Transparency in Regulatory Science — for Whom?
By Lilly Leibu, Digital Editor at N.Y.U. Environmental Law Journal.
This post is part of the Environmental Law Review Syndicate.
Former Administrator of the Environmental Protection Agency (EPA) Andrew Wheeler purported to rid EPA of “secret science” and “[empower] the American people to demand future transparency” with a regulation that stakeholders such as Big Tobacco and coal-funded groups have lobbied for since the mid-1990s. Under his directive, EPA published a notice for the proposed rule “Strengthening Transparency in Regulatory Science” in 2018. The agency broadened its scope and published a Supplemental Notice of Proposed Rulemaking (SNPRM) two years later on March 18, 2020. Between the first proposed rule and the second, EPA received pushback in the form of nearly 995,000 comments and garnered attention from scientists who have vocalized the need to dissolve it completely. Yet on January 6, 2021, the “Secret Science” rule was finalized. The lifespan of this midnight regulation was cut short when a United States district court struck it down, but its haphazard construction and questionable purpose provide a host of other grounds on which to argue for its rejection. Without proactive measures, a future administration may attempt something similar—endangering public health and the environment in the process.
I. Pushback in Court
Arguments made in court opposing this rule have been successful. In Environmental Defense Fund v. EPA, members of the scientific community proved standing by specifying the upstream repercussions of EPA’s new selection process under this Rule—highlighting difficulties in grant proposals, redesign of studies, and adjustments to cohort recruitment. They are not the only ones to call into question the validity of the final rule, as several states, cities, and organizations brought claims almost immediately after its publication. These coalitions challenged the Secret Science rule on the basis of its status as a “procedural rule” and the unlawful application of the Federal Housekeeping Act. In January 2021, the U.S. District Court for the District of Montana agreed that the rule was misclassified as “procedural” and that the agency did not provide good cause to be exempt from the thirty day notice requirement under § 553(d)(3) of the APA. While this is good news, the crux of the issue must be addressed more permanently.
II. Sources Informing EPA’s Use of Science
A. Statutory Basis
Statutory text informs EPA’s obligations to monitor and mitigate environmental and public health threats. The Clean Air Act (CAA), for example, requires EPA to set National Ambient Air Quality Standards (NAAQS) and to review those standards every five years in light of emerging scientific data. EPA carries out its responsibility under the CAA through the use of “periodic review of the latest peer-reviewed studies of each pollutant’s health and environmental effects, with assistance from the Clean Air Scientific Advisory Committee (CASAC).” EPA has received directives from the executive branch as well.
B. Executive Branch Directives
The Office of Management and Budget (OMB) has required all federal agencies to engage in a peer review process for “influential scientific information” since issuing its memorandum in 2004 (M-05-03), with the exception of urgent health and safety decisions. Agency discretion is given in the formulation of a peer review mechanism; however, this broad leeway is narrowed for assessments considered highly influential. Agencies may adopt or adapt the approach taken by National Academy of Sciences (NAS) when selecting a committee of peer reviewers not employed by the government. OMB’s Final Information Bulletin for Peer Review further outlines general requirements for an agency’s peer review plan so as to provide the public with the transparency needed to gain confidence in the rigor of an agency’s reasoning. Information such as the peer reviewers’ report(s), name(s), credentials, and conflicts of interest are included in these plans and are available for public comment regarding their adequacy.
