Exposure to herbicides mixtures in relation to type 2 diabetes mellitus among Chinese rural population: Results from different statistical models
Dandan Wei,Lulu Wang,Qingqing Xu,Juan Wang,Jiayu Shi,Cuicui Ma,Jintian Geng,Mengzhen Zhao,Xiaotian Liu,Jian Hou,Wenqian Huo,Linlin Li,Tao Jing,Chongjian Wang,Zhenxing Mao
DOI: https://doi.org/10.1016/j.ecoenv.2023.115109
Abstract:Background: Although it has been reported that herbicides exposure is related to adverse outcomes, available evidence on the associations of quantitatively measured herbicides with type 2 diabetes mellitus (T2DM) and prediabetes is still scant. Furthermore, the effects of herbicides mixtures on T2DM and prediabetes remain unclear among the Chinese rural population. Aims: To assess the associations of plasma herbicides with T2DM and prediabetes among the Chinese rural population. Methods: A total of 2626 participants were enrolled from the Henan Rural Cohort Study. Plasma herbicides were measured with gas chromatography coupled to triple quadrupole tandem mass spectrometry. Generalized linear regression analysis was employed to assess the associations of a single herbicide with T2DM, prediabetes, as well as indicators of glucose metabolism. In addition, the quantile g-computation and environmental risk score (ERS) structured by adaptive elastic net (AENET), and Bayesian kernel machine regression (BKMR) were used to estimate the effects of herbicides mixtures on T2DM and prediabetes. Results: After adjusting for covariates, positive associations of atrazine, ametryn, and oxadiazon with the increased odds of T2DM were obtained. As for prediabetes, each 1-fold increase in ln-transformed oxadiazon was related to 8.4% (95% confidence interval (CI): 1.033, 1.138) higher odds of prediabetes. In addition, several herbicides were significantly related to fasting plasma glucose, fasting insulin, and HOMA2-IR (false discovery rates adjusted P value < 0.05). Furthermore, the quantile g-computation analysis showed that one quartile increase in multiple herbicides was associated with T2DM (OR (odds ratio): 1.099, 95%CI: 1.043, 1.158), and oxadiazon was assigned the largest positive weight, followed by atrazine. In addition, the ERS calculated by the selected herbicides from AENET were found to be associated with T2DM and prediabetes, and the corresponding ORs and 95%CIs were 1.133 (1.108, 1.159) and 1.065 (1.016, 1.116), respectively. The BKMR analysis indicated a positive association between mixtures of herbicides exposure and the risk of T2DM. Conclusions: Exposure to mixtures of herbicides was associated with an increased risk of T2DM among Chinese rural population, indicating that the impact of herbicides exposure on diabetes should be paid attention to and measures should be taken to avoid herbicides mixtures exposure.