Human health risk assessment of pharmaceuticals in water: an uncertainty analysis for meprobamate, carbamazepine, and phenytoin.

This study presents a step-wise development of a quantitative pharmaceutical risk assessment (QPhRA, hereafter) framework, including Monte Carlo uncertainty analysis for meprobamate, carbamazepine, and phenytoin during (1) accidental exposures of stream water and fish consumption and (2) direct ingestion of finished drinking water for children and adults. Average hazard quotients of these pharmaceuticals (i.e., the ratio of values of chronic daily intake to acceptable daily intake) were found to lie between 1x10(-10) and 3x10(-5) and 99 th percentile values of hazard quotients were found to be less than 1x10(-4) for both sub-populations, indicating no potential risks of adverse effects due to pharmaceuticals exposures. In addition, pharmaceutical concentrations were also observed to be lower than their respective calculated acceptable daily intake-equivalent drinking water levels, indicating no potential human health risks. To the authors' knowledge, for the first time in QPhRA studies, this study has attempted to characterize and quantify effects of factors, such as considerations for sensitive sub-populations using subpopulation-specific toxic endpoints and use of pharmaceutical concentrations in stream and finished drinking waters on risk estimates. Acceptable daily intake was observed to be the primary contributor (>93% variance contribution) in the overall uncertainties of estimates of hazard quotients, followed by fish consumptions and pharmaceutical concentrations in water. Further research efforts are required to standardize use of acceptable daily intake values to reduce large variability in estimation of hazard quotients.