A 2009 memorandum from the Obama administration on scientific integrity also addresses public trust in the scientific bases for regulatory decision-making. Here, the President designated the Director of the Office of Science and Technology Policy (OSTP) as responsible for the scientific integrity of the executive branch. This includes coordinating with agency and department heads as well as OMB to ensure robust methodology, the selection and retention of highly qualified individuals, and the public availability of information unless otherwise non-disclosable. Executive Order 13563 (EO 13563) expands on the expectations of public participation in the regulatory process and establishes a minimum comment period of sixty days for any proposed regulation. Those who wish may comment on “relevant scientific and technical findings” provided “in an open format that can be easily searched and downloaded” on the internet. Memorandum 13-13: Open Data Policy – Managing Information as an Asset (M-13-13), also a product of the Obama administration, reiterated this principle of transparency by creating a framework through which agencies will ensure that the data they use is available for processing and dissemination. The same year, OSTP directed EPA, among other agencies, to develop and submit a plan elucidating how to increase public access to peer-reviewed, scientific research funded by the agency. The open access plan assures that public access does not infringe on proprietary interests, intellectual property, and personal privacy. Proponents of the Secret Science rule seem less intent on preserving an individual’s confidentiality.
III. Transparency in Science as a Double-Edged Sword
A. A Crisis in the Scientific Community
The emphasis made on encouraging thorough, publicly-available data seemed to increasingly trend upwards; yet the scientific community has what many consider a “reproducibility crisis.” Research practices have fallen prey to several biases, such as hindsight bias, confirmation bias, and they have struggled with high-dimensional dataset resulting in false-positive findings. Clinical data sets are complex, and researchers may need to simplify, or reduce, the data such that the models they employ are not made inefficient by the extra “noise” of unnecessary variables. Environmental health research, for example, must adapt “hundreds of terabytes of inputs” for air pollution predictions, and this “curse of dimensionality” may make it difficult for a researcher to reduce the data to the variables associated with both the exposure and outcome. When used for regulatory policies, flawed data analysis would unavoidably lead to flawed and inefficient decision-making. Scientific publications have proposed collaborative, open science that would expose researchers to evaluations made by a more diverse and thus more meaningful and constructive review of methodology and analyses. M-13-13 and its aforementioned contemporaries echoed this approach as a means of ameliorating the identified issues. However, some politicians have exploited this legitimate weakness in scientific publication to attack the scientific conclusions disadvantageous to their agenda without having to challenge the substance of those publications. Their proposal to “construct explicit procedural hurdles” under the guise of “transparency” would open the door for regulators to reject studies on the basis of their reliance on confidential health information.
B. The Origin of the Final Rule
Representative Lamar Smith (R-TX), the Science Committee chairman from 2012 to 2018, was one such politician. He attempted to exploit the concerns on the sufficiency of the peer review process and its capacity for identifying quality interpretations of data. He issued a subpoena to EPA for the raw data used in a study considered foundational to CAA regulations. The study in question, known as the Six Cities Study, is the product of Harvard researchers analyzing 8,000 participants for a timespan of 14 to 16 years. After adjusting for cigarette smoking and other individual risk factors, the authors concluded that there was an association between particulate air pollution and daily mortality rates. A 1995 American Cancer Society study comprised of data from 1.2 million people was also targeted by the Science Committee for access to its raw data. By making this information publicly available, Rep. Smith argues that he would then have the capacity to democratize regulations by sharing raw data with the public and “independent scientists for review.” Rep. Smith goes on to say that EPA does not “give the American people what they deserve—the truth about regulations.” The authors of the Harvard study maintained that removing names and addresses from the data values does not eliminate the possibility of someone identifying a particular subject on the basis of their hometown or their date of death, for example. Rep. Smith, however, continued his efforts by introducing a bill informally known as the “Secret Science Reform Act of 2015” where he sought to restrict EPA from promulgating rules that are not transparent or reproducible. The 2015 bill did not reach the Senate.
C. Proposed Rule of 2018
EPA’s 2018 proposed rule on transparency in regulatory science clearly echoes the language of Rep. Smith’s unsuccessful bill. Both documents mention making data publicly available such that an independent validation process may take place, and they reiterate the importance of “the best available science” as the framework for EPA’s decisions. However, where the bill is short and imprecise, the proposed rule elucidates with greater detail the desired “change[s to] agency culture and practices regarding data access.” EPA categorizes the need for this proposed rule as double-pronged: increasing the transparency of the regulatory process and strengthening the scientific integrity of the agency’s regulatory decision-making. The background begins by claiming that the public itself would bear the burden of compliance costs and, as such, ensuring public availability of the data and models “pivotal to the regulatory action” is an appropriate measure to tackle the former prong. The agency asserts that providing information such that “the public [may] understand, assess, and replicate findings” would better inform them and provide them with greater transparency. Further, EPA erroneously quotes the “replication crisis” as a partial justification for taking this action.
