Multiscale Modelling Reveals Accelerating Community Outbreak Risks of Measles in the United States

Abstract Measles resurgence in high-income countries that previously achieved elimination reveals a critical surveillance failure: current systems rely on county-level aggregates that obscure fine-scale clustering where outbreaks originate. We assembled the nationwide multiscale vaccination database spanning 45 US states (2013–2025), encompassing over 50,000 schools, 13,000 districts, and 3,000 counties. We developed a gravity-based transmission framework and demonstrate that school-level effective reproduction numbers crossed the epidemic threshold in 2022–2023, a transition invisible to aggregated surveillance. Average susceptibility doubled from approximately 5% to 10% following the COVID-19 pandemic. Three independent mechanisms drive aggregation bias: enrollment-weighted averaging systematically dilutes small under-vaccinated schools; gravity kernels attenuate transmission potential with coarser spatial scales; and clustering signal erasure converts heterogeneous risk landscapes into falsely uniform county averages. School district boundaries frequently misalign with counties, creating cross-boundary transmission corridors that push well-vaccinated counties above epidemic threshold through spillover alone. State trajectories diverge markedly, shaped by exemption policies and local infrastructure. Preventing measles re-endemicity requires surveillance systems operating at the spatial scale where vulnerability accumulates.