Tom Neltner, J.D., is Chemicals Policy Director and Maricel Maffini, Ph.D., Consultant
The Environmental Protection Agency (EPA) will soon propose a drinking water standard for perchlorate. The decision – due by the end of May per a consent decree with the Natural Resources Defense Council (NRDC)— will end a nearly decade-long process to regulate the chemical that has been shown to harm children’s brain development.
In making its decision, EPA must propose a Maximum Contaminant Level Goal (MCLG) “at the level at which no known or anticipated adverse effects on the health of persons occur and which allows an adequate margin of safety.”[1] It must also set a Maximum Contaminant Level (MCL) as close to the MCLG as feasible using the best available treatment technology and taking cost into consideration.
To guide that decision, EPA’s scientists developed a sophisticated model that considers the impact of perchlorate on the development of the fetal brain in the first trimester when the fetus is particularly vulnerable to the chemical’s disruption of the proper function of the maternal thyroid gland. As discussed more below, the model was embraced by an expert panel of independent scientists through a transparent, public process that included public comments and public meetings.
In April, a consulting firm published a study critiquing EPA’s model. The authors acknowledged the model as a valuable research tool but did not think it is sufficient to use in regulatory decision-making due to uncertainties. Therefore, the authors concluded that EPA should discard the peer-reviewed model and rely on a 14-year old calculation of a “safe dose” that does not consider the latest scientific evidence and has even greater uncertainties. They didn’t offer other options such as using uncertainty factors to address their concerns about the model’s estimated values.
Given the importance of the issue and the risk to children’s brain development, we want to explain EPA’s model, the process the agency used to develop it, and the study raising doubt about the model.