Why we don’t know if plastics are safe

How can we determine conclusively whether a chemical is safe?

That’s the question Jerry Heindel, a top health scientist in the United States government, wanted to answer. And what better chemical to ask this question about than bisphenol A — one of the most researched chemicals in the world, and one of the most controversial.

BPA, as the chemical is known, is an industrial additive essential in the production of many plastic consumer goods: water bottles, plastic food containers, dental products. And yet, despite hundreds of studies dating back nearly a century, there’s no scientific consensus on whether BPA is harmful to human health.

That’s in part because any effects the chemical has would be long-term, complicated and hard to detect. But it’s also because academic scientists and government regulators disagree on how best to evaluate the safety of chemicals.

The solution Heindel — at the time a health science administrator at the U.S. National Institute of Environmental Health Sciences — came up with was to design a study that would pull the two warring factions together in a single effort to assess the safety of BPA.

This is a story about what went wrong, and why scientists are no less sure than before about whether BPA is safe for human consumption.

Launched in 2012, Heindel’s study is the largest safety analysis ever attempted on the chemical. It mobilized scores of scientists and cost tens of millions of taxpayer dollars.

It has also failed, so far, in its goal of bridging the divide between scientists, imploding into acrimony last year when preliminary results published by the U.S. Food and Drug Administration sparked protests from the academics involved. A final conclusion on the safety of BPA also remains out of reach.

This is a story about what went wrong, and why scientists are no less sure than before about whether BPA — a chemical most of us probably have in our bodies — is safe for human consumption.

Blockbuster plastics

Rising worries about BPA coincided with the proliferation of plastics into every part of our economies and lives.

The word “plastics” covers a gamut of synthetic materials that are easy to manufacture, cheap to produce, and — with the help of chemical additives and processes — incredibly versatile. First mass-produced in the beginning of the 20th century, by the 1950s the material had became integral in everything from clothes and cars to furniture and food packaging.

“At the beginning it was just convenient to build a useful material,” Ninja Reineke, a senior policy adviser at the advocacy group ChemTrust, said. “They didn’t take into account the potential implications.”

Researchers had known as early as the 1930s that BPA has a strange side-effect: It weakly mimics the hormone estrogen, which regulates the female reproductive system. But this curious consequence didn’t stop petrochemical companies from using it — and using it a lot.

BPA is essential for the production of two blockbuster plastics: Epoxy resins, used as coatings for everything from electrical cables to shop receipts, and polycarbonate plastic, a hard, clear substance that has come to replace steel or glass in products from cars and electronics to water bottles and Tupperware.

By the late 1980s, nearly half a million tons of BPA were being produced per year in the U.S. alone.

Then, in 1992, scientists at Stanford University discovered that the chemical was “migrating” from the plastic, meaning it could be leaching from packaging into food or drinks and then into our bodies.

In the early 2000s, a survey found BPA in the urine of more than 90 percent of Americans tested. What it couldn’t say was whether that was a problem.

Low-dose effects

Evaluating the safety of substances that affect the hormone system, like BPA, is hard because they don’t behave like most chemicals.

Traditionally, chemical regulation and the science it is based on rests on the assumption that the higher the concentration of a chemical, the more harmful it is. The dose makes the poison.

But a growing number of scientists believe that some widely used chemicals don’t follow this trajectory, and can disrupt our hormonal systems even at very low levels. For these chemicals, they argue, setting strict exposure limits — a common form of regulation — is ineffective.

Effects on the endocrine system — which through a crowd of hormones regulates our metabolism, growth and development, tissue function, sexual function, reproduction, sleep, and mood — can be hard to trace, because the balance of hormones in the body is precarious and affected by many factors.

And yet, amid all this noise a signal is emerging, according to the members of the Endocrine Society, which is made up of some 18,000 scientists around the world who study the endocrine system: Endocrine-disrupting chemicals like BPA are wreaking havoc on our bodies.

“They can really program your genes and development differently” — Ninja Reineke, a senior policy adviser at ChemTrust

The negative effects listed by the Endocrine Society include reduced fertility, diabetes, obesity and prostate cancer for men, and breast cancer for women.

Those concerned about endocrine disrupters say this group of chemicals has a number of unusual properties that make them behave similarly to hormones: Not only can they cause adverse effects even in very low amounts, these effects don’t necessarily have an obvious trajectory curve when you increase the dose.

