Skip to main content

In Assessing Benefits and Risks of Controversial Chemicals, Experts Find Little Certainty

The battle over the chemical bisphenol A, or BPA, widely used in polycarbonate plastics and epoxy resins in food cans, holds lessons for the assessment of a broad range of chemicals whose possible impact on human health remains elusive, speakers said at a recent AAAS panel discussion.

Some studies in animals suggest that BPA can disrupt the endocrine system, with potentially harmful effects on growth and development. Canada plans to ban plastic baby bottles made with BPA, and some groups in the United States and Europe are calling for the same action.

BPA can leach into food from the protective internal epoxy resin coatings of canned foods and from consumer products such as polycarbonate tableware, food storage containers, water bottles, and baby bottles. It has been found to be present in detectable levels in a large portion of the U.S. population.

Sarah Vogel

But can low-level exposures to the chemical during critical periods of development cause adverse health effects? Trying to answer that question has been difficult and contentious, said Sarah Vogel, a Johnson Family Foundation program officer whose doctoral dissertation was on the scientific, political, and economic history of BPA.

Nor is the question limited to BPA. As Jon Hamilton, an NPR science correspondent who moderated the panel discussion, noted: “BPA is one of many chemicals that have these same problems of nuance and complexity. We are just only beginning to understand them. Regulatory agencies are still trying to figure out what to do with them.”

Jon Hamilton

The 1 November panel on “Coming to the Table on Food Safety — Bisphenol A and Beyond,” was the third in a series of “Science & Society: Global Challenges” discussions co-sponsored by the AAAS Center for Science, Technology and Security Policy; the American Chemical Society; and the Georgetown University Program on Science in the Public Interest.

In large part, the BPA challenge comes down to the adequacy of the science. During the 1990s, Frederick vom Saal at the University of Missouri and others began to raise questions about the reproductive effects of BPA at exposure levels well below the current regulatory safety standard. Two industry-sponsored studies failed to replicate vom Saal’s findings, and the debate was on. The U.S. Food and Drug Administration and the European Food and Safety Authority have defended the current exposure standard for BPA (50 micrograms per kilogram of body weight), although the FDA’s own subcommittee of science advisers criticized the agency for relying solely on the two-industry funded toxicity studies for its assessment of BPA.

Christopher Borgert

Christopher Borgert, a University of Florida toxicologist and president of Applied Pharmacology and Toxicology Inc. in Gainesville, Florida, said there can be misconceptions about the meaning of subtle endocrine effects in the research studies. The endocrine system maintains relatively constant internal body conditions, he said, much like a heating and air conditioning system that clicks on and off constantly in a home. “One can measure very subtle changes in the body, and those may be simply a signal that the endocrine system is doing its job,’’ Borgert said.

Scientists should stick to their data, Borgert argued, and not become advocates for policy or regulatory changes that are not borne out by the experiments to date. For some, he said, the data has become “almost secondary to a belief system.” He urged adherence to three fundamental tenets of science: Verify that you are measuring what you say you are measuring, within a prescribed rate of error; control for extraneous factors that can influence the experiment; and require that the observations are replicated independently by others using similar tools and methods.

“A lot of the data on BPA, when looked through that prism, doesn’t really fulfill the definition of scientific evidence for the adverse effects suggested,” Borgert said. That’s not to say the results aren’t useful for making hypotheses to be tested, he said.

The National Toxicology Program (NTP), an arm of the federal Department of Health and Human Services, found in a 2008 assessment of hundreds of studies of BPA at or below the safety standard that there was “some concern for effects on brain, behavior and prostate gland in fetuses, infants and children at current human exposures to bispheonol A.”

And, it added: “The possibility that bisphenol A may alter human development cannot be dismissed.”

Vogel and others have argued that standard toxicology methods for determining whether a substance is safe need to be reassessed, with more emphasis on low-dose studies and more attention to subtle effects beyond death, malformation, and cancer. In a 2001 report on low-dose studies, the NTP found that “the current testing paradigm used for assessments of reproductive and developmental toxicity should be revisited to see if changes are needed regarding dose selection, animal model selection, age when animals are evaluated, and the end points being measured” following exposure to hormone-mimicking agents.

Borgert said that while there have been hundreds of low-dose studies on BPA, there has not always been a consistency of approach. There are different species and strains of rodents involved, and there remain questions about how the species actually relate to human beings. “Mice aren’t rats,” Borgert said, “and rats aren’t little people.”

Mark Duvall

Mark Duvall, an attorney with Beveridge & Diamond PC, a firm specializing in environmental law, said scientists are performing a valuable public service when they do hypothesize about the potential risks of exposure to chemicals, even when all the evidence is not in. And regulators often act in the face of uncertain science, he said.

“The regulator takes science, often times conflicting science, and other factors, applies it against the regulatory and statutory standards and, in the presence of uncertainty, makes a regulatory decision,” Duvall said. “That has to withstand judicial review and that’s a big challenge.” The decisions are sometimes struck down by courts because “there wasn’t a strong enough evidentiary basis,” he added.

Moreover, Congress sometimes bypasses the regulatory process altogether. Duvall cited a 2008 example in which Congress passed the Consumer Product Safety Improvement Act, which bans lead and several phthalates from toys. “Congress stepped in and short-circuited the regulatory process,” he said.

Often at issue in the political realm is whether and how to apply the precautionary principle, which holds that steps should be taken to control exposure to a chemical in the absence of a scientific consensus that it poses a risk to human health. Under the current regulatory system, critics say, a chemical is essentially innocent until proved guilty. They argue that, under the precautionary principle, manufacturers should be required to prove that a product is safe before it can enter the market.

The European Chemicals Agency has used the precautionary principle to justify policy decisions where there is a substantial degree of complexity and uncertainty about the data. The scientific evidence, Vogel said, may not prove cause-and-effect, but some action still may be justified to reduce risk or prevent harm in human populations, particularly when there is exposure to multiple chemicals whose interactions may be poorly understood.

Duvall said the precautionary principle also was used by the city of San Francisco when it passed an ordinance—later repealed—that banned childcare products and toys containing BPA. Canadian authorities also used the principle in proposing to ban baby bottles with BPA, he said. Legislation introduced by Democrats in the U.S. Congress to overhaul the Toxic Substances Control Act also embraces the precautionary principle, Duvall said. When proof of no harm is lacking, the legislation would require regulators to use caution.

Borgert said science cannot offer the sort of ironclad proof that legislators might prefer.. “I can’t prove the negative, not just for BPA but for scores of chemicals,” he said. And while results of ongoing BPA studies may raise concerns, he said, researchers too often “are failing to distinguish what is demonstrable on the basis of validated science and what we can merely hypothesize.”


Learn more about the AAAS Center for Science, Technology, and Security Policy.

Learn more abouit the AAAS Center for Science, Technology and Congress.