Abstract
In Japan, health risk assessment of chemicals in ambient air should be based on information obtained from epidemiological studies for setting the Guideline Values and Environmental Quality Standards. However, toxicity data derived from epidemiological studies often are limited to the cases of occupational exposure.
Because protective measures against excess exposure have improved during recent years, the cases of overt toxicity due to high doses are expected to be quite rare. Consequently, future health risk assessment will rely on data of animal studies rather than those of epidemiological studies in humans.
To this end, we need to confirm the appropriateness of extrapolating to humans from animals by using the benchmark dose approach and uncertainty factors.
First, we assessed the cancer risk of vinyl chloride monomer and 1,3-butadiene on the basis of inhalation studies in animals, and compared these assessment values estimated from animal studies with the Guideline Values for these substances in ambient air in Japan, which previously were established on the basis of epidemiological studies in humans. The Guideline Values of these substances were within the ranges of the risk assessment values caculated from animal studies.
However, in the case of assessment of non-carcinogenic endpoint (e.g. adverse effects of trichloroethylene and tetrachloroethylene) our analysis showed that the assessment values based on animal studies might differ from those based on epidemiological studies, if large uncertainty factors are adopted.