Instead of a steadily increasing linear graph, the charts used by scientists to illustrate their effects often wiggle like a sine curve.

Members of the Endocrine Society say the effects of chemicals like BPA don’t always show up immediately. Hormone balance is most critical for fetuses and babies. Too much or too little of a hormone — or one of its mimics — early in life can throw off the development trajectory.

“They can really program your genes and development differently,” Reineke from ChemTrust said. “But it may not be obvious until later in life.”

Culture clash

So far, these findings — most of them made by scientists conducting research in universities — have failed to sway the government bodies that regulate chemicals like BPA.

Studies given the most weight by the FDA have to abide by a stringent set of bookkeeping requirements and laboratory best practices. This set of “good laboratory practices” (GLP) originated in a push to make sure that scientists weren’t cutting corners or faking data, but has since become the gold standard for regulators.

But while good laboratory practices are the norm for scientists working at the FDA or in industry-funded labs, many academics don’t have the resources to meet their stringent requirements. Besides, academics have their own gold standard: peer-reviewed journals and university oversight.

The result is a culture clash that can sometimes produce wildly different conclusions — with regulators often discounting studies conducted by academics when assessing chemical safety.

“[Regulators] tend to put much more weight on studies that have been done according to guidelines, with lots of animals and the kind of record-keeping that allows them to repeat,” said John Bucher, from the U.S.’s National Institute of Environmental Health Sciences. “Those kinds of studies are rarely done in the literature.”

The clash is particularly acute in the debate over BPA. The FDA has assessed the chemical several times. Each time, it has found that it’s safe at the levels in which it is present in food.

In the FDA’s latest review of studies dealing with BPA’s potential health impacts in 2014, the agency looked at hundreds of studies, including dozens that found cause for concern at low doses, conducted mostly by university scientists. It determined that only four of them met the standards for setting policy. None of the four found low-dose effects.

An FDA spokesperson said the review was carried out using “predefined, scientifically supported criteria for ranking each study,” and that even if many studies didn’t meet the bar for setting regulatory values, “some studies were identified that did contribute qualitative information to support general conclusions” about whether BPA was harmful. The spokesperson added that extensive reviews of “low-dose” literature on BPA indicated that fears about the chemical’s low-dose effects are not warranted.

Academics in response accused the FDA of cherry-picking data to support a preconceived conclusion. They also said that by not considering the work by university researchers, regulators are ignoring some of the best science on chemicals.

“The scientific apparatus worldwide is more advanced than it’s ever been, and yet it’s being largely ignored in chemical regulation,” said Thomas Zoeller, who studies thyroid hormone action and brain development at University of Massachusetts Amherst.

Ends and means

The main point of disagreement is over what exactly indicates a negative effect.

When determining the impact of a chemical on a specific organ, scientists use an array of “endpoints,” or markers of health effects, to track any changes. These could be the weight of an organ, the development of tumors or lesions, indicators of a disease, or an increase in the release of a bodily fluid or hormone, to name a few.

The FDA maintains a list of “validated” endpoints, which it says are acceptable because the effects can be replicated and they “best characterize risk” posed by a chemical. Before a chemical is identified as hazardous, the FDA said, it must proven that it has not just an effect, but a negative effect, on the body.

Some academics disagree with the FDA’s approach. The agency’s endpoints look mostly for “gross abnormalities,” according to Heather Patisaul, who studies endocrine disruptors and the brain at North Carolina State University.

“They’re using such an insensitive marker. Brain weight doesn’t reflect brain function” — Thomas Zoeller, reseracher

For example, one traditional endpoint for studies on brain development is the weight of the brain — if a chemical causes smaller brains to develop, it’s pretty likely that something is wrong.

But academics like Zoeller and Patisaul say many traditional endpoints wouldn’t pick up the subtleties that a disruption of the endocrine system could cause.

“They’re using such an insensitive marker,” said Zoeller. “Brain weight doesn’t reflect brain function. It’s like, if the engine isn’t working in your car, it’s not really going to be informative to pull the engine out and weigh it.”

They point out that the FDA’s endpoints list hasn’t been updated in decades, since before the debate about endocrine disrupters or low-dose effects began. Instead, they advocate what they call more “modern” endpoints, which focus on signs of specific diseases.

Looking for CLARITY

Enter Jerry Heindel. His idea was to combine regulator-approved practices with academic expertise and modern endpoints.

