By Juliet Eilperin
Washington Post Staff Writer
Monday, September 22, 2008; A09
The Environmental Protection Agency, under pressure from the White House and the Pentagon, is poised to rule as early as today that it will not set a drinking-water safety standard for perchlorate, a component of rocket fuel that has been linked to thyroid problems in pregnant women, newborns and young children across the nation.
According to a near-final document obtained by The Washington Post, the EPA's "preliminary regulatory determination" -- which was extensively edited by White House officials -- marks the final step in a six-year-old battle between career EPA scientists who advocate regulating the chemical and White House and Pentagon officials who oppose it. The document estimates that up to 16.6 million Americans are exposed to perchlorate at a level many scientists consider unsafe; independent researchers, using federal and state data, put the number at 20 million to 40 million.
Some perchlorate occurs naturally, but most perchlorate contamination in U.S. drinking water stems from improper disposal by rocket test sites, military bases and chemical plants. A nationwide cleanup could cost hundreds of millions, if not billions, of dollars, and several defense contractors have threatened to sue the Defense Department to help pay for it if one is required.
The new EPA proposal -- which assumes the maximum allowable perchlorate contamination level is 15 times what the EPA had suggested in 2002 -- was heavily edited by officials of the White House Office of Management and Budget. They eliminated key passages and asked the EPA to use a new computer modeling approach to calculate the chemical's risks.
Under a process the OMB initiated in 2004, federal agencies with an interest in chemicals such as perchlorate, such as the Defense Department, have opportunities to weigh in on the EPA's regulatory decisions before they become final: The Government Accountability Office reported this spring that the Pentagon had pressured the EPA for several years not to regulate perchlorate.
"They have distorted the science to such an extent that they can justify not regulating" the chemical, said Robert Zoeller, a University of Massachusetts professor who specializes in thyroid hormone and brain development and has a copy of the EPA proposal. "Infants and children will continue to be damaged, and that damage is significant."
Zoeller said scientific studies have shown that a small reduction in thyroid function in infants can translate into a loss of IQ and an increase in behavioral and perception problems. "It's absolutely irreversible," he said. "Even small changes in thyroid functions early on have impacts on functioning through high school and even into people's 20s."
A reference to those studies in the EPA's proposal was deleted by OMB officials.
The document states that establishing a drinking-water standard for perchlorate "would not present a 'meaningful opportunity for health risk reduction for persons served by public water systems,' " but it also reveals that many Americans will be exposed to the compound at levels higher than recommended if nothing is done to remove it. Perchlorate impedes the functioning of the thyroid gland, which produces hormones that foster mental and physical development and control metabolism. The notice indicates that the agency plans to finalize its decision by Dec. 1.
The EPA's assistant administrator for water, Benjamin H. Grumbles, said in a statement yesterday, "Science, not the politics of fear in an election year, will drive our final decision."
"Until then, final numbers and strategies are mere speculation," Grumbles added. "We know perchlorate in drinking water presents some degree of risk and we're committed to working with states and scientists to ensure public health is protected and meaningful opportunities for reducing risk are fully considered."
The Senate Environment and Public Works Committee, chaired by Barbara Boxer (D-Calif.), has endorsed legislation requiring the EPA to set a standard for the chemical and to monitor perchlorate in tap water. Yesterday, Boxer lambasted the agency for refusing to establish a federal exposure standard.
"Perchlorate has been a serious, persistent and widespread problem which threatens the health of our families, especially our children," Boxer said. "For the Bush EPA to walk away from this problem and shrug off this danger is, in my view, unforgivable and immoral."
Federal, state and independent scientists have differed over the years as to what represents an acceptable dose of perchlorate in drinking water, though all have set the bar higher than the non-mandatory level in the EPA's new proposal. In January 2002, the EPA issued a draft risk assessment finding that 1 part per billion should be considered safe; in March 2006, the Massachusetts Department of Environmental Protection set a maximum contaminant level of 2 ppb; last year, California adopted a standard of 6 ppb.
A National Academy of Sciences panel prepared a risk analysis in 2005 that, according to the EPA's traditional models, would produce a protective standard of 1 to 6 ppb. The academy's study came under attack because two of the committee's members had financial ties to defense contractors that face legal liability because of perchlorate disposal.
In the EPA's proposed rule, the administration assumes that perchlorate contamination of 15 ppb is safe. But its regulatory document states that "between 16,000 and 28,000 pregnant women" and 900,000 to 2 million Americans could be exposed to higher levels.
The EPA document also finds that bottle-fed infants would be exposed to more than five times the level the National Academy of Sciences deemed safe -- 700 nanograms per kilogram of body weight per day -- if parents mix formula with drinking water containing perchlorate levels of 15 ppb.
OMB officials said during the drafting process that there was "no need" for detailed data to flesh out a table suggesting that infants would be exposed to perchlorate levels above the academy's recommendation.
To determine safe levels of exposure, the administration opted not to use the academy's "reference dose," a formula that includes a tenfold safety factor to protect children and vulnerable populations, and instead used a computer model developed by the Chemical Industry Institute of Toxicology. EPA officials initially inserted language in the document calling this a "novel approach," but the OMB deleted that language.
Federal officials have yet to determine the extent of perchlorate contamination nationwide, but it is known to be widespread. The GAO, which produced a 2005 report calling for a better federal tracking system for perchlorate, found that limited EPA data show the chemical compound has polluted the soil, groundwater and drinking water in 35 states and the District and has contaminated 153 public water systems in 26 states.