Georgia Tech researcher pays a high price for mismanaging an NSF grant

By Jeffrey MervisApr. 29, 2020 , 5:40 PM

When it comes to her research, Eva Lee sweats the details. The Georgia Institute of Technology engineering professor “is extraordinarily talented” at sifting through massive amounts of health care data and finding “novel insights” into how to save lives, improve care, and reduce costs, says physician scientist Brent Egan of the American Medical Association (AMA), who has collaborated with Lee on treating patients with cardiovascular disease.

Lee’s skills are now in high demand. Public health officials from around the world responding to the COVID-19 pandemic are clamoring to use software that she began to develop nearly 2 decades ago. And Lee’s participation in a group of U.S. scientists who raised an early alarm about the pandemic have garnered her national media attention, including on the front page of The New York Times.

In contrast, she’s paid much less attention to the reporting requirements on the grants that have supported her research for more than 2 decades at Georgia Tech. And the 55-year-old applied mathematician is now paying a steep price for that neglect.

In December 2019, Lee pleaded guilty to federal criminal charges of providing false information to the National Science Foundation (NSF) on a project it has been funding for more than 10 years. She assumed taking that step would allow her to get on with her life. “I’m good with math, and I like spending time with my students,” she says. “I thought the plea would let me go back to work.”

But it’s not that simple. On 21 May, a judge is scheduled to impose her sentence. And being convicted of a felony is likely to trigger a review by Georgia Tech officials that could end in the university revoking her tenure and dismissing her.

The circumstances that have led Lee to simultaneously be both a felon and a high-profile coronavirus warrior may well be unique. (NSF officials say they don’t keep statistics about the prevalence of fraud in the program that funded Lee.) Even so, her case is a sobering reminder to the community that, although institutions are responsible for ensuring that every grant they receive is properly managed, individual scientists may face serious consequences if they don’t follow the rules. Such missteps have jeopardized Lee’s career—and even her freedom.

Improving health care

Over the years, Lee has received more than $10 million from a half-dozen U.S. government agencies to support her work in developing optimization models to improve health care. Yet it’s one of the smallest of those awards—an NSF grant of $240,000 over 5 years—that has gotten her into so much trouble.

That grant supported her continued participation in the Center for Health Organization Transformation (CHOT), a multisite project launched in 2008 by NSF’s Industry-University Cooperative Research Centers (I/UCRC) program. By paying $50,000 a year to become members of the center, companies earn the right to help shape its research agenda and to use those findings to improve their operations. Based at Texas A&M University, College Station, CHOT touts itself as the only one of NSF’s current batch of 73 I/UCRC centers that focuses on “innovations in the health care delivery.”

CHOT has provided a framework for Lee to team up with hospitals, academic medical centers, and health care providers in and around Atlanta. For example, her research has helped Grady Memorial Hospital cut waiting times in its emergency room by half and lower postsurgical infection rates for cardiac patients to near zero, says Michael Wright, a former senior vice president for operations at Grady. “There weren’t any ‘a ha’ moments,” says Wright, who served on the center’s Industrial Advisory Board (IAB). “But her model … gave me the data I needed to make the case for needed improvements.”

Lee’s work also helped the Care Coordination Institute improve its services to cardiac patients, says Egan, who participated in that project and is now AMA’s vice president for cardiovascular disease prevention. By searching through the institute’s database of 2 million patients to identify key risk factors, Lee was able to help develop guidelines that improved the quality of care for thousands of patients, Egan says.

NSF investigates

In 2014, NSF renewed CHOT for 5 years, with Texas A&M remaining as the lead institution and Georgia Tech one of six university partners. But Lee says the procedures for complying with the terms of the second phase of the grant changed. Instead of simply signing off on documents already completed by Georgia Tech research managers, as she had done during the first phase, Lee says she was told she needed to do the paperwork herself. And doing that, she admits, led to her downfall.

Lee committed three violations, according to an investigation done by NSF’s Office of Inspector General (OIG), which won’t say how it came to investigate Lee. One was failing to document that Georgia Tech was receiving sufficient funding for the center from the requisite number of companies—a total of $175,000 from at least three, according to NSF’s rules. Lee says she knew exactly what she and her students were doing on every center project. But she was fuzzy on the financial arrangements that enabled that research.

