From total concentration to bioaccessible load: Reassessing soil heavy metal(loid)s risks through probabilistic...

Traditional health risk assessment (HRA) generally overlooks bioaccessibility, potentially leading to overestimated risk estimates. In this study, we systematically investigated the heavy metal(oid)s (HMs) characteristics of 126 soil samples collected from an industrial agglomeration area, based on which, in vitro assays were conducted to test bioaccessible HMs concentrations, and further integrated into both probabilistic HRA and the prediction of internal exposure levels. The study results of probabilistic HRA indicated that both scenarios showed negligible non-carcinogenic risks (with and without bioaccessibility), but the probabilities of non-carcinogenic risks could decrease by 82.83 % for adults and 78.93 % for children when considering bioaccessibility, along with a significant reduction on the overestimation of carcinogenic risks by approximately 70 %. In addition, internal exposure levels predicted by a physiologically based toxicokinetic (PBTK) model demonstrated that incorporating bioaccessibility led to a substantial reduction in urinary HMs concentrations, with decreases ranging from 60.41 % for Cd to 98.50 % for Cr. Moreover, the deviations from the verification value were consistently lower and more stable across all metals and age groups, indicating that incorporating bioaccessibility into risk assessments provides a more accurate and reliable estimation of HMs exposure to avoid overestimation of associated health risks, and to support more informed decision-making in soil environmental health management.