Inferential instability of national sugar and sweetener availability as an indicator of adult obesity trajectories: A global within-between panel audit

The analysis uses a global 2010-2023 panel comprising 3,038 economy-years across 217 economies. It explicitly separates between-economy and within-economy estimands and tests the longitudinal interpretation using an identical-sample temporal analysis with cluster-aware coefficient contrasts, a formal isometric log-ratio sensitivity analysis, independent fixed-effects replication, and wild-cluster-bootstrap inference. The central finding is deliberately calibrated: cross-economy agreement cannot validate national sugar availability for longitudinal obesity surveillance. The study identifies temporal and construct instability without claiming that sugar is protective or that the mechanisms producing the instability have been identified. The manuscript aligns well with PLOS ONEs emphasis on technically sound, transparent and reproducible research of broad relevance. All data required to reproduce the findings, complete metadata, executable code, full-precision results, diagnostic outputs and a completed STROBE checklist are provided as S1-S5. Figures are provided separately as compliant 350-dpi TIFF files. The study used only publicly available, aggregated economy-year statistics and involved no individual participants, identifiable information or biological specimens; institutional ethics review and consent were therefore not required. This is original work; it is not under consideration elsewhere, and the sole author has approved the submission and accepts responsibility for its content. Funding and competing-interest declarations will be entered accurately in the submission portal. An Academic Editor with expertise in nutritional epidemiology, global health metrics, longitudinal panel methods, or food-system surveillance would be well placed to assess the work.