D. Supplemental Notice of Proposed Rulemaking of 2020
When the SNPRM was published, it addressed seven modifications to the original proposed rule. The most notable of these changes were the last three. The agency proposed using only pivotal regulatory or pivotal science if the data and models are publicly available for independent validation. Models or data that use confidential business information (CBI), proprietary data, or Personally Identifiable Information (PII) are considered “publicly available” only under the condition that they have tiered access “sufficient for independent validation.” The agency, when “promulgating regulatory decisions or finalizing influential scientific information,” would grant greater weight to these studies. EPA proposed this approach in order to “[increase] access to data and models [such that] stakeholders [may] reanalyze the data and models and explore the sensitivity of the conclusion to alternative assumptions while accessing only the data and aspects of the models that they need.”  In addition, the Administrator, who is politically appointed, would have the discretion to grant an exemption for this public availability requirement. The administrator may then exercise this discretion when, say, they find it “impracticable” to impose the public availability requirement on an industry’s study. Lastly, EPA toyed with the idea of using the Federal Housekeeping Statute as its sole source of authority in promulgating the rule.
E. Final Rule in 2021
EPA, despite heavy criticism, pushed forward in finalizing Strengthening Transparency in Pivotal Science Underlying Significant Regulatory Actions and Influential Scientific Information. It acknowledged the dissenting comments but maintains that “transparency assumes no political ideology.” In supporting this statement, EPA mischaracterizes the literature referenced discussing the importance of reproducibility and transparency in science for the sake of public trust and the quality of the scientific conclusions. Scientists have responded to this strategy with continued disapproval as “the strength of . . . scientific evidence” is more appropriately ascertained from “data collection and analysis methods” as opposed to reproducibility. The final rule instructs EPA to give “greater consideration to pivotal science whose underlying dose-response data are publicly available or available through restricted access.” EPA removed the requirement that models be independently validated, acquiescing slightly to the concerned commenters and reducing the scope of the regulation back to the initial 2018 parameters. The rule also finalized the use of the Federal Housekeeping Statute as its source of authority and was set to take effect immediately, citing APA § 553(a)(2), (b)(A), and (d) to justify circumventing the 30-day notice requirement. In doing so, the agency created its own weak spot when challenged in court.
The housekeeping statute reads as follows:
The head of an Executive department or military department may prescribe regulations for the government of his department, the conduct of its employees, the distribution and performance of its business, and the custody, use, and preservation of its records, papers, and property. This section does not authorize withholding information from the public or limiting the availability of records to the public.
Essentially, if EPA can establish the Secret Science rule as one of procedure and establish itself as an executive department, it can derive authority from the Federal Housekeeping Statute. Precedent suggests, however, that neither of those requirements are met.
IV. Federal Housekeeping Statute as a Source of Authority
In the 2018 proposal, EPA listed the provisions of eight separate statutes, such as the Clean Water Act, as the basis for providing general authority to take this action, an act akin to blindly throwing darts at a board and hoping at least one of them sticks. Past case law and direct statutory text from those same environmental statutes would seem to preclude EPA from utilizing this technique. In Global Van Lines, Inc. v. Interstate Commerce Commission, the Fifth Circuit held that “agencies may not rely on general statutory grants of rulemaking authority to promulgate regulations that are otherwise inconsistent with more specific statutory directives.”