“I had no idea what [the industry partners] were contributing,” she says. “But I knew that it was a lot more than the required amount. So, I just put down $50,000 for each one.” Five members had contributed a total of $250,000, she told NSF in annual reports covering 2016–18.

But NSF’s OIG found that wasn’t the case. “According to reports produced by Georgia Tech, there was no more than one member between 2016 and 2018, and the amount received in industry membership fees totaled $112,000 from 2016–18,” OIG’s head of investigations wrote to NSF’s chief operating officer on 7 August 2019. (Lee doesn’t dispute that finding, but she says the university was putting company contributions into different accounts than the one maintained for the center. She also notes that the partners were still paying for what her students were doing, which she viewed as tangible evidence that they were still participating in the project.)

The OIG investigation also found that Lee had manipulated the process used to set the center’s research priorities. IAB members hear pitches from each research team and then cast votes at the end of their spring meeting, one of two held each year. But in an interview with OIG in April 2019, Lee “admitted that she [had] voted on behalf of IAB members.”

Lee’s third violation, according to OIG, was attaching the signature of a Georgia Tech grants administrator to the annual reports that she filed with NSF. Lee admitted to OIG that she had lifted the signature from a previous document that had been submitted to NSF.

The probe expands

Based on its investigation, OIG concluded Lee had behaved in a way that represented “a significant departure from accepted practices.” And in September 2019, NSF decided she could not serve as a reviewer or consultant on grant proposals for 3 years.

Lee’s lawyer, Buddy Parker, characterizes the reviewing ban as “a slap on the wrist.” He notes that NSF did not conclude Lee had committed research misconduct, a more serious offense that can lead to a multiyear ban on submitting grant proposals to NSF or another federal agency.

But Lee wouldn’t be in such hot water if NSF’s penalty had been the only consequence of her misdeeds. One day after her 3-hour interview with NSF investigators on 10 April 2019, the U.S. attorney’s office in Atlanta told Lee she was also the target of a federal investigation for aggravated identify theft (for the purloined signature) and intent to fraud (for the incorrect information sent to NSF). The U.S. attorney’s office has declined to comment on the reasons it decided to prosecute Lee.

On 5 December 2019, Lee agreed to plead guilty to two lesser charges: filing a false statement to NSF about the status of the center’s industry members, and lying to OIG during the interview about her behavior relating to voting at the 2018 IAB meeting. Both are felonies. But they don’t come with the mandatory prison sentence that could have resulted from prosecution of the other charges. Instead, the U.S. attorney has asked District Court Judge Steve Jones to impose 8 months of home confinement.

A lack of support

How does a veteran professor at a top-ranked research university get tripped up by what seems like just a little more paperwork? Friends and colleagues don’t excuse Lee’s errors, but they also don’t seem surprised that it happened. Both Lee’s personality and her surroundings likely played a role, they surmise.

“She’s flat-out brilliant. … She also wants to help people,” says Rice University mathematician Richard Tapia, a National Medal of Science winner and her graduate adviser in the 1990s. But Lee can have difficulty with some everyday tasks, he says. “She has trouble tying her shoelaces, she can’t operate a copying machine, and she’s prone to doing stupid things.”

Lee doesn’t disagree with her former mentor and colleague. “I can’t even adjust the thermostat,” she confesses. “I’m not a systems engineer, I’m an applied mathematician.”

Tapia, who has won accolades for his mentoring of minority scientists and whose record of public service led Lee to choose Rice over Stanford University and the Massachusetts Institute of Technology for her graduate studies, says that saying she lacks practical skills “is actually a compliment.” In Lee’s case, he guesses, her passion for her work—including helping underserved populations, both students and those with unmet medical needs—blinded her to her obligation to follow the rules. And she had nobody at Georgia Tech to prevent her from making such poor choices, he adds.

“She has a very high regard for herself,” Tapia says, “and she makes a lot of administrative mistakes. So do I. Fortunately, I have someone whose job is to save me. But Eva doesn’t.”

Mark Prausnitz, a Georgia Tech biochemical engineering professor who has known Lee for 20 years, says he’s been able to use resources within his department to employ an assistant to help with the necessary paperwork on his grants. “But she doesn’t have the benefit of someone like that,” he says of Lee, who is in a different department within the engineering school.

