The National Academy of Science Perchlorate Report

The National Academy of Sciences (NAS) came out yesterday with its report on the health risks from perchlorate in drinking water. Perchlorate is an oxidizer used in solid rocket propellant. Perchlorate leaches readily through soil, and is a groundwater contaminant in many locations in the U.S. Millions of people are potentially exposed to low levels of perchlorate in their drinking water. There is the concern that low-level perchlorate exposure may result in neurodevelopmental effects in young children from disruption of thyroid function. Regulatory agencies have labored for years to estimate perchlorate levels in water that would not be associated with adverse health effects in humans. In 2002, the Environmental Protection Agency (EPA) prepared a health risk assessment, which concluded that a level of one part-per-billion (ppb) in drinking water would not be associated with adverse health effects in humans. That assessment was subsequently challenged by the Department of Defense (DOD) and members of the defense industry, and eventually ended up with the NAS. The NAS’s report was just published (a prepublication version can be downloaded here).

The NAS committee was charged with assessing the state of the science with regard to thyroid toxicity of perchlorate, evaluating the animal toxicity studies used to assess human health risks and, based on this review, determine if the findings in EPA’s risk assessment for perchlorate were consistent with the current scientific evidence.

As yet, I’ve only skimmed through the report (it’s over 200 pages), in order to get this posted in a timely manner. I skipped to the end to see how EPA’s risk assessment fared. The NAS committee disagreed with EPA’s use of animal toxicology data showing changes in the thyroid gland as the no observed adverse effect level (NOAEL) for developing the Reference Dose (a short definition of the RfD is here). It recommended that the inhibition of iodide uptake by the thyroid in humans, a key biochemical event, and a predecessor to adverse effects as the basis for the RfD. It recommended using, in combination with other evidence, a short-term ingestion exposure study involving healthy adult men and women, which identified a no-observed-effect level (NOEL) for inhibition of iodide uptake by the thyroid. Based on this study, the NAS committee recommended a RfD of 0.0007 mg/kg-day, which if you do the arithmetic comes out to something a little more than 20 ppb in drinking water. The EPA’s 2002 risk assessment estimated a RfD of 0.00003 mg/kg-day, corresponding to approximately 1 ppb in drinking water. I'll need to post more later about the reasoning underlying the committee's recommendation.

This should be interesting. Last month, in a story published in the Riverside (Inland Southern California) Press-Enterprise, it was reported that one of the authors of the RfD study recommended by the NAS committee had rewritten a news article on perchlorate, which had been submitted by a free lance writer and published in the same issue of Environmental Health Perspectives as the RfD study. The story had been rewritten to portray perchlorate risks as a less significant public health concern. It’s an interesting account, and the P-E kindly provides “before” and “after” versions of the free lancer’s article (free registration required). While it's possible that this was a standard practice, it’s the sort of thing that lends credence to what some people say about risk assessment.

Even before the report had been issued, the NRDC challenged that the NAS panel had been influenced by the White House, the Defense Department and defense contractors. The scope of that influence allegedly included lobbying of the White House by defense contractors, lobbying of the NAS by DOD and defense contractors, DOD and White House involvement in drafting the charge to the NAS panel, and the seating of panel members with views sympathetic to the DOD and the defense industry.

Sorting all of this out is going to take some time. FOIA requests to the White House and DOD by NRDC resulted in a harvest of documents. For example, in discussing the lobbying of the NAS committee, NRDC’s expectation appears to have been that the committee should have been sequestered like a jury. At this point, it’s hard to know if the meetings and submittal of documents to the committee by DOD and industry members represent the normal interplay during an expert review, or lobbying. I’m no real judge in these matters, and I welcome the views from anyone who has served on expert panels at this level. However, the redactions and refusal to produce documents by the DOD and White House in response to the FOIA request, as reported by NRDC, are troubling at the very least.

Expert scientific panels are not really practicing science, and do not make regulatory policy. However, because they operate at that interface between science, policy and regulation, the stakeholder interaction (industry lobbying, NGO activism) will inevitably be part of the process, and not on a level playing field either. The one thing underpinning the assessment of environmental health risks is the credibility of that process. Individual citizens can’t judge whether or not the rat bioassays or epidemiological studies were conducted properly, or were characterized appropriately in risk assessments. Even when the information is presented in detail, such as in the NAS report, you’re pretty much taking it at face value. The outcome won’t be viewed as trustworthy if the process isn’t perceived as transparent and credible. Maybe David Brin has a point with his views about the transparent society. Stay tuned as this story evolves.

Postscript: the Environmental Working Group (EWG) views the NAS report as an endorsement for a low drinking water action level for perchlorate. That’s a point that bears further examination. Depending on what is the distribution of drinking water exposures, there may not be a lot if difference between 1 and 20 ppb, or it could represent a step-function in potential health risks.