A. EPA is Not an Executive Department for the Purposes of the Federal Housekeeping Statute
EPA is not considered an executive department or military department under the Federal Housekeeping Statute. EPA anticipated this argument by citing a 1970 reorganization plan that created the agency and transferred various responsibilities from other agencies to EPA. Functions of the Department of Agriculture and the Department of the Interior, both executive departments under 5 U.S.C 101, were absorbed by the new agency in Reorganization Plan No. 3. Thus, EPA contends, it has “full section 301 or equivalent authority” and congressional intent supports this conclusion. Notably, DHS was also the product of a transfer of authorities from other agencies like the Department of Agriculture and the Department of Energy and Congress, interestingly, formally established the Department of Homeland Security (DHS) as an executive department in 2002. Congress likely would have made EPA an executive department by now if that were its intention.
B. The Secret Science Rule Does Not Qualify as a Procedure
Moreover, EPA’s shoehorning of this rule into the narrow interpretation of “procedure” does not follow precedent. EPA is adamant in characterizing the Secret Science rule as merely “modify[ing] the EPA’s internal procedures regarding the transparency of science underlying regulatory decisions” and, in doing so, attempts to insulate it from judicial review in subsequent agency actions. Vermont Yankee Nuclear Power Corp. v. Natural Resources Defense Council held that “absent extraordinary circumstances,” it is inappropriate for a reviewing court to “prescribe the procedural format an agency must follow.” The Supreme Court also held in F.C.C. v. Schreiber that an agency’s “adoption of [a] procedural rule favoring public disclosure” may not necessarily be considered arbitrary or capricious. These principles are generally sound in light of the specialization and control an agency would need over its own processes. EPA’s allegedly procedural rule, however, does not “provide guidelines or procedures for [implementation]” one would expect to see. By insisting on its classification as such, the agency is carving an opening for the beneficiaries of lax regulations.
In reviewing the Secret Science rule, The U.S. District Court for the District of Montana found that the “critical factor” a court must use in distinguishing between a procedural and substantive rule is whether the rule allows the agency to freely exercise discretion on a case-by-case basis. When a rule “narrowly limits” this discretion or “establishes a binding norm,” it is considered substantive. Here, EPA, in attempting to convince the court of the rule’s nature, repeated the phrases “purely procedural” and “only applies internally” ad nauseam. The court saw through this, however, and held that it does not distract from the “clear language of requirement” demonstrated by “shall” as opposed to “may.” EPA did not have the authority it sought to vest under the Federal Housekeeping Statute and thus could not evade greater accountability for its actions.
V. The Rule Does Not Achieve the Purposes It Asserts
A. The Rule Does Not Serve the Public
The Secret Science rule acts as a Trojan horse for politicizing what should be an apolitical, open process. The controversy surrounding the widely-used insecticide chlorpyrifos, for example, illustrates the vulnerabilities when selecting the “best available science.” Columbia University’s Center for Children’s Environmental Health (CCCEH) released a study in 2006 linking attention problems and developmental delays in children born to women exposed to the insecticide during their pregnancies. Additional studies have also observed “structural abnormalities in children’s brains, diminished IQ scores and Parkinson’s disease in adults.” EPA proposed a ban on chlorpyrifos under the Obama administration; however, in 2017, then-Administrator Scott Pruitt reversed the action. The agency cites pushback from Columbia to release private information on the children and mothers of the study as a justification for reverting from their original stance. It further stated that the quality of the university’s study “is hard to assess when raw analytical data have not been made available, and the study has not been reproduced.” The agency admitted it “[did] not have a specific reason to believe that CCCEH have inappropriately handled the data or statistical analysis.” Researchers at Columbia offered to explore the possibility of viewing the private datasets in a secured location; however, EPA was not interested. Without being able to make sensitive information publicly available, EPA rejected moving forward with a chlorpyrifos ban. The Secret Science rule would effectively ban the use of these studies. While the rule claims to “strengthen . . . [EPA’s] obligation to ensure the Agency is not arbitrary in its conclusions,” it is unclear how, after the scant reasoning provided, the agency could confidently claim it upheld its obligations to protect public health.
