Testing analysis for TSCA new chemicals embraced by EPA’s Beck has serious omissions

Richard Denison, Ph.D.is a Lead Senior Scientist.  Ryan O’Connell, EDF High Meadows Fellow, and Stephanie Schwarz, EDF Legal Fellow, assisted in the research informing this post.

[UPDATED 5-15-18:  See clarifications and a correction added in brackets below.]

As noted in a previous blog post, EDF recently filed a request for an extension of the public comment period on EPA’s draft Alternative Testing Methods Strategic Plan under the Toxic Substances Control Act (TSCA).  This was due to the lack of public access to documents that Dr. Nancy Beck had prominently alluded to, without identifying, at EPA’s April 10, 2018, public meeting on the draft plan.  EDF requested a 30-day extension starting once the documents were placed in the docket for the draft plan.

On April 27, EPA provided a 15-day extension after placing the documents in question into the docket.  It turns out the documents (a letter and an attached spreadsheet) are from two animal welfare organizations, People for the Ethical Treatment of Animals (PETA) and Physicians Committee for Responsible Medicine (PCRM).  They are not, were not submitted as, and apparently were not intended to be, comments on the draft plan, however; rather, they raise the groups’ concerns over the increase in testing of new chemicals under the new TSCA, based on an analysis they said indicates EPA more frequently included testing provisions when issuing consent orders for new chemicals after passage of the new law than it did before.

Given that the documents PETA and PCRM submitted were not comments on the draft plan and were not submitted in that context, it is all the more curious why Dr. Beck so prominently noted and expressed such enthusiasm for them at the EPA public meeting held about the draft plan – especially because it appears she did so before EPA had conducted any serious review of the documents, which as you’ll see below, is a big problem.

I suspect Dr. Beck’s interest in the PETA/PCRM letter and analysis has little to do with sparing laboratory animals and much more to do with her seeing the documents as providing a useful pretext for her concerted efforts to avoid imposing testing requirements on new chemicals – a clear priority for her previous employer, the American Chemistry Council (ACC).  Industry has incentives to avoid testing due to its costs and the risks that testing may reveal a chemical presents significant risks to health or the environment.

Now that we’ve had a chance to review the PETA/PCRM documents, I want to use this post to highlight two things:

  • First, the PETA/PCRM analysis erroneously understated the extent of testing EPA required prior to the passage of the Lautenberg Act, because it failed to count any of the testing requirements for two-thirds of the new chemicals it examined for which EPA issued consent orders in 2015 and 2016.
  • Additional context is required when assessing the extent of testing of new chemicals EPA was mandating under the new TSCA, which was not provided in the PETA/PCRM analysis.

Serious omissions in the PETA/PCRM analysis

The letter and analysis provided by PETA and PCRM point to a “dramatic increase in animal testing required for new chemicals in 2017, the first full year of implementation” of the new TSCA.  While, as discussed later, there certainly was an increase, the PETA/PCRM documents significantly overstate the extent of increase by mistakenly excluding much of the testing called for in prior years.

The spreadsheet provided by PETA/PCRM has three tabs showing the data they compiled about consent orders issued in 2015, 2016 and 2017. Each tab has two columns indicating what if any testing requirements were imposed in the consent order.  (The two columns are M and N on the 2015 tab, and L and M on the 2016 and 2017 tabs.)  Here is what we found when we sought to verify the PETA/PCRM analysis:

  • For 2015, both of these columns are blank (indicating no testing is called for in the consent order) for 44 of the 59 new chemicals listed. Yet when we examined their consent orders, testing requirements had in fact been imposed on all 44 of these chemicals.
  • For 2016, both of these columns are blank for 4 of the 17 new chemicals listed. Yet when we examined their consent orders, testing requirements had in fact been imposed on all 4 of these chemicals.
  • For 2017, both of these columns are blank or state “no testing required/recommended” for 10 of the 290 new chemicals listed. Yet when we examined their consent orders, testing requirements had in fact been imposed on 6 of these chemicals.

[UPDATE 5-15-18:  Some of the consent orders noted above apply to more than one new chemical, because the chemicals were submitted together and are structurally related. In some of those cases, the consent order’s testing requirement, once it applies, mandates testing of only one of the substances in the group.  However, the requirement to conduct that testing applies to all of the substances; hence, it is accurate to state as we did that each is subject to a testing requirement.]