Have the FDA raise and treat enough animals to satisfy the criteria for their study, and then once they’ve done their research, ship some of the rats off to independent specialists, who can then take a look. At the end, both sides could combine their findings and determine once and for all whether BPA is safe.

The project’s unwieldy name — the Consortium Linking Academic and Regulatory Insights on bisphenol A ToxicitY — was shorthanded to the CLARITY study in an optimistic nod to its goal.

Funded by the U.S. government, the study has in the seven years since its inception attracted the attention of regulators around the world, both for its work on the specific chemical and as an experiment in bridging the rift between academic and regulatory studies.

The European Food Safety Authority, notably, postponed its plans to re-evaluate the safety of BPA until the CLARITY study was published. The agency has also assessed the safety of the substance several times, arriving at the same conclusion as the FDA: that the substance is safe at current levels of consumption.

The alliance between the regulators and the academics was always uneasy. Both sides were, to an extent, staking their credibility on the findings. The FDA had repeatedly found that BPA was safe, and many of the grantee scientists had spent their careers warning about the dangerous effects at low doses.

“Both sides were putting themselves out there,” Heindel said. “If all the grantee studies were positive, the FDA would really have to change what they’re saying about BPA. And the grantees were saying, what if this doesn’t work?”

There were small bumps in the road — “In the course of a 10-year project, there are of course disagreements,” said Bucher, the senior scientist on CLARITY at Heindel’s NIEHS — but as the data was being collected everyone stayed on board.

The 14 academics involved in the study started publishing their results, one by one, and several of them said they found low-dose effects of BPA, similar to their own earlier studies.

“At this point, it’s difficult to see how the two sides will come together” — Jerry Heindel, health scientist

Then, in February, a senior scientist from the FDA published a press releaseannouncing a first draft of the agency’s part of the CLARITY study.

It wasn’t intended to be the final result; the FDA’s conclusions would later be integrated with all the academic studies by the National Toxicology Program for a final report.

But the language was unambiguous: The data from the study, the FDA said, “supports our determination that currently authorized uses of BPA continue to be safe for consumers.”

The other participants were blindsided and accused the agency of breaking its promise to wait until all the work was completed to pass judgment.

Zoeller, who was one of the academic scientists tapped for the CLARITY study, said many of his peers came into the project skeptical of the FDA, and to them the press release only confirmed that the regulator wouldn’t change its mind.

“That was really a poke in the eye,” he said.

Deep divide

The FDA softened its conclusion slightly in the final presentation of its study the next month, but the damage was done.

The controversy crested with two webinars in September, on two consecutive days — one hosted by some academic participants of the study and one hosted by the FDA — where the two sides presented wholly different interpretations of the same data.

Some academic authors hit back at the agency, saying that the FDA’s data “simply do not support [its] conclusion.”

If the FDA “would look at the data in an agnostic way, and ignore their ideology, then I think we would all agree on things,” said Zoeller. “They’re wrong about this.”

The final report, which will integrate the findings of all the studies, is set to be published later this year. A spokesperson from the FDA said the agency “will carefully consider all applicable results from these studies to determine what if any additional action the agency should take to ensure the safe use of BPA as a food contact substance.”

Bucher, who is working on compiling all the results, said the study won’t seek to produce a single conclusion that every scientist involved agrees on. “We’re trying to capture the range of opinions,” he said.

He said he hopes that if the final results indicate a need to change policies on BPA, that regulators are willing to make them. “I’m not saying there are [changes needed] yet,” he said. “But I’m hoping that regulatory agencies would, if required, change the way they look at certain sets of information.”

Heindel, who retired from NIEHS in 2016 and founded a hub for scientists working on endocrine disrupters, said he still believes his model — of combining regulatory and academic studies — is useful and could be replicated in the future. But the study did not reach its goal of determining whether BPA is safe.

If anything, it’s just created more questions.

“The CLARITY study, if things were black and white, would have been completely on one side or completely on the other,” Heindel said. “But it turns out it’s in the middle, and now we have to understand why that happened.”

Heindel’s experiment was seen by many in the scientific community as more than just an effort to assess the safety of BPA; it was a chance to bridge the divide between regulators and academia that makes it so hard to know if a chemical is safe.

It may have deepened it instead.

“At this point, it’s difficult to see how the two sides will come together,” Heindel said.

This is the second article in a series on plastics and the body.