Mixture Effects of Commonly Applied Herbicides on County Level Obesity Rates in the United States: An Exploratory Ecologic Study (2013-2018).
Sarah Otaru, Laura E Jones, David O Carpenter
Abstract
Open AccessMetabolic disorders such as obesity have increased globally in recent decades and are a major public health concern. Previous research suggests that herbicide exposures may contribute to metabolic dysfunction, but few studies have examined mixture effects of multiple herbicides on obesity at a population level. Using county-level data from 2013 to 2018, we examined the associations between obesity rates and the application of 13 commonly applied herbicides in the U.S. We first conducted adjusted single-pollutant mixed-effects models and then used quantile-based g-computation mixture modeling to assess combined herbicide mixture effects on county-level obesity rates. Models were adjusted for demographic and socioeconomic covariates and accounted for geographic clustering. Significant positive associations were identified between county-level obesity rates and applications of glyphosate, 2,4-D, atrazine, acetochlor, metolachlor, and several other herbicides in adjusted single-pollutant models. Glyphosate showed one of the strongest individual associations (β = 0.29 per standard deviation increase, 95% CI: 0.21-0.36). Increases in herbicide mixture were significantly associated with higher obesity rates (Psi = 0.71 per quantile exposure mixture, 95% CI: 0.65-0.76) from mixture modeling. Inclusion of significant interaction terms did not appreciably increase the mixture effect. Glyphosate, 2,4-D, metolachlor, dimethenamid-P, and glufosinate contributed most strongly to the weighted mixture effect. Mixture effects varied by rurality, with stronger associations observed in rural counties, particularly in micropolitan regions. Our findings highlight the importance of considering cumulative herbicide mixture exposures rather than individual chemicals in isolation. The observed rural-urban disparities emphasize the need for targeted public health interventions and policy actions in rural communities, which may be particularly vulnerable to the adverse metabolic impacts of herbicide mixtures.