A framework is developed and applied for semi-quantitative estimation of cumulative risk from complex mixtures of compounds in water supplies. The framework places these risks onto the unifying metric of Disability Adjusted Life Years (DALYs), and harmonizes cancer and non-cancer, morbidity and mortality, effects. The framework can be used to: 1) calculate a measure of cumulative risk for a given supply, and compare this measure across supplies or across the same supply with candidate treatments applied; 2) identify those compounds contributing most significantly to cumulative risk, so risk management measures can be applied most effectively; and 3) quantify the influence of different regulatory limits, for specific compounds, on the cumulative risk from drinking water. Results of application to a hypothetical water supply in which all compounds are at their existing Maximum Contaminant Level (MCL) show the cumulative risk for even a complex mixture may be dominated by a few compounds. In this application, that risk was dominated by as few as 10% of the compounds. The analysis also shows that establishing MCLs based on compounds for which there is an oral slope factor, but where no cancer-based limit has yet been established, probably will have little influence on the relative cumulative risk (as measured by Total Weighted DALY) of different water supplies. This arises primarily because the non-cancer-based MCL is usually more restrictive than the one based on cancer for target probabilities of cancer equal to 1E–4 or 1E–5

No abstract

This paper applies cumulative and aggregate risk methods and weight of evidence determination to re-analysis of epidemiological and clinical studies of exposure to perchlorates. The implications of cumulative and aggregate risk are considered for 28 epidemiological studies on IUI, serum thyroid hormone levels and clinical indicators. Consideration is given to simultaneous exposures to perchlorates, nitrates, thiocyanates and organohalogens in the study populations. The elevation of effects by perchlorates alone is found only in the studies that use urinary perchlorate as the metric of exposure. These studies are beset by a problem with use of urinary perchlorate concentration in that there is large inter-subject variability in the relationship between intake and urinary concentration due to differences in metabolism and disposition of the compounds following ingestion. As a result, an individual placed into the "high urinary concentration" group may be there due to high values of exposure, to long biological clearance half-lives, or due to high transfer fractions from the serum into the urine. The influence could be removed by correcting urinary levels by measured clearance half-times for individuals in a study, but that has not been done in the case of the studies examined here. It is of interest therefore that the studies that use direct measures of intake of perchlorates rather than urine concentration fail to display the hormone effects. The current study uses a "weight of evidence" approach for perchlorates, employing all 28 studies. The result is a slope of the exposureresponse curve (percentage change in hormone effect per unit exposure) of 0.3% per µg/kg-day, with 95% confidence interval of (−0.05%, 1%). This confidence interval for the slope encompasses 0, indicating no statistically significant slope when all data are combined in a weight of evidence determination. This is consistent with the conclusions of the USEPA and EFSA that the epidemiological studies do not provide compelling evidence for a causal association between exposures to perchlorates and either hormone effects or clinically adverse effects. The conclusions are 1) that current epidemiological results do not provide evidence of effects induced by perchlorates apart from the IUI effects, 2) that the same results provide evidence that the IUI effects induced at environmental levels of exposure are associated with down-stream adverse effects and 3) that effective risk management requires the cumulative and aggregate risk framework adopted here, suggesting a need for risk assessors to return to the original studies and provide separate estimates of exposure-response relationships for all four compounds or at the least to control for exposures to nitrates, thiocyanates and organohalogens.