B. The Rule Does Not Demand Higher Quality Studies
If the final rule does not primarily serve the public, then would it at least satisfy the second prong of improving data analyses to counter the reproducibility crisis? According to the American Association for the Advancement of Science (AAAS) and other professionals, no. Environmental epidemiological studies function differently than the clinical studies used in, for example, the Department of Health and Human Services. Scientists monitor the participants because of where they are located and other identifying characteristics. Anonymizing this information—that is, detaching a name from the data—would not completely protect these individuals for the reasons offered by the Harvard scientists in the Six Cities Study. More so, researchers warned that the rule would affect the consideration of past and future studies. Environmental epidemiology “often [looks] retrospectively at previous exposures based on residential location” and serves as a vital source in detecting a particular contaminant’s adverse effects. In the future, individuals may not want to participate in such studies if their data will be non-anonymized, effectively restricting access to key information. Scientific organizations have also wondered what may happen to “situationally unique research” where reproducibility is extraordinarily unethical and infeasible. Observations made on the Deepwater Horizon oil spill, for instance, account for some of the Seattle Aquarium’s water quality and pollution research. EPA, after publishing the SNPRM, addressed this particular concern rather vaguely in an email to the aquarium’s president and CEO, saying that they would not reject situational research on the basis of its non-replicability. However, the agency spokesperson did not expand on how these studies would be incorporated. This emphasis on reanalysis of data could easily lend itself to becoming a means through which industry stakeholders assess data in a light favorable to their objectives. This rule increasingly looks like a Trojan horse.
Limiting or “exclud[ing] inconvenient scientific evidence without acknowledging that it has been excluded” seems like a far cry from strengthening transparency and improving public confidence. In fact, even former President Trump’s EPA Science Advisory Board (SAB) seemed shocked by the proposed rule when it was published in 2018. The field, according to them, has already been moving towards greater transparency while maintaining the confidentiality required of public health studies. SAB had also stated in their memo that the rule’s design received no input from scientists, and it did not seem to consider implementation costs. According to the Congressional Budget Office, the 2018 proposed rule was estimated to cost $250 million initially and then around $1 million to $100 million per year. This led to uncertainty regarding how EPA may conduct its benefit-cost analyses (BCA) and threatened to cut the number of studies relied on by EPA in half. One can imagine that between the costs of having to redact data, make it publicly available, and the time it would take to do so, relevant or innovative studies would be tossed out early on in the data evaluation stage. This method would fail to “examine evidence in its totality” such that, in EPA’s determination of cost-effective regulatory policy, the “best available science” is conflated to mean whichever study is most convenient according to non-scientific parameters. When confronted with high quality studies, EPA will give greater attention to the study with publicly available data—that therefore can be independently verified—than those that include PII, CBI, or other non-anonymized confidential information that cannot be shared with the agency. Scientists may then resort to “becoming politically entrenched dogmatic advocates” as they struggle with the upstream repercussions of the rule highlighted by members of the community in Environmental Defense Fund v. EPA.
The New Administration and Beyond
The selection of quality scientific studies for informing regulations has been plaguing EPA for decades now. This narrowing and broadening and reinterpretation of “best available science” does a disservice to the purpose of the agency, scientific integrity, and, ultimately, the American people. Those who would benefit most from “Strengthening Transparency in Regulatory Science” would be those who want to be regulated the least. For all its emphasis on achieving transparency, the final rule is quite duplicitous in claiming a benevolent goal—wanting to improve the general public’s ability to comment on regulations and address issues in the scientific community—while on the other hand constraining the use of updated, relevant scientific data. It does not employ any of the open science or collaboration suggestions generously provided by many scientific journals, nor does it explicitly provide solutions to ameliorate the privacy concerns echoed in thousands of comments. It is fascinatingly hypocritical and ironically ambiguous.
