How Is the USDA Doing on Scientific Integrity?

In March 2013, the US Department of Agriculture updated its scientific integrity policy, a policy mandated by the Obama Administration for all federal agencies with a significant focus on science. Along with 22 other agencies and departments, the USDA developed a policy in 2011 that the Union of Concerned Scientists assessed to “not make adequate commitments to scientific integrity.” How does the revised policy measure up and does it appear to be working?

A starting point: the 2011 USDA scientific integrity policy

The 2011 version of the USDA scientific integrity policy had some shortcomings. Namely, the policy failed to address many of the guidelines put forth in the December 2010 scientific integrity directive from the White House Office of Science and Technology Policy.

It failed to include two key tenets of scientific free speech: It didn’t grant scientists the “right of last review” on materials publicly released in their name or that significantly relied on their work. And it didn’t provide a “personal views exception”, which would allow scientists to express their personal views on a topic (including views on policy), provided they make clear that they are not speaking for their agency. (For more on why scientists need personal views exceptions in government agency policies, see my previous blog here.)

Also of concern, the policy didn’t contain many specifics about who is responsible and how the stated policies would be implemented.  Importantly, the details of this policy didn’t matter much in any event, since the policy expired on August 5, 2012, one year after its issuance.

New provisions: the good and the not so good

The 2013 update to the USDA scientific integrity policy makes improvements. First off, it removes any time limitation for its relevance. It also adds significantly more detail about who is responsible for carrying out which aspects of the policy. The new policy creates a USDA Departmental Scientific Integrity Officer and requires each agency within the USDA to appoint an agency level scientific integrity officer. These are important efforts for creating the structure and mechanisms to carry out the principles of the policy.

The policy also adds a few new provisions, including one protecting whistleblowers. The policy states that the Department policy is to:

Protect those who uncover and report allegations of research misconduct or other violations of scientific integrity, as well as those accused of research misconduct or other violations of scientific integrity in the absence of a finding of misconduct, from prohibited personnel practices (as defined in 5 U.S.C. 2302(b));

It is great to see the USDA add this provision to protect not only those who report scientific integrity breaches but also those accused, if no misconduct is found. This is important as we know that scientific integrity cases are often complicated and some may be wrongly accused.

Other new provisions to the USDA policy were less positive. A new provision now states that the department policy is to (emphasis mine):

(2) Ensure that scientists may communicate their findings without political interference or inappropriate influence, while at the same time complying with USDA policies and procedures for planning and conducting scientific activities, reporting scientific findings, and reviewing and releasing scientific products. Such communications include research on policy-related issues when appropriate to the role of the agency and scientist; however, the scientists should refrain from making statements that could be construed as being judgments of or recommendations on USDA or any other federal government policy, either intentionally or inadvertently. Communications on such matters should remain within the bounds of their scientific findings. Such scientific and technical communications for non-USDA media (e.g., manuscripts and presentations for scientific journals, workshops, conferences, and symposia) should follow agency technical review procedures and do not generally require review above the agency level.

It is good to see the USDA explicitly prohibit “political interference and inappropriate influence” over scientists’ communication of their work, as we know these are often the source of scientific integrity problems.

The second bolded area raises questions about USDA scientists’ ability to express their personal views. The policy appears to prohibit scientists from speaking on anything outside of their scientific findings, regardless of whether they clarify if they are speaking in their personal capacity. Instead of including such restrictive language, the USDA could have allowed scientists to express personal views with a disclaimer that they are not speaking on behalf of their agency.

Indeed, other federal agencies have effectively dealt with this issue within their scientific integrity policies. The National Oceanic and Atmospheric Administration, for example, clearly articulates scientists’ right to express their personal views in its policy. The NOAA Scientific Integrity Policy states:

NOAA scientists are free to present viewpoints, for example about policy or management matters, that extend beyond their scientific findings to incorporate their expert or personal opinions, but in doing so they must make clear that they are presenting their individual opinions- not the views of the Department of Commerce or NOAA.