PETA and PCRM’s main contention in their letter is that there has been a “dramatic increase in animal testing required for new chemicals in 2017, the first full year of implementation” of the new TSCA.  We have not independently determined whether or not the groups have included all new chemicals subject to consent orders issued during each of those years.  But for those they have included, they have significantly under-counted testing imposed by orders issued in 2015, and also have missed some additional testing in orders from 2016 and 2017.

Together, the PETA/PCRM analysis failed to count any of the testing requirements imposed for two-thirds of the new chemicals it examined for which EPA issued consent orders in 2015 and 2016.

The net result of these omissions is that PETA and PCRM have significantly overstated the extent of increase from 2015 to 2017 in testing imposed by EPA through consent orders for new chemicals.

How did such a large discrepancy develop?  We suspect it’s entirely inadvertent, and that PETA and PCRM relied on the consent order summaries EPA provides in its ChemView data portal, rather than examining the underlying consent orders themselves.  It appears that EPA’s summaries do not always include a description of testing requirements, especially for earlier orders; more recently, EPA’s summaries have been doing so, which may explain why PETA and PCRM missed fewer testing requirements in the later years.

Before making any use of the PETA/PCRM analysis or relying on its conclusions, EPA needs to conduct its own, thorough review and analysis.  [UPDATE 5-15-18:  We have confirmed with EPA staff that some of the testing descriptions have been inadvertently truncated in both the consent order summaries and exported results from ChemView searches.]

Now let me turn to some much-needed qualifications and context when assessing the PETA/PCRM analysis and its implications.

The new TSCA calls on EPA to obtain more, not less, information on chemicals

Among the key reforms made to TSCA by 2016’s Lautenberg Act were expansions in EPA’s testing authority under the law and in EPA’s duties to ensure it had sufficient information on which to base its risk decisions involving both new and existing chemicals.  These reforms were broadly supported as essential fixes to the old law needed to restore the ability of EPA to protect human health and the environment from chemical risks.  Among these provisions are:

  • addition of authority for EPA to require testing through issuing orders, rather than the far more cumbersome rulemaking process that was the only option under the old law;
  • elimination of the requirement that EPA first demonstrate potential risk or very high exposure or release of a chemical before it could require testing, a classic Catch-22;
  • a requirement that EPA base any decision to designate an existing chemical as a low priority for risk evaluation on information “sufficient” to demonstrate the chemical is not a high priority;
  • a mandate that if EPA has insufficient information to evaluate the potential risks of a new chemical, it must issue an order requiring that information and/or imposing conditions sufficient to mitigate any potential risk; and
  • a requirement that EPA consider all “reasonably available information” in making decisions on new and existing chemicals.

The new law also includes provisions requiring EPA to take steps to “reduce, refine or replace vertebrate animal testing” where it can do so while providing “information of equivalent or better scientific quality and relevance for assessing risks of injury to health or the environment of chemical substances or mixtures.”

The primary focus of the new law is clearly to better ensure the safety of both new and existing chemicals, including by identifying and mitigating risks to vulnerable subpopulations such as infants, workers and disproportionately exposed communities.  In doing so, nothing in the law calls on EPA to reduce or avoid testing of chemicals, but rather to reduce testing involving vertebrate animals where viable and scientifically reliable alternatives exist.

So it should be a surprise to no one that, properly implemented, the new law should result in more, not less, testing of chemicals – including testing involving vertebrate animals where scientifically sound alternatives do not yet exist.

What else you should know about the PETA/PCRM analysis

For the reasons just stated, the increase in testing requirements for new chemicals identified by PETA and PCRM should not be surprising and is fully consistent with the letter and spirit of the law.

But, beyond the serious omissions discussed earlier, there are several aspects of the nature and extent of testing identified by PETA and PCRM that need to be understood.

EPA required no up-front testing of new chemicals in 2017

[UPDATE 5-15-18:  We based this section on the PETA/PCRM letter, itself, which identified all of the 2017 testing as either triggered or pended.  Upon closer examination, it appears that for at least five of the 290 new chemicals PETA and PCRM examined, their consent orders do require some up-front testing.  Still, the vast majority of testing required in 2017 was not required to be done up-front.]

