Perchlorate report doesn’t dispel controversy
Revised 1/28/05
Perchlorate is not as hazardous as U.S. EPA scientists estimated, but it is more harmful than many perchlorate manufacturers have suggested, says a National Academy of Sciences (NAS) committee in a long-awaited report released on January 10.
The safe dose recommended by the committee, 0.7 micrograms per kilogram of a person’s body weight per day (µg/kg-day), is 23 times higher than EPA’s draft reference dose but at least 7 times lower than perchlorate industry groups’ recommendations. The report, however, is unlikely to dispel controversy about safe levels of perchlorate in drinking water because calculating such levels involves additional information about exposure sources as well as choices about whom to protect, according to Alan Roberson, regulatory affairs director at the American Water Works Association in Washington, D.C.
The U.S. National Academy of Sciences (NAS) has recommended a reference dose of 0.7 micrograms per kilogram per day. To convert this dose into a drinking-water standard, regulators will have to determine or estimate how much perchlorate exposure comes from the food and water a person consumes per day, based on their body weight. The U.S. EPA¨s default assumptions are a 70-kilogram male who drinks 2 liters of water per day and gets 20% of the dose through water, but regulators can use other numbers. California¨s proposed drinking-water standard of 6 ppb assumes that water contributes 60% of exposure.
| Drinking-water standard (ppb) | ||||
| Water consumption | % Perchlorate dose from water | |||
| 70-kg male | 100% | 60% | 20% | |
| 1 liter/day | 49.0 | 29.4 | 9.8 | |
| 2 liters/day | 24.5 | 14.7 | 4.9 | |
| 50-kg pregnant female | 100% | 60% | 20% | |
| 2 liter/day | 17.5 | 10.5 | 3.5 | |
| 2.5 liters/day | 14.0 | 8.4 | 2.8 | |
| 4-kg baby consuming only fluids | ||||
| 0.6 liters/day | 4.7 | n/a | n/a | |
Despite numerous questions about these choices and the drinking-water issue during a January 10 press call, committee members refused to comment on it. They explained that the assumptions used to derive drinking-water standards involve public-policy choices that were beyond the committee’s charge. However, off the record, several committee members say they believe the data support the conclusion that a standard of about 20 ppb would be protective of pregnant women, fetuses and newborns. “The data show that 200 ppb is protective for healthy adults and 20 ppb is protective for sensitive populations,” one member says.
“This is a political decision that gives EPA flexibility,” says Roberson, who adds that from this point, crafting a regulatory standard usually takes 5 or 6 years. Tasks facing EPA include evaluating the 192-page report, producing another draft risk assessment, and making a determination to proceed with a regulation. Further data on perchlorate in food will be acquired by the U.S. Food and Drug Administration (FDA) and other groups—FDA’s preliminary survey found perchlorate in lettuce and milk from across the country. In the meantime, states are likely to carry on developing their own standards, Roberson says.
Perchlorate disrupts thyroid function by competitively inhibiting iodine uptake in a dose-dependent fashion. Because thyroid hormones play a major role in brain development, EPA and the NAS committee focused on making sure that fetuses and infants would be protected.
The committee based its dose recommendation on the 2002 Greer study, in which healthy men and women were given perchlorate to determine at what dose iodine inhibition occurs (Environ. Health Perspect. 2002, 110, 927–937). That study found no significant inhibition at 7 µg/kg-day, a conclusion supported by four additional studies. The committee applied an uncertainty factor of 10 to protect the fetuses of pregnant women who might have hypothyroidism or iodine deficiency and hence came up with the 0.7µg/kg-day recommendation.
Risk assessments are usually based on an identification of adverse effects in a sensitive population. But for perchlorate, the committee emphasized that the reference dose should be based on inhibition of iodine uptake by the thyroid in humans, which is not an adverse effect but the key biochemical event that precedes any health effects caused by perchlorate exposure. “Iodine uptake inhibition is more reliable, quantitatively valid, and unequivocally demonstrated in human experiments,” says committee chair Richard B. Johnston, Jr., of the University of Colorado School of Medicine in Denver. “Using human data in a controlled setting with dosing by mouth is the optimal way to derive a reference dose,” he says.
The committee’s analysis differed from EPA’s, in part because they gave more credence to human than animal studies. “We had three of the best thyroid people in the country and many physicians. We are comfortable with human data and confident that this is a conservative health-protective approach to perchlorate risk assessment,” Johnston says.
The National Resources Defense Council, at a January 10 press briefing, accused the White House and the Department of Defense of unduly influencing the report and questioned whether some committee members had conflicts of interest. NAS executive officer William Colglazier denied the allegations, saying that there was nothing unusual about how the study was conducted.
