A Multi-Route Exposure Assessment of Chemically Contaminated Drinking Water

This report provides an example of how a single source of contamination could potentially contribute to all routes of exposure. A modeling approach was used to estimate multiple exposure routes in an attempt to assess the health significance of gasoline-contaminated drinking water supplies. This model consisted of a two-compartment, indoor air quality equation that calculates the contribution made by ambient and indoor air contaminated by a pollutant volatilized from drinking water to that pollutant's inhalation burden. In addition, the model uses the traditional equations for assessing a pollutant's oral and dermal burdens. Benzene, toluene and xylene were used as surrogates for gasoline contamination to determine the contribution of contaminated water to adult and child body burdens from indoor air, oral (drinking water and food) and dermal exposure routes. The contribution thus calculated for each chemical was compared to the EPA's Office of Drinking Water Health Advisories.

In terms of acute exposure, the use of chemically contaminated water for showering purposes may generate vapor in the confined area of the bathroom at levels sufficient to cause or contribute to mucous tissue irritation, as commonly reported in affected homes. High temperatures and humidity may also contribute to these effects, especially in the bathroom. In terms of chronic exposure, the use of chemically contaminated water at EPA-recommended guideline amounts in an affected home may result in inhalation, oral and dermal exposures leading to cumulative doses exceeding adult and child total daily body burdens based on EPA's Health Advisories. Thus, this model indicates that the traditional standard/guidelines derivation processes should be reevaluated to consider the pollutant contribution from multiple routes of exposure.

The New Jersey Departments of Health and Environmental Protection conducted a study in which concentrations of several pollutants including benzene in the breathing zone were measured during a 15-minute shower in homes with contaminated water. The findings suggest that the air quality model used in the present study may satisfactorily predict the airborne concentrations of pollutants in, at least, the bathroom after showering with contaminated water (Pearson rank correlation coefficient of 0.773 with p = 0.0012 for n =14).

The findings of the present study support the use of an adjustment factor for all exposure durations to account for exposures to other sources of the contaminant, i.e., urban, occupational, and food. A value of 20% seems appropriate based on the study's findings. However, it is more likely that this value may be too high considering that occupational and food exposures for most of the contaminants evaluated in the study were not accounted for.

The generated values in this study are theoretical and, therefore, should be cautiously interpreted. Limitations of the model include inappropriate use of steady-state dermal absorption rates for the pollutants considered in this study, poor air stripping data, and background indoor air quality data, among others. However, given the limitations of the model, the inherent conservative approach employed in calculating body doses and drinking water guidelines, and the scant reports of ill health effects in households with gasoline-contaminated drinking water, the findings presented in this report should not be discounted and should suggest research efforts to confirm or refute the findings presented in this report. Primary research efforts should concentrate on evaluating dermal uptake of pollutants under initial rate conditions in varying temperatures of soapy water.

The Maine Departments of Human Services and Environmental Protection are gathering case history information on several selected homes with contaminated drinking water that have not yet installed charcoal filters. The information gathered will include the results of air monitoring in the bathroom with the shower flowing, an assessment of the pattern of contaminated water use and health questionnaire, among others. The brief health questionnaire will concentrate primarily on acute effects to assess irritant potential of gasoline vapors in bathrooms. Similar information will be gathered from control homes for comparison.