AUTHOR=Ng Carmen D. , Zhang Pluto , Kowal Stacey TITLE=Validating the Social Vulnerability Index for alternative geographies in the United States to explore trends in social determinants of health over time and geographic location JOURNAL=Frontiers in Public Health VOLUME=Volume 13 - 2025 YEAR=2025 URL=https://www.frontiersin.org/journals/public-health/articles/10.3389/fpubh.2025.1547946 DOI=10.3389/fpubh.2025.1547946 ISSN=2296-2565 ABSTRACT=ObjectiveTo create county-, 5-digit ZIP code (ZIP-5)–, and 3-digit ZIP code (ZIP-3)–level datasets of the Social Vulnerability Index (SVI) and its components for 2016–2022 to validate the methodology beyond county level, explore trends in SVI over time and space, and demonstrate its usage in an enrichment exercise with health plan claims.Materials and methodsThe SVI consolidates 16 structural, economic, and demographic variables from the American Community Survey (ACS) into 4 themes: socioeconomic status, household characteristics, racial and ethnic minority status, and housing type and transportation. ACS estimates of the 16 variables for 2016–2022 were extracted for counties and ZIP code tabulation areas and for ZIP code geographies, crosswalked to ZIP-5, and aggregated to ZIP-3. Areas received a percentile ranking (range, 0–1) for SVI and each variable and composite theme, with higher values indicating greater social vulnerability.ResultsSVI estimates were produced for up to 3,143 counties, 32,243 ZIP-5s, and 886 ZIP-3s. SDoH trends across the US were largely consistent from 2016 to 2022 despite slight local changes over time. SVI varied across regions, with generally higher vulnerability in the South and lower vulnerability in the North and Northeast. When linked with health plan claims data, higher SVI (i.e., higher vulnerability) was associated with greater comorbidity burden.ConclusionSVI can be estimated at the ZIP-3 and ZIP-5 levels to provide area-level context, allowing for more routine integration of socioeconomic and health equity–related concepts into health claims and other datasets.