Pages

Tuesday, August 22, 2023

The Data Fraud Case - And Still More

An analysis in the Chronicle of Higher Education suggests that lawsuits are a blunt instrument for dealing with allegations of data manipulation. The problem in the recent high-profile Harvard case is that the response to the allegation seems to be that maybe it happened, but it wasn't me who did it. So, like self-driving cars, we seem to have self-manipulating data.

When Scholars Sue Their Accusers
Francesca Gino is the latest. Such litigation rarely succeeds.

By  Adam Marcus and  Ivan Oransky
8-18-23

Francesca Gino has made headlines twice since June: once when serious allegations of misconduct involving her work became public, and again when she filed a $25-million lawsuit against her accusers, including Harvard University, where she is a professor at the business school. The suit itself met with a barrage of criticism from those who worried that, as one scientist put it, it would have a “chilling effect on fraud detection.” A smaller number of people supported the move, saying that Harvard and her accusers had abandoned due process and that they believed in Gino’s integrity.

How the case will play out, of course, remains to be seen. But Gino is hardly the first researcher to sue her critics and her employer when faced with misconduct findings. As the founders of Retraction Watch, a website devoted to covering problems in the scientific literature, we’ve reported many of these kinds of cases since we launched our blog in 2010. Plaintiffs tend to claim defamation, but sometimes sue over wrongful termination or employment discrimination, and these kinds of cases typically end up in federal courts. A look at how some other suits fared might yield recommendations for how to limit the pain they can cause.

The first thing to know about defamation and employment suits is that most plaintiffs, but not all, lose. Mario Saad, a diabetes researcher at Brazil’s Unicamp, found that out when he sued the American Diabetes Association in the very same federal district court in Massachusetts where Gino filed her case.

Saad was trying to prevent Diabetes, the flagship research journal of the American Diabetes Association, from publishing expressions of concern about four of his papers following allegations of image manipulation. He lost that effort in 2015, and has now had 18 papers retracted.

That same year, The American Journal of Clinical Nutrition prevailed in a case brought by a researcher who tried to stop its editors from retracting one of her papers.

In 2016, Fazlul Sarkar lost a court bid to unmask the name of an anonymous commenter on PubPeer, a platform for such comments, whose criticism he blamed for costing him a job offer. Sarkar has now had 41 papers retracted.

In 2021, Rhawn Joseph, the “prince of panspermia,” had a $50 billion — no, that’s not a typo — case against Springer Nature over a retraction tossed by a federal judge. Joseph, as the judge in the case noted, “considers himself ‘one of the leading figures in the search for extraterrestrial life,’” and panspermia proponents posit that life arose elsewhere in space and then made its way to Earth. He has, he says, proved conclusively that there is life on Mars, but that NASA and the scientific community have responded with “slander, defamation and death threats.”

And this year, a judge upheld a decision against Pacira Biosciences, which had sued the American Society of Anesthesiologists over papers published in its flagship journal that were critical of Exparel, an analgesic sold by Pacira used to prevent pain after surgery.

Not all plaintiffs do so poorly. Elsevier settled a case in 2011 brought by an advocate for intelligent design who objected to the retraction of his paper. The author, Granville Sewell, had argued that “the Second Law of Thermodynamics makes it impossible for evolution to improve living organisms,” according to Joe Felsenstein. But as Felsenstein, a biologist, notes, “The obvious reply is that the biosphere is not an isolated, closed system, that to come near having one, we must also include the sun which undergoes a huge increase of entropy as it radiates energy, that more than compensates for the much smaller decrease of entropy involved in the evolution of life.”

Still, Elsevier agreed to publish a retraction notice that said the move was “not because of any errors or technical problems … but because the Editor in Chief subsequently concluded that the content was more philosophical than mathematical” and therefore not a good fit for the journal. The journal also offered its “sincere and heartfelt apologies” to Sewell.

Elsevier lost another case brought by a researcher who said the company had used “untruthful and unverified” language in a retraction notice. That notice, published in 2011, said the authors had “misappropriated data, plagiarized and concealed the authorship of work.” But after the researcher sued, the journal toned down the notice to say simply that “some of the findings have previously been published … without proper cross-referencing.”

And two researchers at the University of Pittsburgh earned a settlement with a journal that left their paper retracted but changed the wording of the notice to make them appear less culpable.
Then there are those cases where critics use the courts to press their claims. The fitness giant CrossFit won nearly $4 million in legal sanctions after charging that The Journal of Strength and Conditioning Research had defamed the company in a since-retracted paper. That paper had claimed nine of 54 participants dropped out of a study of CrossFit workouts because of “overuse or injury.” But it turned out that just two had dropped out — and neither because of such issues. The paper was eventually retracted for lack of institutional-review-board approval.

And Joseph Thomas, a whistle-blower at Duke University, earned $34 million in a case that cost the university a $112.5-million settlement — not to mention untold legal fees. Thomas — who was a technician in a lab that studied the effects of pollution on the lungs — alleged that Duke had knowingly included faked data in grant applications and reports for grants totaling $200 million.

The issues might have gone unnoticed were it not for the fact that the technician blamed for them, Erin Potts-Kant, had been caught embezzling more than $25,000 from the lab. That crime prompted an examination of data she generated, which had made it into some 50 studies. The scrutiny revealed serious issues, and 18 papers co-authored by Potts-Kant have now been retracted, and she has been banned for life from receiving federal funding — but Duke scientists and officials tried to cover it up, according to Thomas. In the end, Duke paid a hefty price, the largest such settlement by a university under the False Claims Act.

