05 August 2009

Bipartisan Panel of Industry, NGO and Academic Experts Proposes New Rules for Science in Federal Regulation

I was a participant on the study team that put together the report discussed in the following press release. The full report can be found here. Comments welcomed.
Wednesday, August 5, 2009

Wednesday, August 5, 2009
Contact: Paul Bledsoe, (202) 204-2403
Additional Contact: Eileen McMenamin, (202) 379-1633

Bipartisan Panel of Industry, NGO and Academic Experts Proposes New Rules for Science in Federal Regulation

Goal to Distinguish Science Questions from Policy Debates

Agencies Should Empanel Scientific Advisory Committees Solely of Members with Relevant Scientific Expertise; Improve Disclosure and Transparency

Washington, D.C. — A bipartisan panel of top scientific and regulatory experts today released recommendations calling on the White House and federal agencies to make specific changes in the regulatory process to clearly distinguish scientific questions from policy disputes. The report comes from the Science for Policy Project (SPP). A project of the Bipartisan Policy Center, SPP is co-chaired by former Rep. Sherwood Boehlert (R-NY), past chair of the House Science Committee, and Donald Kennedy, former editor of Science; and eleven other ideologically diverse members from business, academia, government and non-profits. The complete list of panel members appears below.

"The fundamental theme of the report is that the Administration needs to put in place procedures to try to distinguish science questions from policy questions," said Boehlert. "Often, policy disputes are cast as fights over science. This damages the credibility of science and obscures the real issues that ought to be debated. For example, how much risk a substance poses to human health or the environment is a science question; how much risk is acceptable is a policy question."

"Our recommendations would make the regulatory process more rigorous and transparent," said Kennedy, who headed the Food and Drug Administration (FDA) under President Jimmy Carter. "That should lead to better, more thoughtful regulations and greater consensus when the government is considering whether or not to regulate. And perhaps most important, when there are disagreements, officials and the public will have a clearer sense of what they are about."

The report recommends requiring new information when regulations are proposed by agencies such as the Environmental Protection Agency and the FDA, and enhancing the credibility of federal advisory committees to ensure the integrity of science in regulatory policymaking. Many of these recommendations are relevant to the White House and federal agency effort to implement President Obama's March 9, 2009 Presidential Memorandum on Scientific Integrity. An interim version of the SPP report was released in March, and the White House has reviewed that report as part of its work to issue guidelines on scientific integrity and regulatory reform, which are expected soon.

The report's recommendations include:
  • Clarity regarding key science questions needed to write specific regulations. Federal Register notices for proposed regulations should make clear what science questions and what policy questions needed to be answered to formulate the regulation and what science was most influential in drafting the regulation. The notices might also make clear what additional science would help resolve remaining questions and might offer policy alternatives that are consistent with the science.
  • Greater focus on science in advisory committees. Federal agencies should empanel scientific advisory committees – committees that are composed solely of members with relevant scientific expertise – to address science questions relevant to policy, and those panels should not make policy recommendations. Members of such committees should be Special Government Employees, a category that makes them subject to conflict of interest and other ethics rules.
  • Greater transparency in committee appointment process. The process for appointing committees should be more transparent, with agencies taking steps including seeking names through the Web and asking for comments on proposed individuals on the Web.
  • More disclosure by committee members. Members of scientific advisory committees should be required to disclose to the government and to the public far more information on their backgrounds than is currently the case so the government and the public can evaluate their qualifications and determine whether their service raises any concern about conflicts of interest or bias.
  • Clearer conflict of interest rules. The government should set clear rules about what constitutes a conflict of interest for a member of a scientific advisory committee. The government should distinguish clearly between conflict of interest and bias. Agencies should try not to appoint advisory committee members with conflicts of interest, but if such individuals are appointed, the appointment should be publicly announced and justified. The standard for appointing members with conflicts should be changed to the one used by the National Academy of Sciences. Chairs or co-chairs of committees should never have conflicts.
  • Greater transparency in committee selection might allow a limited number of closed committee meetings. If procedures are put in place to make the selection of advisory committee members more transparent, then the government could consider allowing scientific advisory committees to have a limited number of closed meetings under specific circumstances.
  • Transparency in the use of scientific literature. The process agencies and scientific advisory committees use to review the scientific literature should become more transparent and thorough. They should be explicit about the criteria they are using to determine which scientific papers to review and how those papers are being evaluated, and should pay more attention to the methodology used in cited work. In general, papers that have not been peer reviewed should be treated with skepticism, but they should not be automatically excluded.
  • Legitimate use of Confidential Business Information. The Confidential Business Information designation, which limits public access to information, is legitimate but appears to be overused. Agencies should consider ways to ensure that the category is being used legitimately, such as requiring businesses to provide a brief but substantive justification for requesting protection.
  • Greater participation in, and improved quality of peer review. Federal agencies, universities and scientific journals need to experiment with ways to encourage more scientists to act as peer reviewers and to experiment with different peer review procedures to see what will improve the quality of reviews.
The report makes detailed recommendations about the formation and use of scientific advisory committees, including setting requirements for disclosure and determining and dealing with conflicts of interest and bias.

