Restoring Respect For Research

President Barack Obama first issued the call to “restore science to its rightful place” in his January 2009 inaugural address, but actually establishing policies to help separate science from politics at 21 federal agencies has taken more than three years.

The final versions of agencies’ new scientific integrity policies were due to be posted on a public website by March 30. The White House Office of Science & Technology Policy (OSTP) announced in a blog post earlier this month that all but three—Departments of Defense, Homeland Security, and Labor—met that deadline. The last three plans are expected by the end of April.

“This process has been time-consuming, but, I would argue, exceedingly important,” OSTP Director John P. Holdren wrote on the office’s blog earlier this month. “Through it all, the prime importance of scientific integrity—the need to ensure that Americans can trust the results of federally supported science—has been elevated and made explicit in numerous ways.”

The new policies are being created at a busy time for those in the scientific integrity community. Rules regulating everything from federal advisory committees to professional ethics for federal employees to research on human subjects are being revised. And the National Academy of Sciences (NAS) is revisiting a landmark 1992 report on scientific misconduct, because the research environment has changed drastically in the past 20 years.

The goal of the scientific integrity policies is to foster the public’s trust that federal research—and the resulting publications and regulations—is not being influenced by politics or special interests. Obama’s charge was spurred in part by the George W. Bush Administration’s perceived slights of science.

After mentioning scientific integrity in his inaugural speech, the President issued a memo to all federal agencies in March 2009 outlining his suggested reforms and directing Holdren to recommend a plan to accomplish those reforms within 120 days.

“The days of science taking a backseat to ideology are over,” Obama said in an April 2009 speech to NAS. He called on Holdren to “ensure that federal policies are based on the best and most unbiased scientific information.”

The 120-day mark came and went without any public acknowledgment from OSTP. As months kept ticking past, interested parties in the research community wondered what had happened.

“We were very excited. But why did it take so long?” Mark S. Frankel recalls asking. Frankel is director of the Scientific Responsibility, Human Rights & Law Program at the American Association for the Advancement of Science (AAAS). “Nothing much was happening for almost two years.”

Inside OSTP, the task of designing a policy turned out to be more difficult than the President’s advisers had anticipated. The office had created an interagency panel to hash out a preliminary scientific integrity policy and to review public comments about what should be in such a policy.

“Determining how to elaborate on the principles set forth in the [President’s] memorandum in enough detail to be of real assistance in their implementation, while at the same time retaining sufficient generality to be applicable across executive departments and agencies with a wide variety of missions and structures, has been particularly challenging,” Holdren wrote in a June 2010 blog post. He added that OSTP hoped to have a policy out within a few weeks.

But the next word from OSTP came in December, when Holdren put out a four-page memo outlining what federal scientific integrity policies need to do. Specifically, he tasked each agency to include four elements in their policy.

The first element should be aimed at improving the use of scientific information in government policy-making. The second requires clear policies for when and how federal scientists can speak to the media. The third governs how advisory committee members are recruited and how their advice is conveyed. And the final element dictates that plans include ways for government scientists to participate in the larger scientific community by, for example, publishing papers or serving on journal editorial boards.

Each agency was left to decide for itself how to go about creating the policy. For most agencies, this process involved revising existing science integrity policies.

For example, the Environmental Protection Agency already had policies in place that addressed specific aspects of Holdren’s memo. But this process forced EPA to evaluate its policies all together, says Mark Miller, a senior scientist who helped coordinate the EPA committee that evaluated the agency’s policies.

The reevaluation made it clear that the new policy needs to apply not only to the agency’s 7,000 scientists but also to managers, grantees, and contractors. In the process, EPA created “a culture of scientific integrity,” Miller says. “There is a better understanding of what is expected.”

EPA also made a bigger financial commitment to scientific integrity, assigning staff specifically to work on scientific integrity and setting up a permanent committee, says Robert Kavlock, the acting deputy assistant administrator for science. Because it is a regulatory agency, any change in the public’s trust impacts EPA directly. “Doing science well and doing it right is certainly of huge importance to the agency,” he explains. “If we don’t do credible work, we can’t protect human health.”

Nonregulatory agencies like the National Science Foundation don’t face the same political pressures. But NSF’s funded scientists do work that could indirectly impact government regulations, so it also took the charge seriously, says Fae L. Korsmo, who led the scientific integrity changes at NSF, which is strictly a grant-making institution.

NSF created a broad committee to ask, “Are we doing what we need to be doing to be transparent, ethical, and act with integrity?” Korsmo explains. After its evaluation, NSF made changes in how it presents information about its advisory committee members to the public. It also tells scientists what to do, should they wish to speak publicly about controversial topics against the recommendation of NSF’s media office.

Holdren’s memo encouraged all agencies to solicit public comments, but only a few did. Those agencies that did ask for public comment include the National Aeronautics & Space Administration, the National Oceanic & Atmospheric Administration, EPA, NSF, and Departments of Interior and Education. Other agencies simply posted policies quietly on their websites.

The resulting policies are a mixed bag, says Francesca T. Grifo, who heads the Scientific Integrity Program at the Union of Concerned Scientists, a nonprofit science advocacy group. Some are dozens of pages long and create new mechanisms to encourage and enforce scientific integrity at the agency. Other policies, like those from Departments of Agriculture and Energy, are in the form of a memo from an agency official confirming the scientific integrity principles. Additionally, some agencies have a mechanism in place to revise the policies when new situations arise or problems come to light.

“If they are implemented with good intent, they should make a difference in how policy is made in the executive branch,” AAAS’s Frankel says. But that won’t change everything. “One should never forget that politics and science go hand in hand in government, and science is not always the decisive factor.”

For Grifo, the policies themselves are not as important as the dialogue that went on to create them. “It is largely about transparency,” she explains. If the institutional culture doesn’t support the policy, she adds, “that is an untenable situation for a scientist to be in.”

Even a few months ago, Grifo was still hearing in her surveys of federal employees that scientists feel uncertain about where they stand. “Good science doesn’t happen when people are keeping their heads down,” she says. “You need a raucous, rowdy discussion.”

An important result of this exercise, according to Grifo, would be if these new policies make federal scientists aware of how they should handle political interference in their research. “We are not naive enough to think that we will ever eliminate these things entirely.”