What we don’t know about chemicals CAN hurt us

We can no longer operate under the assumption that what we do not know about a chemical substance cannot hurt us. Tragic results associated with too many toxic substances have taught us that lesson all too well. Chemicals, not people, must be put to the test.

Senator James Pearson (KS) co-sponsor of the 1976 Toxic Substances Control Act (March 26, 1976)

The lifeblood of chemical safety evaluation is data. We know that dioxin, PCBs, asbestos, benzene, lead, mercury, formaldehyde, PFOS, and trichloroethylene cause cancer, birth defects and other diseases because scientific studies tell us so.

Test data doesn’t materialize out of thin air. If we don’t do studies, we won’t know whether chemicals can harm people and may erroneously assume that they are safe. We’ve learned this lesson the hard way: time and again, ubiquitous chemicals like asbestos and PCBs have caused widespread damage that might have been prevented if timely testing had been conducted.

The data we have on the hazards of the tens of thousands of chemicals in commerce is woefully inadequate. Research conducted by the federal government and academic institutions can help fill this gap but, on its own, will fall far short of meeting our needs. Most people would agree that safety testing is a basic responsibility of the companies that profit from making and selling chemicals. But the chemical industry has a poor track record on chemical testing and experience has shown that it will not step up unless the government forces it to do so.

The Trump EPA is failing a key test

From its enactment in 1976, a central tenet of the Toxic Substances Control Act (TSCA) has been that “adequate data should be developed with respect to the effects of chemical substances and mixtures on health and the environment and that the development of such data should be the responsibility of those who manufacture and those who process such chemical substances and mixtures.” In 2016, at the urging of Safer Chemicals Healthy Families and other groups, Congress strengthened TSCA to give EPA increased authority to require the industry to test chemicals. Yet not a single chemical has been subjected to these expanded testing requirements. Even worse, the current EPA leadership seems to believe that it can make chemical safety decisions based on inadequate data and that the absence of data means the absence of risk.

Chemical testing has a central role in TSCA

The 2016 TSCA amendments created a new integrated process for prioritizing and then conducting risk evaluations on chemicals of concern and at the same time broadening TSCA’s testing authority. These two improvements were intended to work hand-in-hand: Congress expected that, by requiring more testing, EPA would obtain the “best available science” for determining the risks of chemicals and in turn would conduct robust, comprehensive risk evaluations.

Demonstrating this linkage, TSCA section 4 was amended to authorize EPA to require testing “for the purposes of prioritizing a chemical substance” and “to perform a risk evaluation.” And to enable testing to be conducted quickly and efficiently, Congress empowered EPA to issue testing orders instead of relying on resource-intensive rulemaking.

Retreating from pre-prioritization

The Obama EPA recognized the interdependence of high-quality safety evaluations and strategic data collection in its January 2017 proposed rules for prioritization and risk evaluations. These proposals sought to establish a “pre-prioritization process” to assess the sufficiency of available data well in advance of listing chemicals as high-priority and to “do a significant amount of upfront data gathering . . . if information gaps are identified.”

But the chemical industry opposed this process and the Trump EPA dropped it from the final rules it published in July 2017. This retreat sent a strong signal that the agency was not serious about using its new testing authority to support its initial risk evaluations.

Pigment violet 29: untested hypotheses, preliminary data, and sweeping extrapolations collide

We are now seeing the unfortunate consequences of this decision. In December 2018, EPA released its first draft risk evaluation, for Pigment Violet 29 (PV29). The draft evaluation concludes that this chemical does not present an unreasonable risk of injury but bases this sweeping conclusion on limited hazard and exposure information that most scientists would consider inadequate to demonstrate the absence of risk. Comments by Safer Chemicals Healthy Families and other groups strongly faulted the draft evaluation for giving PV29 a clean bill of health without a credible scientific basis.

