Science Inventory

A Summary of Publications on the Development of Mode-of-Action Information and Statistical Tools for Evaluating Health Outcomes from Drinking Water Disinfection By-Product (DBP) Exposures


EPA is announcing the availability of A Summary of Publications on the Development of Mode-of-Action Information and Statistical Tools for Evaluating Health Outcomes from Drinking Water Disinfection By-Product (DBP) Exposures. This set of publications develops mode-of-action data and statistical methods for expanding beyond single chemical human health risk assessments of exposure to disinfection by-products (DBPs) in drinking water to the evaluation of complex mixtures of DBPs. Five publications are publically available. The publications are related to the EPA’s “Four Lab Study” experiments, which have been conducted to provide toxicology and chemistry data on environmentally-relevant mixtures of DBPs and to improve estimation of the potential human health risks associated with exposure to DBP mixtures.


U.S. EPA. A Summary of Publications on the Development of Mode-of-Action Information and Statistical Tools for Evaluating Health Outcomes from Drinking Water Disinfection By-Product (DBP) Exposures. U.S. Environmental Protection Agency, Washington, DC, 2011.


This project summary and its related publications provide information and methods for the human health risk assessment of complex DBP mixtures, including toxicological mode-of-action determinations for reproductive and developmental effects for DBPs, development and analysis of analytical chemistry and toxicology data on DBPs, and risk assessment methodology for complex chemical mixtures in drinking water.


Chemical contaminants are formed as a consequence of chemical disinfection of public drinking waters. Chemical disinfectants, which are used to kill harmful microorganisms, react with natural organic matter (NOM), bromide, iodide, and other compounds, forming complex mixtures of potentially toxic disinfection byproducts (DBPs). The types and concentrations of DBPs formed during disinfection and the relative proportions of the components vary depending on factors such as source water conditions (e.g., types of NOM present), disinfectant type (e.g., chlorine, ozone, chloramine), and treatment conditions (e.g., pH and temperature). To date, over 500 DBPs have been detected in treated waters. However, typically more than 50% of the organic halide mass produced by chlorination disinfection consists of unidentified chemicals, which are not measured by routine analyses of DBPs.

The protocols and methods typically used to evaluate chemical mixtures are best applied to simple defined mixtures consisting of relatively few chemicals. These approaches rely on assumptions (e.g., common mode of action, independent toxic action) regarding the type of joint toxic action (e.g., dose-additivity, synergism) that might be observed. Such methods, used for site assessments or toxicological studies, are often not sufficient to estimate health risk for complex drinking water DBP mixtures. Actual drinking water exposures involve multiple chemicals, many of which may be unknown. In addition, the toxicological effects of drinking water mixtures, including knowledge of the mode-of-action, is unknown in most cases. Because the potential for joint toxic action is thought to depend on dose levels and toxicological action of the mixture components, it is difficult to assess the likelihood of observing either additivity (e.g., dose addition or response addition) or some interaction effect (e.g., synergism or antagonism) among the mixture components. Thus, information and methods need to be developed that advance the field of chemical mixture risk assessment, improve the conduct of simple DBP mixture assessments, and enhance the ability to evaluate the potential toxicity of whole DBP mixtures.

Epidemiologic studies have reported some associations between DBP exposures and health effects in humans, including bladder cancer and reproductive and developmental effects. Although credible, these epidemiologic studies do not comprise a conclusive body of evidence for determining whether DBP exposures are risk factors for reproductive or developmental toxicity; thus, additional DBP toxicology studies represent a research need. To address this issue, EPA undertook the Four Lab Study project to evaluate the chemistry and toxicology of whole mixture DBP concentrates. These concentrates were produced in a water matrix, thus, providing an opportunity for animal exposures to occur through the same oral ingestion pathway and medium as human exposures. In this project, the reproductive and developmental consequences of multi-generation oral exposure to DBP water concentrates were examined, thereby evaluating the chemistry and toxicology of both the known and unidentified DBPs through experimentation with the whole DBP mixture.

This project summary and its related publications provide information and methods for the human health risk assessment of complex DBP mixtures. Colman et al. (2011) identifies twenty-four DBPs as developmental toxicants and four adverse developmental outcomes associated with human DBP exposures: spontaneous abortion, cardiovascular defects, neural tube defects, and low birth weight. This publication examines data relevant to the mode-of-action for DBP developmental toxicity that can be used to group DBPs for mixture risk assessments. Pressman et al. (2010) describe a new procedure that was used to produce chlorinated drinking water concentrates for the Four Lab Study animal toxicology experiments. This publication identifies more than 100 DBPs in the concentrates, quantifies concentrations for 75 of the priority and regulated DBPs and measures the levels of total organic halide, which is a summary measure for the whole DBP mixture including the unidentified fraction. This extensive chemical characterization of the DBP mixture is important to the analysis of the toxicological experimental outcomes. The results are being used to evaluate the chemical stability of the mixture over the duration of the animal study, attribute toxic effects to specific known DBPs or to the unidentified DBP fraction, and to develop general methods for assessing the toxicity of complex mixtures. The statistical methods in the Rice et al. (2008) paper represent chemical mixture risk assessment approaches that were developed for application to the Four Lab Study chemistry and toxicology data. Rice et al. (2008) present a toxicologically-based risk assessment strategy for identifying the individual components or fractions of a complex mixture that are associated with its toxicity. Finally, the publications by Hertzberg et al. (2008) and Simmons and Teuschler (2010) present discussions of DBP chemistry and toxicity data, as well as heath risk assessment methodologies, utilizing information that was developed during the planning and conduct of the Four Lab Study experiments.

This collection of five publications on complex mixtures of drinking water DBPs was developed collaboratively among scientists within U.S. EPA’s Office of Research and Development and under contracts with Battelle and Syracuse Research Corporation. The citations for these publications are as follows:

Development of risk assessment methods is a major component of the work done by EPA's National Center for Environmental Assessment, which provides guidance and risk assessments aimed at protecting human health and the environment. The development of environmental toxicology data, analytical chemistry data and environmental engineering technologies are critical elements of evaluating DBP mixtures, for which project leadership largely resides in EPA’s National Health and Environmental Effects Laboratory, National Exposure Research Laboratory and National Risk Management Research Laboratory, respectively. This research on complex drinking water DBP mixtures supports the needs of the Office of Water to evaluate and provide safe drinking water under the Safe Drinking Water Act Amendments of 1996, which specifically call for the EPA to “develop new approaches to the study of complex mixtures, such as mixtures found in drinking water….” As new regulations are proposed and promulgated, changes are generally made in water treatment processes and disinfection scenarios. Such innovations can result in fundamental changes in the composition of DBP mixtures to which the public may be exposed and for which human health evaluations will be needed. Toxicology and chemistry data and risk assessment methods such as those developed within this project will be useful for evaluating potential human health risks from exposure to current and future DBP mixtures.

Related Links:

Record Details:

Product Published Date: 05/27/2003
Record Last Revised: 05/17/2016
OMB Category: Other
Record ID: 56913