Modelled and measured arsenic exposure around a power station in Slovakia

In 1999–2000 an epidemiological study was carried out to investigate arsenic exposure and risk of non-melanoma skin cancer in the population around a power station in Slovakia. As part of the exposure assessment for that study, markers of physiological exposure to arsenic (urinary arsenic and nail arsenic) and measures of environmental arsenic were taken. Residential history and arsenic emissions were combined in a measure used to classify the study population’s exposure. Here, we take a model designed to estimate the daily arsenic absorption and urinary arsenic excretion developed by the US EPA and compare its output to the values for measured urinary arsenic, measured environmental arsenic and modelled food arsenic used in the previous study. We found that the US EPA model underestimated arsenic exposure in this population. Potential misclassification compared to the exposure used previously was also quantified, with poor agreement between a measured biomarker of current arsenic exposure and modelled estimates of arsenic uptake and excretion.

► This paper assesses arsenic excretion and absorption as modelled by a USEPA-derived exposure model.
► Modelled urinary arsenic was compared to measured urinary arsenic.
► Median modelled arsenic was lower than median measured arsenic excretion by a factor of 3.5.
► Modelled arsenic absorption and excretion were poor predictors of measured exposure.