On January 20, 2021, President Biden released an executive order reaffirming the administration’s position on science-based regulations and the prioritization of environmental and public health. The new rule will undergo review in order to consider “suspending, revising, or rescinding the agency action” as soon as possible. But this puts the verification of quality scientific studies in regulatory policies back where it started––as a pawn subject to political whims. Biden’s team could provide a solution through constructing its own regulation so as to “[overwrite] the Trump rule” and “formally wipe [it] out… for future administrations.” The Congressional Review Act (CRA) could also serve as a more longstanding alternative. A successful resolution under the CRA could eliminate the opportunity for any future president to limit EPA’s use of science in its regulations.
 Augusta Wilson, Big Tobacco’s Smoke and Mirrors Revived by Pruitt’s Science Transparency Policy, The Hill (June 4, 2018, 5:00 PM), https://thehill.com/opinion/energy-environment/390638-big-tobaccos-smoke-and-mirrors-revived-by-pruitts-science?rl=1.
 Strengthening Transparency in Regulatory Science, 83 Fed. Reg. 18,768 (Apr. 30, 2018).
 Strengthening Transparency in Pivotal Science Underlying Significant Regulatory Actions and Influential Scientific Information, 86 Fed. Reg. 469 (Jan. 6, 2021).
 Envtl. Defense Fund v. EPA, No. 4:21-cv-00003, 2021 WL 270246 at *7 (D. Mont. Jan. 27, 2021).
 Stephen Lee, EPA’s ‘Secret Science’ Rule Taken to Court by States, Cities, Bloomberg Law (Jan. 19, 2021, 3:38 PM), https://news.bloomberglaw.com/environment-and-energy/epas-secret-science-rule-taken-to-court-by-states-cities.
 Shouse et al., Cong. Rsch. Serv., RL30853, Clean Air Act: A Summary of the Act and Its Major Requirements (2020).
 The Clean Air Act: Solving Air Pollution Problems with Science and Technology, Env’t Prot. Agency https://www.epa.gov/clean-air-act-overview/clean-air-act-solving-air-pollution-problems-science-and-technology (last visited Apr. 12, 2021).
 Off. Mgmt. & Budget, Memorandum on Issuance of OMB’s “Final Information Quality Bulletin for Peer Review” M-05-03, (Dec. 16, 2004).
 Id. at 26.
 Exec. Off. of the President, Office of the Press Sec’y, Memorandum on Scientific Integrity (Mar. 9, 2009).
 Exec. Order No. 13,563, 76 Fed. Reg. 3,821 at 3,822 (Jan. 21, 2011).
 Exec. Off. of the President, Office of Sci. & Tech. Policy, Memorandum on Increasing Access to the Results of Federally Funded Scientific Research (Feb. 22, 2013).
 Id. (Hindsight bias refers to one’s tendency “to see an event as having been predictable only after it has occurred” while confirmation bias references the tendency to direct one’s attention only to the evidence that supports one’s “expectations or favored explanation.”).
 Min Zhu, et al., Dimensionality Reduction in Complex Medical Data: Improved Self-Adaptive Niche Genetic Algorithm, Computational and Mathematical Methods in Medicine, Nov. 16, 2015 at 1, https://doi.org/10.1155/2015/794586.
 Christine Choirat, et al., Data Science in Environmental Health Research, 6 Current Epidemiology Reports 291 (2019), https://doi.org/10.1007/s40471-019-00205-5.
 Wilson, supra note 1.
 Email from Christopher C. Horner, Counsel, Bracewell & Patterson LLP to Timothy N. Hyde, Senior Dir. of Pub. Issues, R.J. Reynolds Tobacco Co. and Randy Tompson, Dir. of Pub. Issues, R.J. Reynolds Tobacco Co. (Dec. 23, 1996, 1:56:01 PM) (proposing a strategy to counteract EPA’s re-examination of environmental tobacco smoke), https://www.industrydocuments.ucsf.edu/tobacco/docs/#id=jfww0019.
 Wilson, supra note 1.
 Geoffrey Kabat, What Is Really at Stake in the House Committee on Science, Space, and Technology Subpoena of EPA Data, Forbes (Sep. 23, 2013, 8:00 AM), https://www.forbes.com/sites/geoffreykabat/2013/09/23/what-is-really-at-stake-in-the-republican-partys-subpoena-of-epa-data/?sh=16cd9c861ebc.
