Beyond the Hospital: Spatial Inequality and Structural Divergence in Stroke Mortality Across the United States (2020-2024).

BACKGROUND: Stroke is a time-critical cerebrovascular emergency where survival depends heavily on rapid prehospital recognition and transport. While in-hospital acute stroke care has advanced significantly, the structural distribution of where stroke deaths occur, particularly out-of-hospital, remains a critical blind spot. The COVID-19 pandemic severely disrupted emergency medical services (EMS) and patient care-seeking behaviors, but the longitudinal geographic and structural legacy of this disruption on terminal stroke care is undercharacterized. METHODS: We analyzed Centers for Disease Control and Prevention Wide-Ranging Online Data for Epidemiologic Research (CDC WONDER) mortality data (2020-2024) for adults aged ≥45 years with stroke as the underlying cause of death. Geographic redistribution was evaluated across National Center for Health Statistics (NCHS) urbanization categories using unadjusted mortality counts. Changes in terminal place of death were tracked longitudinally. To evaluate geographic concentration, we calculated empirical Bayes-smoothed standard mortality ratios for state-level data. Spatial inequality for in-hospital versus out-of-hospital mortality was quantified using Lorenz curves and population-weighted Gini indices, with differences assessed via bootstrap resampling. RESULTS: From 2020 to 2024, stroke mortality showed a distinct geographic redistribution, which we operationalize as suburbanization, characterized by the largest relative increases in medium (+8.4%) and large fringe metros (+7.3%), alongside stagnation in large central metros and slight declines in noncore areas. A significant structural shift occurred in terminal care, highlighted by an acute surge in home-based deaths during 2021 (n=41,423) and a continuous rise in hospice care. Spatial analysis revealed substantial geographic concentration for both out-of-hospital (Gini=0.5321) and in-hospital stroke mortality (Gini=0.5045), reflecting persistent, systemic regional disparities across the entire stroke-care continuum. CONCLUSIONS: While suburban and mid-sized metros exhibited the largest absolute increases in out-of-hospital stroke deaths, this spatial redistribution is based on unadjusted counts and may reflect demographic shifts rather than individual-level risk changes. Nevertheless, the high geographic concentration of stroke deaths suggests underlying structural vulnerabilities in both prehospital and regional healthcare networks. Integrating spatial Gini indices into stroke surveillance identifies vulnerable regions requiring targeted EMS interventions and public health resources.