The U.S. National Institutes of Health (NIH) thinks it may have figured out how China’s foreign talents recruitment program is undermining its system for making awards and ensuring ethical behavior by its grantees.
In an interview yesterday with Science, Michael Lauer, director of NIH’s extramural research program in Bethesda, Maryland, described a two-pronged strategy that NIH believes China’s Thousand Talents Program has pursued to improperly reap the benefits of NIH-funded research. One entails breaching NIH’s vaunted system of reviewing grant proposals to share information with colleagues in China. The second consists of setting up shadow labs in that country to replicate NIH-funded research.
Lauer offered no new evidence to support those assertions and no data on how often these tactics have been used. But his description adds considerable detail to previous NIH statements addressing concerns by Congress and officials in President Donald Trump’s administration that federal research agencies aren’t doing enough to combat attacks on U.S. science by foreign entities, particularly China.
Last year, NIH officials shocked the U.S. biomedical research community by firing off letters to more than 60 institutions regarding NIH-funded scientists it believed may have failed to disclose all foreign ties or violated NIH’s policies on peer review. This spring, the MD Anderson Cancer Center in Houston, Texas, and Emory University in Atlanta dismissed several faculty members of Asian descent in the only two investigations that have been made public. Lauer says some 250 scientists were put under the microscope initially, and 180 of those cases remain active as NIH wades through how other institutions have responded.
A report by a federal watchdog agency out today chides NIH for ignoring national security concerns in selecting peer reviewers. The new report, by the Office of Inspector General (OIG) within the Department of Health and Human Services (HHS), NIH’s parent body, suggests NIH consider paying “extra attention” to screening scientists “who would be reviewing grant applications with particularly sensitive subject matter or that have lucrative commercial applications.”
The report doesn’t provide any data on the scope of the problem. Lauer estimates that “maybe 10%” of the 180 active cases involve peer-review violations. But, he says, that percentage could be misleading.
“We don’t know the scale of the problem,” he admits. “And we are concerned that the scale is much worse than what we are seeing.” Most reviewers are also grantees, Lauer notes, meaning that any steps NIH takes to address its concerns about grantees would also extend to the pool of reviewers.
Leaks and shadows
Lauer says NIH’s investigations are not limited to those interacting with Chinese institutions. But he repeatedly cited that country’s Thousand Talents Program when describing what NIH is doing to tackle the problem. He flagged two of its tactics as especially troubling.
One is to encourage scientists to become a member of an NIH study section and then share grant applications under review with Chinese colleagues. “We know that one of the goals of the Chinese Thousand Talents Program is to obtain information,” Lauer says. “So, the leakage of information through peer review is an item of great concern to us.”
One of the fired MD Anderson scientists used his position on an NIH study section to share grant applications with Chinese colleagues, despite knowing his actions violated NIH policy, Lauer asserts. “He’d send them off to China, often with commentary,” Lauer says. “And he would say this material is confidential.”
The second threat to research integrity, according to Lauer, comes from inviting NIH-funded researchers to set up so-called shadow labs in China as part of the country’s foreign talent recruitment program. Such an arrangement, Lauer says, allows a Chinese institution to gain direct access to the research that NIH is funding.
“It’s a very patient approach,” Lauer says. “You can build a tremendous amount of knowledge about basic, preclinical science by setting up shadow labs in China that are mirrors of U.S. labs. And then, once the research gets to a point of being translational, it’s already in China.
“It lets you get around the problem of expert controls,” he adds, referring to the process of winning U.S. government approval to transfer sensitive technologies to another country. “You don’t have to worry about tech transfer—you’ve already transferred the technology to yourself.”
Who goes first?
Knowing how China’s foreign talents program operates has helped NIH figure out a way to spot scientists who may have crossed the line, Lauer says. One telltale sign, he says, is how scientists list their affiliations on papers they publish.
“What we are looking for is an NIH-funded scientist who listed one affiliation in China and another at their American institution,” he begins. Such dual affiliations are not uncommon, but Lauer says “we are especially interested in cases in which the Chinese affiliation is listed first. That’s important because we have seen, in the contracts that these scientists sign with Chinese institutions, that they are explicitly told to make sure that the Chinese affiliation is listed first.”
Why is that so important to China? According to Lauer, being listed first “enhances their ratings on the various citation indices. The research gets credited to the Chinese institution. And then the Chinese institution can say we are one of the most highly cited institutions in the world.”
Lauer’s staff spent an average of 10 hours probing the publishing history of each of the 250 scientists it initially flagged. “It’s not as straightforward as it might sound,” Lauer says about how his staff scraped publication data to learn more about their collaborations and sources of funding. “But they were able to put together a list of scientists who fit our phenotype and who we might want to look at in more detail.”
Guilt by geography?
NIH has said that not all the scientists it is scrutinizing are Chinese. But many in the scientific community worry that the steps it and other agencies are taking in response to the perceived threat from foreign entities could target certain racial and ethnic groups.
“While we must be vigilant to safeguard research, we must also ensure that the U.S. remains a desirable and welcoming destination for researchers from around the world,” warns a 4 September letter signed by 60 scientific organizations, including AAAS (which publishes Science). “Finding the appropriate balance between our nation’s security and an open, collaborative scientific environment requires focus and due diligence.”
Federal officials seem to recognize they are walking a fine line. For example, the OIG report recommends that NIH “leverage the expertise” of HHS’s Office of National Security (ONS), which is responsible for vetting NIH employees and foreign visitors to its Bethesda campus. And ONS regards nationality and home institution as possible risk factors. But both OIG and NIH say they want to avoid weighing geography in assessing potential risk.
“We have no basis to say” that being a foreign national or working at a foreign institution poses an inherently greater security risk, says Erin Bliss, assistant inspector general for evaluation and inspections. “And NIH has talked to us about how important collaborations with foreign researchers are, in terms of advancing science. But we do recommend that NIH work with national security and intelligence experts to figure out what are the appropriate risk factors that they should be considering in targeting their oversight activities.”
Lauer says the OIG report “is spot on” in suggesting that it tap the expertise of ONS and other federal agencies dealing with national security. But he questions its suggestion to look more closely at research on sensitive topics.
“We don’t fund any classified research,” he says. “So that may be more of an issue for the Department of Defense or the Department of Energy.”
Limited resources are a major constraint, Bliss and Lauer agree. That appears to rule out background checks. Vetting all 27,000 reviewers used each year would require an additional 100 full-time employees, NIH told OIG.
Existing program staff can play an important role, Lauer says, but they can’t carry the entire load. “Our SROs [scientific review officers] aren’t investigators, and they aren’t detectives,” Lauer says. “But we do tell them, ‘If you see something, say something.’ They already receive extensive training on what might constitute inappropriate behavior by a reviewer.”
NIH told OIG it would take “at least 6 months to a year” to come up with a “risk-based approach for identifying those peer reviewer nominees who warrant extra security.” And that’s not likely to be the final word. OIG is working on another report about how NIH can monitor reviewers already on the job.
That’s quite a challenge, Lauer admits, pointing to a December 2018 recommendation from an outside advisory body to the NIH director that NIH block reviewers from downloading applications and instead require them to read them only online. “Once an application is downloaded, we lose control over it,” Lauer points out. “So, banning the practice would give us more control.”
But that recommendation has yet to be adopted. “Everybody downloads applications,” he admits. “And there was a tremendous amount of pushback to the idea. But I would say it’s still on the table.”