 Kelly Servick, House Panel Subpoenas EPA for Air Pollution Data, Sci. (Aug. 2, 2013, 6:15 PM), https://www.sciencemag.org/news/2013/08/house-panel-subpoenas-epa-air-pollution-data.
 Douglas W. Dockery et al., An Association Between Air Pollution and Mortality in Six U.S. Cities, 329 New Eng. J. Med 1753 (1993).
 Christopher Rowland, House GOP Demands Harvard Study Data, Boston Globe (Sept. 6, 2013, 8:57 PM), https://www.bostonglobe.com/news/nation/2013/09/06/landmark-harvard-study-health-effects-air-pollution-target-house-gop-subpoena/2K0jhfbJsZcfXqcQHc4jzL/story.html.
 Servick, supra note 29.
 Press Release, Comm. on Sci., Space, & Tech., Smith Subpoenas EPA’s Secret Science (Aug. 1, 2013), https://republicans-science.house.gov/news/press-releases/smith-subpoenas-epa-s-secret-science.
 Rowland, supra note 32.
 H.R. 1030, 114th Cong. (2015).
 Strengthening Transparency in Regulatory Science, 83 Fed. Reg. 18,768 at 18,770.
 Id. at 18,772.
 Id. at 18,774.
 Id. at 18,770.
 Strengthening Transparency in Regulatory Science, 83 Fed. Reg. 15,396 at 15,399.
 Id. at 15,403.
 Id. at 15,397. The significance of this will be discussed in the following sections.
 Strengthening Transparency in Pivotal Science, 86 Fed. Reg. 469, at 475.
 Id.; Memorandum from The Office of the Press on Scientific Integrity, supra note 15.
 Juliet Eilperin & Brady Dennis, EPA Finalizes Rule to Limit Science Behind Public Health Safeguards, Wash. Post (Jan. 5, 2021, 10:51 am), https://www.washingtonpost.com/climate-environment/2021/01/04/epa-scientific-transparency/.
 Strengthening Transparency in Pivotal Science, 86 Fed. Reg. 469, at 483.
 Models are of great importance in environmental health studies. See Marie-Abèle Bind, Causal Modeling in Environmental Health, 40 Annual Rev. of Pub. Health 23 (Jan. 11, 2019), https://doi.org/10.1146/annurev-publhealth-040218-044048.
 Strengthening Transparency in Pivotal Science, 86 Fed. Reg. 469.
 Id. at 472.
 5 U.S.C. § 301 (2011).
 N.Y. Attorneys General et al., Comment Letter on Proposed Rule Strengthening Transparency in Regulatory Science, Supplemental Rule (Aug. 16, 2020) (citing Global Van Lines, Inc. v. Interstate Commerce Comm’n, 714 F.2d 1290, 1293-97 (5th Cir. 1983)), https://www.law.nyu.edu/sites/default/files/multistate_comments.pdf.
 5 U.S.C. § 101 (2011) (listing all agencies considered as Executive or military departments).
 83 Fed. Reg. 15,396 at 15,397.
 Id at 15,398.
 James Goodwin, The EPA’s ‘Censored Science’ Rule Isn’t Just Bad Policy, It’s Also Illegal, Union of Concerned Scientists (Nov. 22, 2019), https://blog.ucsusa.org/guest-commentary/the-epas-censored-science-rule-isnt-just-bad-policy-its-also-illegal.
 83 Fed. Reg. 15,396 at 15,397.
 435 U.S. 519, 541 (1978).
 381 U.S. 279, 294 (1965).
 Env’t. Defense Fund v. EPA, No. 4:21-cv-00003, 2021 WL 270246 (D. Mont. Jan. 27, 2021).
 Id. at *20.
 Strengthening Transparency in Pivotal Science, 86 Fed. Reg. 469, at 472.
