Inside EPA, Maria Hegstad
Published December 31, 2008
EPA is strongly criticizing a draft Food and Drug Administration (FDA) report encouraging pregnant women and young children to eat more fish than the agency believes it is safe to consume because of the risk of consuming mercury and its effects on fetal and children's development.
"We have very serious concerns about the [FDA] report. There are serious scientific flaws, and this is not a product that EPA should endorse as it does not reach the level of scientific rigor routinely demonstrated by the Agency," writes Peter Preuss, director of EPA's National Center for Environmental Assessment (NCEA), in Dec. 5 comments on FDA's report. Relevant documents are available on InsideEPA.com.
The draft FDA report, An Evaluation of Risk to U.S. Consumers from Methylmercury in Commercial Fish Products, Including a Quantitative Assessment of Risk and Beneficial Health Effects from Fish, has not been officially released by FDA. EPA's comments were written after agency officials viewed the draft report as part of an interagency review process.
EPA and FDA signed a memo of understanding in June 2005 to cooperate to develop public health advisories "regarding the risks and benefits of consuming commercial and noncommercial fish and shellfish," according to the document. The agencies reached the agreement after EPA and FDA released a controversial 2004 advisory on eating fish contaminated by mercury. It warned pregnant woman, nursing mothers and young children to avoid eating four types of fish with high mercury levels, and to limit consumption of recreationally caught albacore tuna and fish.
But a source with the Environmental Working Group (EWG) says FDA did not inform EPA that it was working on a new assessment of the risks and benefits of eating fish, and that EPA belatedly learned of the report last summer. "The most likely outcome for most people is a small beneficial effect on fetal neurodevelopment from the mother's consumption of commercial fish, even though the fish contain methylmercury," according to the FDA report.
An EPA spokeswoman released a statement that the agency is "working closely with other agencies in the scientific review of this report."
The EWG source described FDA's assessment as "mind-blowing" because it does not consider the relative amounts of mercury that different species of fish contain. "They are just telling everyone to eat more fish," the source says. "It's just reckless." The source points to research done by EPA's former top mercury specialist, Kate Mahaffey, who concluded that one in six pregnant women had too much mercury in her blood. In a letter to EPA Administrator Stephen Johnson, EWG urges the agency not to adopt FDA's recommendations.
FDA predicts the greatest benefits from the lean protein, omega 3 oils and minerals in fish when more than 12 ounces of fish are eaten weekly -- EPA and FDA's existing joint safety standard for pregnant women and young children, according to the FDA report.
But EPA points to what it considers numerous flaws in FDA's risk assessment. "The scientific, statistical and methodological limitations of this work are too great for it to serve as a tool for government decision-making regarding risks from methylmercury in seafood -- or for characterizing and communicating these risks to the public," Preuss writes.
EPA criticizes a lack of transparency in FDA's assessments, saying it is "lacking in the approaches used to distinguish between low mercury containing fish and high mercury containing fish," and warning this could cause problems in communicating risk to the public if the document remains unchanged. "The report tends to portray fish consumption simplistically and in a dichotomous way," the comments state.
EPA describes particular concerns with FDA's "misuse and misinterpretation" of EPA's safe daily oral exposure standard, or reference dose (RfD), for methylmercury in FDA's assessment. In comments attached to Preuss' letter, EPA staff reminds FDA that the RfD is a safe population threshold value, not a prediction of risk at a specific value. The EPA comments also criticize FDA for confusing uncertainty and safety factors and remind FDA that the RfD is intended to protect all populations, not just susceptible ones.
And EPA takes exception to FDA's characterization that most RfD values have uncertainty of about one order of magnitude. EPA notes that while this is the case, its IRIS assessment of methylmercury found no evidence for a threshold in the human studies it considered. "It is not necessary to trivialize existing, non-linear methylmercury risk assessments in order to establish a rationale for a predictive risk assessment," the comments state.
EPA criticizes FDA's assumptions and modeling in its exposure analysis, arguing that there is a great deal of data about the amount of fish Americans eat on regular basis, but that FDA did not make use of it in its modeling. It also questions FDA's use of a study of methylmercury effects in adult men to predict maternal blood mercury levels.
EPA also questions the endpoints FDA selected to determine the effects of methylmercury on children's development. FDA's endpoints, a young child's age at first walking and first talking, "are not the most sensitive measure of adverse neurodevelopmental effects seen in methylmercury exposed populations," EPA argues.
"This is an extremely important public health issue and until the concerns raised by EPA have been fully addressed, NCEA cannot provide support for the document and cautions against releasing it," Preuss writes.