For example, reliable experimental data on PV29 were available for only 5 of the 15 critical health effects that EPA’s Safer Choice program uses to identify non-hazardous chemical products. Moreover, EPA’s determination of “safe” exposure levels was entirely based on a screening study for reproductive and developmental toxicity that EPA guidelines term “insufficient by itself to make an estimate of human risk without further studies to confirm and extend the observations.” Similarly, without any carcinogenicity data, the draft risk evaluation asserted that PV29 is “unlikely to be a carcinogen” – a classification that under EPA guidelines for assessing cancer risk is only justified where “animal evidence (sic) demonstrates lack of carcinogenic effect in both sexes in well-designed and well-conducted studies in at least two appropriate animal species.”

While we can’t say for sure that PV29 has adverse health effects, we certainly don’t have the data to say with confidence that it is not dangerous. EPA has no business assuring the public that PV29 is safe when it simply doesn’t know.

Had EPA used its TSCA section 4 authority when it selected PV29 for a risk evaluation in the fall of 2016, this data vacuum could have been avoided. A testing order could have set in motion several studies on health effects that would now be in hand to inform the ongoing risk evaluation.

Dropping the ball on upcoming chemical risk evaluations

Despite the recommendations of Safer Chemicals Healthy Families, EPA made no effort to conduct an early assessment of data gaps for the other nine chemicals now undergoing risk evaluations. Several of these chemicals lack information on health concerns like endocrine effects and developmental neurotoxicity as well as human exposure and environmental release.

We are about to see these problems magnified as EPA prepares to embark on the next round of risk evaluations. By the end of this year, the law requires EPA to designate 20 chemicals as high-priority and 20 as low-priority. EPA just announced the 40 chemicals for which it plans to make these designations.

Since the high-priority candidates are all on EPA’s 2014 “Workplan” list, it could have begun to identify data gaps on these chemicals and issued testing orders to fill them shortly after the new law took effect. Yet no such efforts have been undertaken. While the 20 high-priority candidates have received some testing, they have not been studied for all health effects and available data is likely insufficient for comprehensive determinations of risk. Unless EPA takes immediate action, it will be too late to fill these gaps.

The candidates for low-priority listing will not undergo further evaluation because EPA must conclude that existing “information is sufficient to establish” the absence of health or environmental risks. These chemicals, too, have not been tested under TSCA. In the likely event that unanswered questions arise about their health or environmental effects, low-priority listing will be unjustified.

Ignoring safety testing needs for new chemicals

Section 5 of TSCA should provide a first line of defense against the hundreds of new chemicals introduced each year by ensuring that they do not enter commerce without a careful evaluation of their potential risks. Recognizing that EPA needed more authority to obtain data on new chemicals, the TSCA amendments directed EPA to require testing where available information “is insufficient to permit a reasoned evaluation of the[ir] health and environmental effects.” But the Trump EPA is not only ignoring this mandate but actually requiring less testing today than under the old law.

Following a recent change in policy, EPA has allowed dozens of new chemicals to enter production without testing, based on scientifically dubious determinations that they are “not likely to present an unreasonable risk of injury.” For chemicals with suspected adverse effects, EPA is relying on voluntary exposure controls to demonstrate the absence of risk. However, EPA’s calculations of “safe” exposure levels are uncertain because no data are available on the new chemical itself, the data on analog (similar) chemicals are often incomplete or of limited relevance, and exposures are poorly characterized. Although these uncertainties preclude a “reasoned evaluation” of risk, EPA is turning a blind eye to the need for testing under the law.

A cynic would say that industry and the Trump EPA are resisting testing because they know it will reveal new and unsuspected hazards for chemicals now assumed to be safe. But Congress rejected this mindset when it enacted TSCA in 1976 and again when it strengthened the law in 2016: it wanted EPA to lift the veil of secrecy over the chemical industry because it recognized that we cannot protect the public from dangerous chemicals without timely data. By returning to the pre-TSCA era of toxic ignorance, EPA is rejecting this accepted principle of chemical risk management.