Such cases can be extremely expensive — not only for the defense, whether the costs are borne by institutions or insurance companies, but also for the plaintiffs. Ask Carlo Croce and Mark Jacobson.

Croce, a cancer researcher at Ohio State University, has at various points sued The New York Times, a Purdue University biologist named David Sanders, and Ohio State. He has lost all of those cases, including on appeal. The suits against the Times and Sanders claimed that a front-page story in 2017 that quoted Sanders had defamed Croce. His suit against Ohio State alleged that he had been improperly removed as department chair.

Croce racked up some $2 million in legal bills — and was sued for nonpayment. A judge has now ordered Croce’s collection of old masters paintings to be seized and sold for the benefit of his lawyers, and has also garnished Croce’s bank accounts. Another judgment means that his lawyers may now foreclose on his house to recoup their costs. Ohio State has been garnishing his wages since March by about $15,600 each month, or about a quarter of his paycheck. He continues to earn more than $800,000 per year from the university, even after a professorship and the chair were taken away from him.

When two researchers published a critique of the work of Mark Jacobson, an energy researcher at Stanford University, in the Proceedings of the National Academy of Sciences, Jacobson sued them along with the journal’s publisher for $10 million. He dropped the case just months after filing it.

But thanks to a so-called anti-SLAPP statute, “designed to provide for early dismissal of meritless lawsuits filed against people for the exercise of First Amendment rights,” a judge has ordered Jacobson to pay $500,000 in legal fees to the defendants. Jacobson wants Stanford to pay those costs, and California’s labor commissioner said the university had to pay at least some of them because protecting his reputation was part of Jacobson’s job. The fate of those fees, and who will pay them, is up in the air, with Jacobson once again appealing the judgment against him.

Even when defendants prevail, they do not always emerge unscathed. David Sanders, the Purdue researcher Croce sued, won that case, and his employer covered his legal fees. He remains a tenured professor. But his college and department “did not recognize my scientific-integrity efforts as valuable,” he told Retraction Watch two years ago when reflecting on the case.

“Indeed, it was conveyed to me that some of my investigatory endeavors, although they were not directed towards articles with authors from my university, were unwelcome,” Sanders told us. “There was a concomitant withdrawal of resources from my laboratory and me.”

In general, such lawsuits are a terrible way to respond to criticism — even, perhaps, illegitimate criticism. For one, judges and juries are, by design, not experts in any of the scientific subject matter. Here’s what Judge Timothy S. Hillman of the U.S. District for the District of Massachusetts noted in a ruling on Saad’s suit against the American Diabetes Association: “The dispute between Dr. Saad and the ADA over the reliability of the data in his articles is not fit for resolution in the form of a defamation lawsuit. Instead, this is a case where ‘the trial of ideas plays out in the pages of peer-reviewed journals, and the scientific public sits as the jury.’”

The fact that such cases are brought to court cannot but have a chilling effect. Would-be-critics self-censor. Publishers fail to correct the record. And universities protect fraudsters. Just ask Paul Brookes, who anonymously created a now-defunct site called Science-Fraud.org in 2012. His work drew praise, but it also drew the ire of his targets, who called for a lawsuit. When his university told him it would not cover his legal expenses, he shut down his site. In the years since, the substance of his efforts has proved right time and time again.

Regardless of how such cases and legal threats tend to play out, it seems highly unlikely that they will go away. If anything, in an age where scientific sleuthing is having a welcome moment, they appear likely to multiply. We’ve reported on two new cases since July.

With that in mind, we have two suggestions. The first is that sleuths use PubPeer to make their critiques anonymously. The site, a nonprofit founded in 2012, allows for comments on any papers that contain a digital object identifier (DOI) — a unique string of characters designed to track publishing events related to an article — or PubMed ID. Commenters may use their real names, or a pseudonym. The platform has proved resilient in the face of legal threats and has led to countless corrections, retractions, and university investigations. (Full disclosure, one of us is a volunteer member of PubPeer’s board of directors.)

The second is that critics might consider publishing their work under the umbrella of media organizations like Retraction Watch that have experience in this area — and defamation insurance. Underwriters tend to set a high bar for such policies, and they are out of reach of most individuals.

Working with media organizations will not prevent all lawsuits. But doing so will make them less likely because experienced journalists who know the definition of libel will serve as trainers and backstops — and will often consult with legal counsel before publishing. Those of us who have been doing this work a long time have learned from school, training, and the lawyers we’ve worked with certain rules of attribution and vocabulary. For example, we would have advised Brookes to avoid using the word “fraud” in his URL.

And partnerships could make the financial burden far less crushing if a suit does occur, since the media outlets (reputable ones, anyway) will have the shield of libel insurance. Raising a deductible from supporters is much easier than raising $1 million to defend against a suit. (A group of scientists have launched a $250,000 GoFundMe campaign to pay the legal costs of Gino’s accusers, for example.)

These kinds of collaborations will lead to more impact, and hopefully to even more skilled critics. Publishers have made it clear they have little interest in this work, and are happy to outsource it, so unless critics keep up their efforts, the scientific literature will remain sullied.

See you in court — or perhaps not.

Source: https://www.chronicle.com/article/when-scholars-sue-their-accusers.

No comments:

Post a Comment