, across diverse country contexts, defensible intervention selection depends not only on which intervention is chosen, but on how choices are made and by whom. A transparent, stakeholder-owned co-creation process, involving the target audience of adolescents, helped country teams navigate trade-offs between evidence, local burden, feasibility, fit with routine practice, and ethical obligations, while exposing important evidence gaps for individual- and community-level ambient air-pollution interventions in LMICs.</p>