The chemical's links to reproductive problems are hotly disputed.
Bisphenol A is found in everything from plastic baby bottles to the liners of tin cans—and it may or may not be bad for you.
Since the mid-1990s, the estrogen-like chemical has been the focus of an escalating, ugly debate between two groups of scientists. One group has argued vehemently that the chemical is dangerous and must be banned even as the other, equally vigorously, has defended its safety.
In late July, a group of scientists published a report concluding that exposure to even low levels of bisphenol, particularly during development, can cause serious reproductive problems (Reprod. Toxicol., doi:10.1016/j.reprotox.2007.07.005).
It's certainly confusing and I wish it weren't. Michael Shelby, US National Toxicology Program
But less than two weeks later, an independent panel assembled to advise the US government expressed “negligible concern for adverse reproductive effects.”
The public disagreement has incited accusations of bias on both sides and left consumers bewildered. Should they avoid the chemical? Or not?
“It's certainly confusing and I wish it weren't,” says Michael Shelby, director of the US National Toxicology Program's (NTP) Center for the Evaluation of Risks to Human Reproduction, which convened the government panel. “According to the panel results, if there is concern, it's not great,” Shelby says.
Critics have accused the government panel of industry bias, forcing the NTP in April to dismiss a contractor that had ties to the chemical industry. The US Environmental Protection Agency has also come under fire for considering scientists with conflicts of interest to assess the safety of acrylamide, a neurotoxin found in fried and baked goods.
Some experts also question the government panel's methods, in particular the decision to exclude as sources of information more than half of the 124 papers published on bisphenol A's effects on development.
“I was shocked,” says Beverly Rubin, a bisphenol A expert at Tufts University in Boston. “[The review] was bizarre, sloppy and very arbitrary. They discounted a lot of very good work and then left in a lot of work that's not so good.”
What's more, the panel accepted 70% of industry-funded papers, but only 30% of those from academia, notes Ana Soto, a developmental biologist at Tufts. “Why would they do that?” Soto asks. “It's mind-boggling.”
Robert Chapin, a researcher at the pharmaceutical company Pfizer and chair of the advisory panel, denies allegations of industry influence. “We established scientifically valid criteria and then we held those up to each study in turn,” he says.
Chapin says Soto and others may be too passionate to be scientifically rigorous. “This might be a case where people are putting advocacy before science,” Chapin says.
Some of the uncertainty about bisphenol A is the result of a dearth of human studies, which have to last through the long lag time between exposure and any effects. The National Children's Study, scheduled to begin next year, aims to follow 100,000 American children from the womb until age 21, and will assess exposure to bisphenol A as well as to heavy metals, pesticides and other pollutants. But the first results won't emerge for at least five years.
The NTP is also not likely to have an official position on bisphenol A for at least six months. In the meantime, Shelby recommends that those consumers who are concerned should switch to bisphenol A–free alternatives, such as glass baby bottles instead of plastic, and fresh and frozen foods as opposed to canned goods.
Kimberly Thompson, a risk analysis expert at Harvard University, advocates that individuals become informed so that they can assess their own risk. “[But] it's very tricky,” she says. “People really do rely on experts to understand whether they should or shouldn't be concerned.”
Related links in Nature Research
Related external links
About this article
Cite this article
Willyard, C. Allegations of bias cloud conflicting reports on bisphenol A's effects. Nat Med 13, 1002 (2007). https://doi.org/10.1038/nm0907-1002a