Policies don’t necessarily change practices

So how is this new USDA policy working now that it has been in effect for a few years? We know from past UCS work that policies don’t always match practices. Agencies can have great agency culture that protects scientists without relying on a policy. And on the other hand, agencies can have wonderful policies in place that simply aren’t implemented. The only way to know is to hear from scientists inside the agency.

In 2012, the Union of Concerned Scientists conducted a survey of federal scientists within its Science Network. Though not comprehensive in nature, the survey yielded some interesting anecdotes suggesting that USDA scientific integrity and communications policies may not be living up to their intentions.

“The policies are written so vaguely that the [USDA] can and does suppress papers that may be inconvenient for other government agencies and industry.” — Anonymous USDA scientist, 2012

“ ‘Loose lips sink ships’ appears to be management’s motivation.” — Anonymous USDA scientist on the agency’s social media policy, 2012

To be fair, this survey did not account for the 2013 updates to the USDA policy. What do we know about how the policy is being implemented currently? We can look to more recent cases for that.

Is the USDA muzzling scientists?

As reported in the Washington Post and elsewhere, former USDA scientist Jonathan Lundgren, who recently left the agency, has filed as a whistleblower, claiming the USDA suppressed his research, which focuses on pesticide use and its impacts. Lundgren was a Senior Research Entomologist and Lab Supervisor for the USDA Agriculture Research Service. Among other complaints, Lundgren says that USDA disciplined him in retaliation for speaking publicly about his research and has stifled his ability to publish and talk to the media. When USDA dismissed his complaint, Lundgren appealed. In response, an internal Scientific Integrity Review Panel was convened and concluded that “the allegation was unsubstantiated and there was not a preponderance of evidence to establish that the USDA [scientific integrity policy] was violated.” Then in February, the USDA Inspector General Phyllis Fong announced that she would be opening an investigation into the high volume of allegations around scientific integrity the department has been seeing, including Lundgren’s case.

Without getting into the details of the allegations, its worth looking at the degree to which the USDA scientific integrity policy is or isn’t equipped to address these kinds of issues. For one, the lack of clear guidance on if and how employees might express their personal views leaves scientists like Lundgren vulnerable if they choose to speak about research that might have policy implications or otherwise be controversial. The provision that “scientists should refrain from making statements that could be construed as being judgments of … USDA” casts a fairly wide net under which the agency could claim scientists acted inappropriately.

Second, Lundgren’s case shows that the department hadn’t fully developed how it would handle scientific integrity accusations and appeals. Interestingly, the USDA scientific integrity policy update did add a provision on protection of whistleblowers and did add significant language on how such policies would be carried out. Note the above figure from the USDA Scientific Integrity Policy Handbook. But it appears some of this procedural language hasn’t been fully implemented, at least in Lundgren’s case.

PEER also finds shortcomings in the policy. The group filed a petition for rulemaking at the agency asking it to revise its scientific integrity policy to include more clarity on handling scientific misconduct cases.

Changing agency culture

Regardless of how Lundgren’s case evolves, it highlights an important challenge at the USDA: How do you clearly and consistently implement a policy through a large and geographically dispersed agency? My past analysis of scientific integrity across the government suggests that these agencies may have more difficulty with policy implementation than those that are more centralized.

This case also reminds us that policies are easier to change than culture. Agencies can improve their policies on paper, but the work doesn’t stop there. Agencies need to take steps to facilitate a change in practices across the agency to ensure they are consistent with the new policy. Agency leadership must make clear to employees that scientific integrity is valued and that the agency takes the policy seriously. If this type of positive reinforcement isn’t consistently applied, employees are apt to forget or revert back to practices under past administrations when scientific integrity wasn’t a priority.

The Obama Administration has made strides in promoting the importance of scientific integrity across the government, but we still have a ways to go before the scientific integrity policies are effectively implemented at the USDA and elsewhere.