Estimating the carbon footprint of oral semaglutide (Rybelsus): a comparative analysis of the medicine carbon footprint model and manufacturer life-cycle assessment in pharmaceutical production.

OBJECTIVES: Pharmaceuticals are a major contributor to healthcare's greenhouse gas emissions yet product-level carbon footprints are rarely disclosed. Glucagon-like peptide-1 receptor agonists (GLP-1 RAs) are increasingly being used for the management of type 2 diabetes and people living with obesity, but their carbon impact remains poorly characterised. We aimed to compare carbon footprint estimates for oral semaglutide (Rybelsus), a peptide-based medicine, generated using the medicine carbon footprint (MCF) framework and manufacturer-reported life cycle assessment (LCA) data. DESIGN AND SETTING: Comparative methodological analysis using cradle-to-gate boundaries to compare pharmaceutical carbon footprint estimates for oral semaglutide. METHODS: Per-tablet MCF estimates for 3, 7 and 14 mg Rybelsus were annualised and compared with product-specific LCA results reported by Novo Nordisk. Differences between methods were assessed descriptively across the three dose strengths. Absolute differences and symmetric percentage differences were calculated. RESULTS: Annualised MCF estimates were 8.9, 20.0 and 39.3 kg CO₂e/year for 3, 7 and 14 mg tablets, respectively. Corresponding EU LCA estimates were 17.0, 25.2 and 39.9 kgCO₂e/year. The mean symmetric percentage difference was -29%, with the largest discrepancy observed at the 3 mg dose (-62.6%) and the smallest difference at the 14 mg dose (-1.5%). CONCLUSION: For oral semaglutide, MCF generated lower carbon footprint estimates than manufacturer-reported LCA values, particularly at lower doses, where fixed device/excipient and packaging emissions are proportionally larger. MCF offers a rapid, accessible and indicative estimation method when proprietary data are unavailable. Improved transparency and harmonised boundaries would strengthen comparability between pharmaceutical carbon footprint estimation approaches. These findings are specific to the Rybelsus formulation of oral semaglutide and may not be generalisable to other formulations or GLP-1 RAs.