Restrictions on the Right to Publish at the EPA

Published Jul 22, 2009

Though the free exchange of ideas is a pillar of the scientific enterprise, excessive and protracted clearance policies for the publication of scientific work at the Environmental Protection Agency (EPA) have both undermined the quality of agency science and harmed the EPA's scientific reputation.

In the early 1990s, the EPA began to assess its requirements for internal and external peer review of its scientific work.1,2 This effort culminated in the agency's 1998 Peer Review Handbook,3 an attempt to ensure quality control over the EPA's products. However, this handbook does not provide much guidance on how EPA scientists can get clearance to publish in scientific journals.  The following cases demonstrate how some divisions have instituted policies that unduly restrict the expression of scientific opinion:

  • In a law review journal, Todd Stedeford, a former EPA scientist, described the clearance process at the EPA's National Center for Environmental Assessment (NCEA) as "a policy of prior restraint that has escalated to censorship on publishing." He notes that the policy "appears to violate the First Amendment rights of government scientists who wish to contribute articles written outside of their official duty hours."4

    According to Stedeford, the NCEA's clearance process is far more extensive than that outlined in the Peer Review Handbook. A scientific article with "policy implications" can require up to four levels of internal review before the author can submit it to a scientific journal, which would then subject it to its own peer-review process. Manuscripts submitted from the EPA's Integrated Risk Information System (IRIS) require up to seven levels of internal review.5 In Stedeford's experience, the agency may hold up a manuscript at any stage based on the "prejudices and whims of each reviewing official," and he reports that there is no consistency between various stages of review:

    "The author has first-hand experience of the inefficiency and disorganization of this system. A manuscript that he co-authored and submitted for clearance in February of 2006 (intended for a peer-reviewed scientific journal) was still in clearance as of July 19, 2006…Further, the corresponding author of that manuscript, Dr. Ching-Hung Hsu, has since left the Agency because of the draconian restrictions placed on publishing."6

    More disturbingly, Stedeford says he was required to submit articles written on personal time for EPA clearance.7 Dr. Stedeford left the EPA shortly after the publication of the law review article.
  • In an interview with UCS, another EPA scientist (who asked to remain anonymous) also described problems with getting clearance to publish in scientific journals.8
  • In another interview, Ami Zota, a non-EPA research scientist working on environmental health, related her experience collaborating with EPA Office of Research and Development (ORD) scientists on a research project and a subsequent paper. In early 2008, Zota's EPA coauthors sent the paper through the agency's internal peer review process, which Zota said was timely, and the EPA reviewer provided helpful technical comments. However, EPA policies also required the signature of the coauthors' supervisor before clearance could be granted, and at that stage the review became non-technical and "political" in nature.

    The paper included standard disclaimers that the findings represented the opinion of the authors and not EPA policy. The supervisor nevertheless told Zota that a handful of sentences in the discussion section were "probable red flags," and that the authors would have to send the manuscript to EPA headquarters, which could delay publication by several months. The supervisor suggested that Zota delete all references to children and children's health—topics that were not the paper's primary focus, but that follow-up investigations would address.

    Zota accepted some of the supervisor's suggestions, but she felt it was her scientific prerogative to frame those topics and future research in the way she saw fit. After a few conference calls, the supervisor granted clearance without further upstream review. However, Zota called the experience "intimidating," and noted that "I now understand how nuanced efforts to censor science might occur at the EPA."9

Internal peer review of scientific publications can greatly improve the quality of EPA science. Indeed, because government agencies have a responsibility to provide high-quality scientific information to the public, we can expect a greater level of review at the EPA than in academia or the private sector. The EPA also has a legitimate interest in speaking with one voice on official policy. However, political review of scientific publications for the merest hint of a policy implication can cross the line into censorship of legitimate scientific opinion.

Note: This page is an excerpt from the 2008 UCS report Interference at the EPA.

1. National Research Council. 2000. Strengthening science at the U.S. Environmental Protection Agency: Research management and peer review practices. National Academies Press: Washington, DC.
2. Expert Panel on the Role of Science at EPA. 1992. Safeguarding the future: Credible science, credible decisions. Washington, DC: U.S. Environmental Protection Agency.
3. Science Policy Council (SPC). 2006. U.S. Environmental Protection Agency peer review handbook, 3rd edition.
4. Stedeford, T. 2007. "Prior Restraint and censorship: Acknowledged occupational hazards for government scientists." William & Mary School of Law Environmental Law and Policy Review 31(3): 725-745.
5. National Center for Environmental Assessment (NCEA). 2004. Memorandum from the director of the National Center for Environmental Assessment to NCEA staff. Clearance procedures for NCEA work products. December 4. Quoted in Stedeford 2007.
6. Stedeford 2007.
7. Ibid.
8. Anonymous EPA scientist. 2008. UCS interview, January.
9. Zota, A. 2008. Phone interview with UCS, February 26. Ami Zota is an environmental health scientist. sic