Superfund Report, Maria Hegstad
Published February 23, 2009
EPA's recently released draft assessment of human exposure to polybrominated diphenyl ethers (PBDEs) -- chemicals widely used in electronics and furniture as flame retardants -- is being criticized by industry for leaving out data that industry has already provided to the agency while environmentalists are criticizing the agency for underestimating exposures.
The groups' concerns could undermine agency efforts to assess the risks posed by the chemicals, which some agency officials have said should be quickly completed to address growing environmental contamination in the Columbia River area and elsewhere, as well as product safety concerns.
The PBDE exposure assessment, released in December, is a key part of EPA's ongoing plan to investigate the human health and environmental impacts of the persistent chemicals -- which have been linked to liver and thyroid toxicity, neurological problems and reproductive issues -- and possible alternatives. The exposure assessment is undergoing a review by external experts, with the document scheduled for completion within the year, says an EPA source.
The draft assessment acknowledges uncertainties stemming from modeling, noting EPA has "very limited data" on children and infants' body burdens of PBDEs as well as limited information on rates of exposure through household dust. Still, the "overall weight of evidence" supports the report's findings that 80 to 90 percent of "United States exposures occur in the indoor environment through contact with house dust," with the remaining exposures to PBDEs primarily through food. The report also concludes that "infant and childhood body burdens appear likely to be significantly higher than that of adults."
According to the draft assessment's findings, Americans have a higher body burden of PBDEs than Europeans. There is also "some suggestion" that women have higher body burdens than men. PBDEs bioaccumulate in lipids, the fat-soluble molecules in the body. American body burdens are measured in the range of 30 to 100 nanograms per lipid weight (ng/lwt) while measurements for other countries are below 10 ng/lwt, according to the report.
Agency sources said recently that it is important for EPA to quickly complete its exposure and risk assessment of the chemical to help clean up high levels of the chemical in the Columbia River basin as well as spur development of safer alternatives.
But comments provided to the agency raise concerns about the agency's data.
One PBDE manufacturer, Albermarle Corporation, is questioning why EPA did not include hazard and exposure information that manufacturers in 2002 provided EPA's pesticides and toxics office (OPPT) through a voluntary data generation program, known as the Voluntary Children's Chemical Exposure Program (VCCEP).
The data covered the three formulations of PBDEs that have been widely used in commercial products, decabromodiphenyl ether (deca), octabromodiphenyl ether (octa) and pentabromodiphenyl ether (penta).
"All the submissions were done at the request of EPA and yet here is EPA doing an assessment without this information," says a source with the company. "It's disappointing when you put all that time and effort into a voluntary program."
The source declined to comment on OPPT's ongoing efforts to compel more data about deca through VCCEP. The American Chemistry Council, an industry association, is negotiating an enforceable consent agreement with EPA, after the agency terminated ACC from VCCEP last summer for not providing the additional data and threatened to issue mandatory testing orders.
An EPA source says staff who developed the new draft exposure assessment -- in EPA's research and development office -- was not aware of the VCCEP submissions to OPPT.
Meanwhile, environmentalists are raising concerns that EPA may be underestimating exposure. The Environmental Working Group in its comments praises EPA's "careful work" on the draft assessment and calls the need for a thorough exposure assessment "acute." But the group raises concerns that the assessment "underestimates exposures for the fetus, infant and young child."
The draft exposure assessment does include a chapter that calculates exposure scenarios for several susceptible subpopulations, including infants and children. But it does not calculate specific exposure scenarios for fetuses, which EWG considers critical. "PBDE exposures begin in utero; exposure assessments for infancy and early childhood should consider the background levels in newborns from transfer of a mother's PBDEs across the placenta," writes EWG analyst Sonya Lunder. Comments are available on InsideEPA.com.
The critical question is whether Americans are exceeding the daily safety standards that EPA published last summer for PBDEs, says a source with EWG, adding that the exposure assessment "tiptoes up to the reference dose without coming to a conclusion." EPA set a reference dose of 0.007 milligrams per kilogram per day for oral ingestion of deca. Part of the problem, for comparison purposes, is that EPA performed a cumulative exposure assessment but calculated safety standards for each of formulation separately, the EWG source says.
The source notes, however, that it makes sense to perform an assessment of the exposures as a group, since they are usually found together in samples and people are exposed to a mixture of PBDEs. EWG's other main concern is that the assessment concentrates on the exposure levels of the average, or median exposed members of any study group. However, exposure scenarios in some homes are much higher, leading to much higher levels of PBDEs in those residents, the source says, adding that EPA's safety standards should protect those exposed at those highest levels as well as the median.
Albermarle, one of the few companies that still makes deca, also complains that EPA's draft risk assessment "fails to distinguish" deca from other PBDEs, and as a result, says the assessment is "of little utility for risk assessment." Albemarle also argues that EPA's draft assessment does not follow the agency's data quality guidelines.
"I don't see how or why you'd need a cumulative risk assessment when you've got actual toxicity data on each of the three chemicals," the industry source says. "Clearly the toxicity of each of the three products is different -- unless you took the sum of the exposure assessment and factored in the differences in the toxicity, I don't see how you could use it in a risk assessment."
EPA, however, "is concerned about all PBDEs," says the agency source. Deca is the only formulation still produced in the U.S., after manufacturers agreed to halt production of octa and penta in 2004. However, multiple PBDE congeners continue to be found in tests of humans and the environment. The complexity of the mixtures used in the commercial formulations is part of the difficulty of assessing the chemicals; there are some 209 different PBDE congeners. Each formulation is made up of a mixture of congeners, making it challenging to determine what formulation contributed to the congeners detected from a specific exposure incident.
Albemarle raises this in its comments, questioning some of the studies EPA uses in its assessment and noting that many labs are not experienced in testing for specific congeners.
