Association of Regional Agricultural Smoke Exposure With Sociodemographic Factors in Rural and Urban Communities.
K D Slater, Bonnie N Young, Bonne Ford, Susana Adamo, Emily Fischer, Emily Gargulinski, Giovanna L Henery, Jeffrey R Pierce, Zoey Rosen, Olivia Sablan, Amber Soja, Lisa A Wiese, Christine L Williams, Sheryl Magzamen
Abstract
Open AccessAmbient air pollution remains a leading environmental risk factor for morbidity and mortality in the U.S, though most research is conducted in urban areas. Our study assessed how sociodemographic factors indicative of social vulnerability were associated with smoke from agricultural burns in Florida. We assessed census-level sociodemographic variables among four counties adjacent to the Everglades Agricultural Area (n = 409 census tracts, 2016-2020). Smoke day counts from local agricultural fires were based on satellite plumes identified from the National Oceanic and Atmospheric Administration Hazard Mapping System. Primary analysis fit a negative binomial model with bidirectional stepwise regression, followed by an adjusted geospatial model with a Queen-continuity adjacency matrix. Sensitivity analysis focused on rural-only census tracts. Rural areas had higher concentrations of people of color and poverty compared to coastal urban areas. Median (Q1, Q3) smoke days by census tract was 36 (31, 45), with the highest concentrations in rural central and western regions. Primary model results skewed toward mostly urban tracts, where an interquartile ranges (IQR) increase in median household income was associated with a 12% decrease (95% confidence interval (CI) -14.5%, -5.2%) in smoke days. Among rural-only census tracts, an IQR increase in percentage of residents living 200% below the poverty line and non-English speaking residents were associated with 23% (95% CI: 1.2%, 37.7%) and 120% (95% CI: 20.5%, 176.5%) increases in smoke days, respectively. Sociodemographic factors associated with health and environmental vulnerability were context dependent. Within rural regions, poverty, race and ethnicity played more important roles in exposure risk, whereas wealth mitigated risk among urban areas.