PO.PS01.12 · 人群科学
Mapping risk: spatial analysis of 2,4-D herbicide emissions and colorectal cancer
作者与单位
摘要 Abstract
Colorectal cancer (CRC) incidence remains disproportionately high in rural agricultural regions, including Arkansas, where environmental exposures may contribute to persistent cancer disparities. Yet, ambient community-level herbicide exposures have been minimally studied as potential CRC risk factors, representing an important knowledge gap. To identify candidate environmental contributors, we integrated county-level air pollutant emission estimates from the EPA AirToxScreen dataset with age-adjusted CRC incidence from the National Cancer Institute State Cancer Profiles. Screening 133 pollutants in Arkansas identified 2,4-dichlorophenoxyacetic acid (2,4-D), one of the most widely used herbicides in U.S. agriculture, as the exposure most strongly correlated with CRC incidence (Spearman ρ = 0.338, p = 0.003). Linear regression indicated that each one-log-unit increase in 2,4-D emissions corresponded to 6.28 additional CRC cases per 100,000 population (p = 0.001), with similar associations in rural and urban strata. Extension to 2,555 U.S. counties demonstrated a consistent association nationally (beta = 5.94, p < 0.001). Spatial error regression improved model fit and confirmed the association after accounting for geographic autocorrelation (beta = 2.55, p = 0.001). Adjustment for county-level poverty modestly strengthened the association (beta = 2.81, p < 0.001), suggesting that both socioeconomic conditions and herbicide exposure may jointly contribute to CRC burden in agricultural regions. While ecological and not causal, this study is among the first to evaluate ambient, population-level 2,4-D exposure rather than occupational exposure, and demonstrates a spatial epidemiologic approach for identifying environmental carcinogen candidates relevant to precision public health and rural cancer equity. These findings support the need for follow-up studies integrating individual-level exposure biomarkers, mechanistic data, and targeted CRC screening strategies in high-exposure regions.
利益披露 Disclosure
D. Baxter, None..
D. Dixon, None..
A. Morris, None..
T. Zheng, None..
Y. Zhu, None.