First, none of the testing PETA and PCRM identified for 2017 had to be undertaken before the subject new chemicals could enter the market.  Such a requirement would be termed “up-front” testing.  Given that the vast majority of new chemicals are submitted for EPA review without their manufacturers providing any health or environmental safety data, what is actually quite shocking is that in no case PETA and PCRM identified was EPA requiring up-front testing.

Instead, EPA’s 2017 testing requirements take two other forms:

  • For nearly two-thirds of the tests PETA and PCRM identified, the manufacturer of the new chemical can indefinitely continue to produce and sell it without ever conducting the testing. The manufacturer would only have to do the tests if it sought to ease the restrictions EPA imposed on the new chemicals’ production or use.  This would occur through modification of the company’s consent order with EPA, assuming the test data submitted by the company warranted relaxing the applicable conditions.  This kind of testing is what EPA calls “pended testing.”
  • For the other third of the tests, the tests are only required if and when a company wants to increase the volume of the chemical produced above a specified level, or after a specified period of time has passed.  This kind of testing is what EPA calls “triggered testing,” with the volume or time triggers specified in the consent order.  The likelihood of this testing actually being triggered is difficult to discern, because EPA generally redacts the volume or time limits in the public versions of its consent orders.

Not only do most of the 2017 tests PETA and PCRM identified fall into the pended category; most (54% using PETA and PCRM’s numbers) of the laboratory animals involved in the testing do too:  they would be used only if a company opted to conduct pended testing in order to get modifications to its consent order.

The increase in testing is due to more consent orders issued, not more testing per new chemical

PETA and PCRM’s letter does not mention the fact that many more consent orders were issued in 2017 than in prior years:  PETA and PCRM examined 290 consent orders issued in 2017, compared to 17 in 2016 and 59 in 2015. Based on the testing requirements in these orders, it does not appear that EPA was requiring more tests to be conducted on a given new chemical in 2017 vs. prior years; rather, testing requirements were imposed on more new chemicals because more new chemicals were subject to consent orders, and both older and more recent consent orders have routinely included testing.

Heightened scrutiny of new chemicals leading to more consent orders is precisely what Congress intended in amending TSCA in 2016.  If anything, this could and well should have resulted in an increased amount of testing per chemical, which it has not.

All of this may be moot, as it predates EPA’s reversing course on new chemical reviews

Ironically, in some ways, much of what PETA and PCRM have raised and what we have discussed above may be moot.  That is because EPA has made clear its intent to move away from issuing consent orders for new chemicals – despite the clear requirements in the law that it do so when it has insufficient information on a new chemical or finds that the chemical may present an unreasonable risk or lead to high release or exposure.  As discussed at length on this blog, EPA now intends to rely instead on a company’s unenforceable premanufacture notice (PMN), potentially coupled with issuance of a Significant New Use Rule (SNUR), which would only require a company to notify EPA if it intends to make or use a chemical in a manner that deviates from specified conditions.  Neither a PMN nor a SNUR can impose testing requirements on a new chemical, however.  Indeed, industry mounted considerable resistance to the testing EPA was imposing through consent orders and took advantage of its heavy influence over agency policymaking under this administration.  As a result, EPA reversed course and opted for approaches that will not allow EPA to require more testing of new chemicals.

In the face of statutory mandates calling for more testing, it is small wonder that Dr. Beck got so excited when she saw PETA and PCRM’s letter and analysis.  Both the flaws in the analysis and the requirements of the law should have EPA looking closer before it leaps.


This entry was posted in EPA, Health Policy, TSCA Reform and tagged , , . Bookmark the permalink. Both comments and trackbacks are currently closed.


  1. Joseph Manuppello
    Posted May 14, 2018 at 6:14 pm | Permalink

    I am Joseph Manuppello, senior research analyst with the Physicians Committee for Responsible Medicine (PCRM). I conducted the analysis of animal testing requirements for new chemicals reported in EPA’s ChemView database. While Dr. Denison raises several valid points, EDF’s analysis contains meaningful omissions, the details of which PCRM has sent to Dr. Denison, privately.

    • Richard Denison
      Posted May 15, 2018 at 1:39 pm | Permalink

      Joseph: Thank you for your questions and input. Based on it, I have added some clarifications and a correction to the blog post. With these, we stand by this post.

      Best regards, Richard