Burdens of mercury in residents of Temirtau, Kazakhstan: II: Verification of methodologies for estimating human exposure to high levels of Hg pollution in the environment Original Research Article

A considerable amount of work has been conducted developing exposure estimate models for quantitative evaluation of mercury (Hg) intake and human health risks, but few have assessed the applicability and the validity for evaluating the risks posed by Hg in the environment and have achieved very mixed results. The present study focused on verifying estimated daily Hg intake using exposure equations with either the deterministic or probabilistic (the Monte Carlo) approaches. The simulated daily Hg intake doses were compared with those established from measured Hg concentrations in the hair of 289 participants. The results showed that the single-value deterministic method for simulating Hg exposure levels overestimated the level of risk by a factor of 1.5 when compared with the highest concentration of Hg observed in the hair of the study population. Contrarily, the average daily Hg intake doses simulated using the probabilistic simulation were similar in distribution to the biomarker data. When the reference dose (RfD) of 0.1 μg/kg body weight/day was adopted as the acceptable dose for daily intake rate, there were approximately 19% estimated to have potential Hg exposure risks based on the Monte Carlo simulation. This percentage was favourably similar to the 17% determined from Hg concentrations in the hair samples. The difference between the probabilistic simulation and the data derived from hair Hg levels was considered mainly due to the uncertainties in unconfirmed questionnaire-based survey data, small sampling sizes and the surrogates used in the exposure models. The findings implied that the existing exposure models together with the probabilistic approach were appropriate for research of human exposure to Hg.