Combining Occurrence and Toxicity Information to Identify Priorities for Drinking‐Water Mixture Research

Abstract

Characterizing all possible chemical mixtures in drinking water is a potentially overwhelming project, and the task of assessing each mixture's net toxicity even more daunting. We propose that analyzing occurrence information on mixtures in drinking water may help to narrow the priorities and inform the approaches taken by researchers in mixture toxicology. To illustrate the utility of environmental data for refining the mixtures problem, we use a recent compilation of national ground-water-quality data to examine proposed U.S. Environmental Protection Agency (EPA) and Agency for Toxic Substances and Disease Registry (ATSDR) models of noncancer mixture toxicity. We use data on the occurrence of binary and ternary mixtures of arsenic, cadmium, and manganese to parameterize an additive model and compute hazard index scores for each drinking-water source in the data set. We also use partially parameterized interaction models to perform a bounding analysis estimating the interaction potential of several binary and ternary mixtures for which the toxicological literature is limited. From these results, we estimate a relative value of additional toxicological information for each mixture. For example, we find that according to the U.S. EPA's interaction model, the levels of arsenic and cadmium found in U.S. drinking water are unlikely to have synergistic cardiovascular effects, but the same mixture's potential for synergistic neurological effects merits further study. Similar analysis could in future be used to prioritize toxicological studies based on their potential to reduce scientific and regulatory uncertainty. Environmental data may also provide a means to explore the implications of alternative risk models for the toxicity and interaction of complex mixtures.

The quality of our Nation’s waters: Quality of water from public-supply wells in the United States, 1993–2007: Overview of major findings

“…Identifying and characterizing all possible contaminant mixtures in drinking water and assessing their toxicity is not practical (Squillace and others, 2002;Ryker and Small, 2008). Many contaminant combinations are possible, and which combinations are potential concerns because of their combined toxicity often is unknown.…”

Spatial Variation in Ambient Air Toxics Concentrations and Health Risks between Industrial-Influenced, Urban, and Rural Sites

“…47 Although HQ values are not directly comparable between pollutants because RFCs do not indicate the same level of toxicity, the differences in toxicity associated with HQ values add a level of uncertainty to the interactive analysis but should not greatly affect the prioritizing of interactions for future study.…”

“…f pjk is the hazard of the kth pollutant relative to the total additive hazard of all of the chemicals interacting with pollutant j. M jk is the interaction magnitude, for which EPA has suggested a default of 5, but M jk has been shown to be up to 10 for low-dose mixtures. 47 B jk is the weight of evidence that chemical k will affect chemical j's toxicity. B jk ranges from Ϫ1 to 0 for antagonistic effects and from 0 to 1 to synergistic effects.…”

Trace elements and radon in groundwater across the United States, 1992-2003

“…Characterizing all possible mixtures in a water sample is a difficult task; assessing the toxicity of any of these possible mixtures is even more challenging (Ryker and Small, 2008). Thus, for this analysis, which focused on trace elements, data on only selected trace elements were used in characterizing sample mixtures.…”