Four years ago tomorrow, President Obama signed a memorandum for the heads of executive departments and agencies on scientific integrity. He asked the Director of the Office of Science and Technology Policy (OSTP) to confer with them – specifically calling out the Office of Management and Budget – and recommend a plan to achieve the highest level of integrity.
And for good reason. The United States has enjoyed prosperity and health in large part because of its strong and sustained commitment to independent science. Interference in science threatens our nation’s ability to respond to complex challenges to public health, the environment and national security and raises the possibility of lasting harm to the federal scientific enterprise.
Dr. John Holdren later released guidance for the agencies as they set about writing their policies and plans for implementing the president’s principles.
Dr. Holdren’s guidance was a welcome first step for addressing the endemic problems UCS has tracked at federal science agencies. We have documented and described examples of falsifying data and fabricating results, selective editing of reports, intimidating and coercing scientists, censoring and suppressing scientists, hiding, suppressing and delaying the release of scientific findings, disregarding legally mandated science, allowing conflicts of interest, and the corrupting of scientific advisory panels.
In addition, deeper changes in the structure and policies of the executive branch can also threaten scientific integrity. These include, for example, centralized decision making, homogenized agency decision making, reduced transparency, unnecessary bureaucracy, retaliation against whistleblowers, allowing officials to move too easily between industry and government posts, removing science from decision making, and weakening enforcement and monitoring.
To protect the integrity of science, the Holdren memo outlined four areas for policy development:
- foundations of scientific integrity,
- public communications;
- use of federal advisory committees; and
- the professional development of government scientists and engineers.
The memo did not specifically mention many important details including reforming the regulatory process, protecting monitoring programs from targeted elimination and the enforcement of current science-based laws.
In the ensuing years, agencies have found different ways to comply with Dr. Holdren’s guidance. More than 20 agencies, offices and bureaus that deal with scientific issues have created their versions of scientific integrity policies. They vary considerably in breadth and quality.
Today, we are releasing our analysis of the strengths and weaknesses of these policies. Our bar is higher and more comprehensive than the one set by OSTP. Simply repeating the words of the Holdren memo, which some agencies opted to do, will not lead to the changes needed to protect and promote scientific integrity. We looked for evidence that an agency considered the policies it had in place and developed specific plans to fill gaps that left threats to scientific integrity unaddressed.
Twenty-two agencies submitted policies or memos to OSTP. The results:
Six agencies submitted policies that actively promote and support a culture of scientific integrity. These include the Centers for Disease Control, the Department of the Interior, the Environmental Protection Agency, NASA, the National Oceanic and Atmospheric Administration, and the National Science Foundation.
Five agencies submitted policies that also promote and support scientific integrity but need additional work to ensure meaningful long-term progress. These include the Department of Homeland Security, the Department of Commerce, the Department of State, the Food and Drug Administration, and the Marine Mammal Commission.
Eleven agencies submitted policies that do not make adequate commitments to achieve the preservation and promotion of scientific integrity. These include the Department of Defense, Department of Education, Department of Energy, Department of Health and Human Services, Department of Justice, Department of Labor, Department of Transportation, Department of Veterans Affairs, the Office of the Director of National Intelligence, US Agency for International Development, and the US Department of Agriculture.
The good news is that all 22 policies repeat some important parts of the Holdren guidance – for example, all note that officials should not suppress or alter scientific or technological findings and most pledge to:
- make Federal Advisory Committee conflict of interest waivers public;
- refrain from editing the products of advisory committees;
- select candidates for scientific positions based primarily on their scientific credentials;
- ensure that the research used in policy decisions undergoes peer review; and
- promote and facilitate the professional development of scientists such as allowing government scientists and engineers to become editors or serve on editorial boards of professional or scholarly journals.
But many policies have significant flaws:
- Seven agencies require permission from supervisors or public affairs before scientists can discuss their research results;
- Nine agencies limit the first amendment rights of their scientists by not allowing them to speak about their research as private citizens;
- Seven agencies do not have final policies available on their websites or have policies posted that have expired;
- Fourteen agencies have policies that do not apply to contractors and other non-career employees;
- Not one agency has updated its policy to reflect the Whistleblower Protections and Enhancements Act of 2012;
- Not one agency is posting a record of who meets with agency officials and only 2 regularly post the calendars of their top officials;
- Not one agency agreed to separate scientific and technological information from draft regulations or other pre-decisional documents so that the science considered in decision-making can be released and protected from political influence by the White House or other agencies; and
- Not one agency chose to better define what constitutes a conflict of interest. A meaningful definition would address who is covered under conflict of interest rules (just the individual, or also his/her spouse and children?), how far back disclosure should go (one year? five years?), or what constitutes a conflict (a cup of coffee? $50,000?) as was suggested in a recent report by the Research Integrity Roundtable of the Keystone Center.
Some agencies that both undertake and use a great deal of scientific analyses did not submit policies. Examples include the Social Security Administration, the Department of Housing and Urban Development, and the Department of the Treasury. While commissions are independent, the Marine Mammal Commission still submitted a policy. I wish the Consumer Product Safety Commission, the Nuclear Regulatory Commission, Securities and Exchange Commission, Federal Elections Commission, National Transportation Safety Board and other commissions had done the same.
I commend the changes that have been achieved, but in no way are we done. A future administration, especially one willing to attack science, could still wreak havoc on our scientific enterprise with disastrous consequences for our health and safety.
I want to pick up produce in my local market knowing that our food inspectors are using independent science to declare it safe from contamination. I want to take the medicines my doctor prescribes feeling sure that if a reviewer at the Food and Drug Administration feels industry pressure to push a drug along despite faulty clinical trials, that there was a place to report that and consequences for their supervisor. I want air quality standards to be based on science with a margin for safety – just as the clean air act requires so that all our families can breathe easier. I want to know that if a reporter calls the Environmental Protection Agency to talk to a scientist, that they do not miss their deadline while the public affairs bureaucracy decides to whom they may speak. And good leadership at the agencies, good scientific integrity policies, and full implementation of these policies can give us all that and more.
So let’s work to make sure that in the next four years we get this done and done right!