Georgia Tech, like all large research universities, has an office that is responsible for managing external grants. But faculty members can’t always count on getting the help they need from Georgia Tech’s Office of Sponsored Programs (OSP), Prausnitz says. “Some [people in the office] are excellent and very responsible, while others are harder to work with,” Prausnitz says. “The individual makes all the difference.”

Lee says OSP staff were “extremely helpful” during the first phase of her grant, despite considerable turnover. But that collegiality vanished during the second phase, she says. And without that help, she says, she wound up cutting corners.

Georgia Tech’s response

Not surprisingly, Parker doesn’t think his client deserves the punishment prosecutors have recommended. “Dr. Lee stole no monies from NSF,” he says. “She stole no monies from Georgia Tech. Dr. Lee caused no harm to NSF and Georgia Tech.” Parker plans to ask the judge to put Lee on 6 months’ probation, a sentence he believes recognizes the value of her services to hospitals and to public health officials.

Lee acknowledges her mistakes, and Parker says he understands that NSF was obligated to report its findings to the Department of Justice. However, Parker decries what he calls “an abandonment … of prosecutorial discretion” and the government’s refusal to agree to an out-of-court settlement.

Parker and Lee are even more outraged by Georgia Tech’s response to the NSF investigation. One week after NSF interviewed her, Lee was suspended with pay and banned from campus. She was denied access to her research data and her Georgia Tech email account and was told not to talk to anyone connected to the university.

Parker accuses the university of trying to cover up its own mistakes. Its program office “for over 3 years failed to support Dr. Lee in the administrative work with NSF,” Parker asserts. “OSP’s incompetence, its deliberate ignorance of its responsibilities to file with NSF the necessary forms, render it equally culpable in wrongdoing. Yet, when questioned by NSF, [Georgia Tech] pointed the finger at Dr. Lee, claiming she is responsible, not us.”

Georgia Tech officials have called Lee’s suspension “standard administrative practice” in the wake of a federal investigation of a faculty grantee. The university lifted its communications ban after Lee pleaded guilty in December, but has not allowed her to access her computers. “Employees on administrative leave are not allowed access to Georgia Tech resources,” the university said last week in a statement clarifying its stance.

University officials declined to answer specific questions about its oversight of the CHOT award, including why Lee needed to take a larger role in managing her award. But W. Blair Meeks, assistant vice president for external communications, said “Georgia Tech disagrees with Dr. Lee’s claim” that it chose not to get involved in overseeing the second phase of the project. Meeks said the university “also disagrees … that she was left on her own.”

A career in the balance

Lee says her prolonged separation from campus has wreaked havoc on her lab, which 1 year ago included 15 Ph.D. students, 10 master’s students, and 40 undergraduates. Some graduate students have been assigned new mentors; Prausnitz, for example, took on two students who were preparing to defend their dissertations. Others remain in limbo, hoping Lee will be allowed to return to campus—as she desires—and resume her research and teaching duties.

That decision rests with Georgia Tech. The university has the authority to strip a faculty member of tenure and dismiss someone who has committed a felony. But it’s not mandatory that they do so.

Prausnitz, Tapia, Wright, and Egan are among those who have written letters to the judge asking him to show leniency. They are hoping their missives will also influence the university’s ultimate decision on Lee’s status.

Meeks acknowledged that the wheels are turning. “We are not doing interviews about this because Dr. Lee has not been sentenced and she is undergoing employment review,” he said. “We do not want to disrupt either process.”

Still, Tapia thinks Lee faces long odds. “It’s going to take someone with a progressive attitude” to reinstate her, he says. “Most universities aren’t happy if you are nonstandard, and Georgia Tech has been very bureaucratic in handling her case.” Prausnitz worries Lee will be hard-pressed to find another academic position if she has to leave Georgia Tech.

For the moment, Lee says she is simply awaiting her sentencing, and working 18-hour days at home, consulting with government officials and public health scientists about how best to marshal resources to battle COVID-19. Applying any lessons learned from her mishandling of the NSF grant, she says, will have to wait until the world begins to re-emerge from the latest pandemic.