 John. P.A. Ioannidis, All Science Should Inform Policy and Regulation, Pub. Library of Sci. (May 13, 2018), https://journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.1002576.
 Union of Concerned Scientists, Comment Letter on Proposed Rule Strengthening Transparency in Regulatory Science, Supplemental Rule (May 15, 2020), https://ucs-documents.s3.amazonaws.com/science-and-democracy/ucs-csd-strengthening-transparency-in-regulatory-science-suppl-comment.pdf.
 Corin Hiar, In Battle over Pesticide Ban, Trump’s EPA Aims to Undermine the Science, Sci. (Aug. 23, 2018, 2:15P PM), https://www.sciencemag.org/news/2018/08/battle-over-pesticide-ban-trump-s-epa-aims-undermine-science.
 Chlorpyrifos: EPA’s Request for Columbia’s Raw Data, EPA, (last accessed Dec. 9, 2020) (emphasis added), https://www.epa.gov/ingredients-used-pesticide-products/chlorpyrifos-epas-request-columbias-raw-data.
 Geena Reed, We Told OMB Why EPA’s So-Called Transparency Rule Is a Trap, Union of Concerned Scientists (Oct. 2, 2020), https://blog.ucsusa.org/genna-reed/we-told-omb-why-epas-so-called-transparency-rule-is-a-trap.
 Lisa Friedman, E.P.A Rejects Its Own Findings That a Pesticide Harms Children’s Brains, N.Y. Times (Sept. 23, 2020), https://www.nytimes.com/2020/09/23/climate/epa-pesticide-chlorpyrifos-children.html.
 Sharon Lerner, Republicans Are Using Big Tobacco’s Secret Science Playbook to Gut Health Rules, The Intercept (Feb. 5, 2017, 9:03 AM), https://theintercept.com/2017/02/05/republicans-want-to-make-the-epa-great-again-by-gutting-health-regulations/.
 83 Fed. Reg. 18,768 at 18,769.
 Marianne Lavelle, EPA’s ‘Secret Science’ Rule Meets with an Outpouring of Protest on Last Day for Public Comment, Inside Climate News (May 19, 2020), https://insideclimatenews.org/news/19052020/secret-science-epa-transparency-rule-wheeler-pollution-environment/.
 Douglas W. Dockery & C. Arden Pope, The Threat to Air Pollution Health Studies Behind the Environmental Protection Agency’s Cloak of Science Transparency, 110 Am. J. Pub. Health 286 (Mar. 2020), https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7002955/.
 Lavelle, supra note 82.
 Reed, supra note 78.
 Scott Waldman, Scientists Favor Transparency, but Say EPA Plan Will Limit It, Sci. Am. (Apr. 25, 2018), https://www.scientificamerican.com/article/scientists-favor-transparency-but-say-epa-plan-will-limit-it/.
 Scott Waldman, EPA’s Own Advisory Board Questions ‘Secret Science’ Plan, Sci. Am. (May 16, 2018), https://www.scientificamerican.com/article/epas-own-advisory-board-questions-secret-science-plan/.
 Juliet Eilperin & Brady Dennis, Pruitt Unveils Controversial ‘Transparency’ Rule Limiting What Research EPA Can Use, Wash. Post (Apr. 24, 2018), https://www.washingtonpost.com/news/energy-environment/wp/2018/04/24/pruitt-to-unveil-controversial-transparency-rule-limiting-what-research-epa-can-use/.
 83 Fed. Reg. 15,396 at 15,399.
 Ioannidis, supra note 71; Lee, supra note 5.
 Exec. Order No. 13,990, 86 Fed. Reg. 7,037 (Jan. 25, 2021).
 Stephen Lee & Jennifer Hijazi, Biden Team in a Bind over Reversing EPA’s ‘Secret Science’ Rule, Bloomberg Law (Jan. 6, 2021, 2:57 PM), https://news.bloomberglaw.com/environment-and-energy/biden-team-in-a-bind-over-reversing-epas-secret-science-rule.