Variability and uncertainty in chemical exposures for regulatory risk assessments

Occupational exposure assessment is an important part of regulatory risk assessment for industrial chemicals. The current procedures used to assess exposure for existing substances either use measurements of occupational exposure or, if these are not available, modelled data, using the EASE software tool. There is some consideration of the variability of exposure measurements, but the procedures do not explicitly include any evaluation of the uncertainties involved. The present study is based on a new paradigm where both modelled and measured exposures are combined within a Bayesian statistical framework. As part of this approach we have incorporated estimates of both exposure variability and uncertainty. We have explored this new approach using a number of occupational exposure scenarios involving inhalation exposure, both in comparison with predictions made by EASE and values produced as part of the European chemicals regulatory process. The results from this work were promising and we have concluded that this new approach could form the basis for an improved exposure estimation methodology, particularly when more detailed consideration of exposure is needed to conclude the regulatory risk assessment. Further development work is necessary to achieve an improved practical exposure assessment methodology.