Construction of Models To Predict the Effectiveness of E-Waste Control through Capture of Volatile Organic Compounds and Metals/Metalloids Exposure Fingerprints: A Six-Year Longitudinal Study

Abstract

The significant health implications of e-waste toxicants have triggered the global tightening of regulation on informal e-waste recycling sites (ER) but with disparate governance that requires effective monitoring. Taking advantage of the opportunity to implement e-waste control in the Guiyu ER since 2015, we investigated the temporal variations in levels of oxidative DNA damage, 25 volatile organic compound metabolites (VOCs), and 16 metals/metalloids (MeTs) in urine in 918 children between 2016 and 2021 to demonstrate the effectiveness of e-waste control in reducing population exposure risks. The hazard quotients of most MeTs and levels of 8-hydroxy-2'-deoxyguanosine in children decreased significantly during this time, indicating that e-waste control effectively reduces the noncarcinogenic risks of MeT exposure and levels of oxidative DNA damage. Using mVOC-derived indexes as a feature, a bagging-support vector machine algorithm-based machine learning model was constructed to predict the extent of e-waste pollution (EWP). The model exhibited excellent performance with accuracies >97.0% in differentiating between slight and severe EWP. Five simple functions established using mVOC-derived indexes also had high accuracy in predicting the presence of EWP. These models and functions provide a novel human exposure monitoring-based approach for assessing e-waste governance or the presence of EWP in other ERs.