The study also focuses on the evaluation of research studies by agencies and scientific advisory committees, and criteria such as peer review, conflicts of interest, sponsor control of research and publication, data availability, and risk and uncertainty.

"The overall purpose of our report," said Kennedy, "is to inaugurate an era in which the science behind regulatory proposals will emerge from a more transparent and credible process. Disagreements over political ideology, economics and values that are at the heart of many regulatory disputes will be debated more openly and fully if these reforms are implemented. Science will be better protected from unwarranted attacks and political values will be more fully debated. Better regulatory policy will protect the public from both needless regulations and from needless dangers."

Boehlert added, "The very fact that this ideologically diverse group was able to reach consensus on specific recommendations for reform shows that the time is ripe for these changes."

The Science for Policy Project is funded by the David and Lucile Packard Foundation, the William and Flora Hewlett Foundation, and ExxonMobil Foundation. The project was directed by David Goldston, the former chief of staff of the House Science Committee. Copies of the full report are available at www.bipartisanpolicy.org.

Members of the Bipartisan Policy Center's Science for Policy Project:*

HON. SHERWOOD BOEHLERT (co-chair): The Accord Group
DONALD KENNEDY (co-chair): Stanford University
ARTHUR CAPLAN: University of Pennsylvania
LINDA J. FISHER: E. I. du Pont de Nemours and Company
LYNN R. GOLDMAN: Johns Hopkins University
JOHN D. GRAHAM: Indiana University
DANIEL GREENBAUM: Health Effects Institute
MICHAEL P. HOLSAPPLE: ILSI Health and Environmental Sciences Institute
KEVIN KNOBLOCH: Union of Concerned Scientists
KENNETH OLDEN: City University of New York
ROGER A. PIELKE, JR.: University of Colorado
SHERRI K. STUEWER: Exxon Mobil Corporation
WENDY E. WAGNER: University of Texas

*panelist affiliations listed for identification purposes only.


  1. Free the science! Forget regulating it, how about some Laissez Faire? Then policy disputes would be a non-issue (and we all know how well central planning worked out for Soviet Genetics).

  2. Peer-review is mentionned, but they need to go one step further: before going into policy, the science must be replicated. That also means supporting better access to the data, algorithms and other relevant info that was used in a study. To often, researchers impede to process of replicability (I have the Mann story in mind). Defining what is science is also important (not all science is built equal). Does a model whose validity has not been proven constitute science? (I have climate models in mind here). Once this is done, they can supply the policy mill.

  3. Yes, bigcitylib, and two other Foundations.

    Do you have any other questions? That one was easy.

  4. Because if Exxon is behind it then its likely to be a bunch of bureaucratic thumb-twiddling rules designed to slow down the regulatory process so as to benefit Exxon which, from taking a quick glance at it, it appears to be.

  5. bigcitylib, a most excellent comment on the content of the report.

    I'm surprised that you failed to mention that not only does ExxonMobile Foundation fund the Science for Policy Project, the members of the Project itself includes someone associated with ExxonMobile and (horrors) E. I. du Pont de Nemours and Company, seven Universities, two Institutes, and two other organizations.

  6. I agree with much that is in the paper, and I am particularly in agreement with the need for more transparency..sometimes NAS reports seem more like an oracles (go to the Temple of Science and ask the oracle) and less like a linear construct between facts found or data reviewed and conclusions drawn.

    I did have a question as to what is a natural scientist for the purpose of scientific panels..it sounds as if economists and social scientists are out?
    My only other suggestions would be for some policy science research. If the output of scientific panels is supposed to be "science" then I would argue that they should be replicable. I would suggest having at least three panels composed of scientists randomly selected from a pool of disciplines and points of view (stratified so that the key disciplines and points of view are present in each). Then their findings could be compared and the common ones among panels determined to be the "scientific point of view for future reference."
    We could try this for some policy that is really important, economically and very controversial.. say climate policy?

    I have a hypothesis that both peer review panels and science for policy panels results are strongly (but not totally) a function of interpersonal dynamics and small p politics. It would be interesting to me to test this hypothesis by controlling for that variable.

    Bigcitylib, Ah, but slowing down the regulatory process is a two edged sword as we found vis a vis the Data Quality Act. It leads to slowing down in the imposition of new (good) regulations that some people may like, but also slowing down in the removal of old (bad) regulations which the same people may dislike.
    One person's "bureaucratic thumbtwiddling" is another person's NEPA process or public involvement.