Read our COVID-19 research and news.

The U.S. Department of Defense is one of five research agencies faulted for gaps in monitoring foreign influences.

E. PETERSEN/SCIENCE

Report finds holes in U.S. policies on foreign influence in research

A new report by a congressional watchdog says U.S. agencies need to flesh out and clarify their policies for monitoring the foreign ties of the researchers they fund.

The report, by the Government Accountability Office (GAO), is likely to spur efforts in Congress aimed at preventing China and other nations from using funding and other connections to gain improper access to research funded by the U.S. government. But at least one of the agencies under scrutiny—the National Science Foundation (NSF)—is pushing back on the idea that its policies are lax. It is warning that tougher rules could hinder its ability to fund the best science.

The GAO report was requested by Senator Chuck Grassley (R–IA), chairman of the Senate Committee on Finance, who in hearings has prodded research agencies to “pick up their game” when it comes to preventing improper foreign influence. It examines the practices of the government’s five biggest funders of academic research: the National Institutes of Health (NIH), NSF, NASA, the Department of Energy (DOE), and the Department of Defense (DOD). The report recommends they adopt explicit and uniform policies on what grantees need to do to comply with federal laws relating to three issues:

  • financial conflicts of interest (COIs);
  • nonfinancial conflicts that include unrealistic time commitments or duplication of research; and
  • disclosure of all sources of research funding, both foreign and domestic.

Although all five agencies have disclosure policies, the GAO report says DOD and DOE lack an agencywide financial COI policy. None of the agencies defines or asks grantees to describe potential nonfinancial conflicts. Those gaps and ambiguities, GAO concludes, have led to “incomplete or inaccurate information from researchers that … impede the agency’s ability to assess conflicts that could lead to foreign influence.”

GAO also chides the outgoing administration for failing to deliver long-promised guidance that is being developed by the White House Office of Science and Technology Policy (OSTP) in an interagency process begun nearly 2 years ago. “Most agencies are waiting for the issuance of OSTP’s guidance before they update their policies,” the report notes, adding that the delay has deprived them of “timely information needed to fully address the threats.”

Fresh numbers

To date, federal research agencies have taken different approaches in dealing with unwanted foreign influence over their grantees. The GAO report documents, in some cases for the first time, the extent of those activities at the five agencies it examined.

The data suggest NIH has been the most aggressive, by far. Three years ago, an in-house team at the biomedical behemoth began to try to identify grantees who had not properly disclosed foreign ties. One key element involved comparing information grantees had provided in their funding applications about non-NIH sources of support with acknowledgements of support in the footnotes of their publications.

To date, that effort has turned up 455 researchers “of possible concern,” the GAO report notes. And the effort appears to be ongoing; in June, NIH officials said they had vetted 399 such cases. Of those, NIH told GAO that six have led to criminal complaints filed by the U.S. Department of Justice. An additional 32 cases were referred to the inspector general of NIH’s parent agency, the Department of Health and Human Services.

The report doesn’t provide a similar tally for the other agencies. “They vary in how they collect the data, and it’s hard to separate potential cases of foreign influence from other types of alleged violations,” says Candice Wright, acting director of GAO’s science, technology assessment, and analytics office, which conducted the study.

At the same time, GAO was able to collect some preliminary or incomplete numbers from the four other agencies; those figures suggest none has taken NIH’s proactive stance. Instead, GAO says, those agencies rely heavily on tips from sources outside the agency—including the FBI or an individual with insider knowledge of an alleged violation—to trigger an investigation.

For example, GAO says, “NSF estimates that it has taken administrative action against nearly 20 grant recipients who failed to disclose foreign ties.” That number matches what NSF officials reported this summer. (Wright says GAO received no information on the size of the initial pool of allegations.)

NASA “has 14 open cases of grantee fraud with a foreign influence component,” according to the GAO report, which hints it is a growing problem. “The number of such case has approximately doubled in the last year,” it notes.

At the Pentagon, one unnamed unit has nine open cases “involving foreign influence at U.S. universities,” the report notes. DOE’s inspector general, meanwhile, has reported “21 active cases involving foreign influence.”

Why they want to know

Grassley asked GAO to focus on foreign influences over researchers working in the United States who receive federal funds. So GAO honed in on the lack of federal policies explicitly designed to detect efforts by foreign entities to game the traditionally open U.S. research enterprise, say, by telling grantees to keep mum about the relationship or by trying to shape the direction of the research.

But agency officials say upholding research integrity consists of more than just learning about who else might be funding someone applying for a grant. For example, NSF says its disclosure policies are designed to obtain a wide range of information that helps the agency with its grantmaking. Knowing such things as an applicant’s background, collaborators, and access to relevant resources helps NSF make better decisions on who to fund, explains Rebecca Keiser, NSF’s chief of research security strategy and policy. And every bit of information is useful: “All means all” sources, she emphasizes.

In contrast, Keiser says, NSF’s policy governing conflicts of interest is meant to ensure the results of the funded research have not been skewed because of any number of outside factors. The most obvious are financial conflicts, in which a scientist stands to profit from the outcome.

But there are also nonfinancial conflicts that could sway the results. One example is when a researcher takes on more work than they can handle. That overbooking is called a conflict of commitment. The researcher’s institution is the arbiter of whether any particular relationship—such as with a company or foreign university—crosses the line, Keiser adds, and how the problem should be resolved.

As federal officials press for more reporting rules on potential foreign influence, it’s important that they not conflate conflict of interest and conflict of commitment policies, Keiser says, a point NSF Director Sethuraman Panchanathan emphasized in a four-page written response to the GAO report. “Not every foreign relationship represents a conflict of commitment,” Keiser says. “And we wanted to make that clear to GAO.”

“We look at disclosures to determine capacity and potential overlapping research,” she continues. “We want investigators to be comfortable disclosing any connection that bears on their work, without fear that it will automatically be labeled a conflict.”

What’s JCORE?

Keiser is part of the interagency OSTP group called the Joint Committee on Research Environments (JCORE) that is examining foreign influence policies. It has come up with a definition of both financial and nonfinancial conflicts as part of the pending guidelines relating to foreign collaborations. Although OSTP Director Kelvin Droegemeier and other committee members have made numerous presentations this year to the academic research community, GAO found they have been flying under the radar of their intended audience.

The agency asked 52 rank-and-file scientists—chosen because they hold large awards from at least two of the five agencies being examined—whether they were familiar with JCORE and its attempt to refine federal policy on foreign influence in research; 49 said they didn’t know about it. “I’m concerned by that number,” Keiser says. “We obviously need to do more outreach.”

Congress is likely to provide one such forum in the months ahead. A Grassley staffer says the issue remains “a high priority for” the senator, who is in line to lead the powerful Judiciary Committee should Republicans retain control of the Senate. “The government has a ton of blind spots” when it comes to foreign influence, according to the aide, “and the GAO has done a good job identifying those gaps.”