A retraction often marks the end of a dispute over published scientific research. But in the case of a $300 digital fertility tracker marketed by Valley Electronics, it was only the beginning.
After a study on the effectiveness of its Daysy thermometer was retracted, Valley sued Chelsea Polis, a researcher who had publicly lambasted its findings, for defamation. When a federal judge threw out the case, Valley appealed, arguing that Polis went too far by calling the company “unethical” and labeling its study “junk science.” The appeal — set for March 22 — underscores the growing tensions between new, loosely regulated technologies and watchdogs insisting that published claims be supported by rigorous science.
To the critics, poking holes in studies is an inherent part of the scientific process — it’s baked into peer reviews and reproducibility research. But to companies raising millions of dollars or trying to rake in new customers, each round of debate can also look like a threat to the bottom line.
“Bad science can be used to prop up useless products and interventions, so we have to be able to criticize bad science,” said Jonathan Jarry, a science communicator at McGill University’s Office for Science and Society.
Jarry and other experts say that whether they’re successful or tossed aside, lawsuits like the one filed against Polis could have a chilling effect on science.
“These lawsuits make (scientific criticism) less likely to happen,” said Elizabeth Hall-Lipsy, a law and science policy professor at the University of Arizona. “People say, ‘Well, I saw what happened to that other person, so maybe I don’t publish my blog. Maybe I don’t make a comment on that product.’”
These kinds of legal battles can drain researchers of money, energy, and time as they play out over years. The dispute over Valley’s fertility thermometer began in March 2018, when a group of scientists published a paper in the journal Reproductive Health that analyzed whether people who used the thermometer and a paired app to track their fertile windows were more likely to get pregnant.
The authors of the study included Daysy’s medical director, Niels van de Roemer, and Martin Koch, who later joined Daysy’s scientific advisory board. They reported that just two of the 125 people in the study became unintentionally pregnant while using the Daysy thermometer, which uses basal body temperature and other information to predict whether someone might be able to conceive. (The Food and Drug Administration has not cleared the Daysy thermometer to be used as a contraceptive, and it isn’t currently marketed as such — though other apps that rely on similar metrics have received clearance.)
Behind the scenes, Polis — a reproductive epidemiologist — had been raising questions after seeing Valley’s marketing claims, including social media posts that seemed to claim the device was as effective at preventing pregnancy as an IUD.
“I care about the quality of studies used to make claims regarding the effectiveness of various contraceptive methods; about the prevention of unintended pregnancies; and about providing reliable information to people on which to base their contraceptive decision,” Polis wrote to van de Roemer in a September 2017 email reviewed by STAT. She also asked if more data had been published on Daysy’s effectiveness as a contraceptive device. She had “serious concerns regarding the marketing claims Valley Electronics makes regarding the use and effectiveness of Daysy for pregnancy prevention.”
Eventually, she made those concerns public. In June 2018, she published a commentary in the same journal critiquing the study’s methods. She also filed a complaint with the FDA about Valley’s marketing, according to documents reviewed by STAT.
In May 2019, the journal retracted the study, citing concerns “about the reliability of the estimates of contraceptive effectiveness for the Daysy device when being used together with the DaysyView app.”
When the study was retracted, Polis spoke with a BuzzFeed reporter. A few weeks later, she also wrote about it on her personal website in a strongly worded blog post titled “How an unethical company (Daysy) responded to retraction of their study.” She included links to her blog in some comments on Instagram posts.
On May 11, 2020 — almost one year after the retraction — Valley Electronics sued her.
“It was a complete shock,” said Polis, who at the time was a research scientist at the Guttmacher Institute. She has since left to become an independent research consultant. “I felt like the financial future of my family was in jeopardy, I didn’t know if I could talk about this to anybody.”
The company sued her for defamation, claiming her statements were untrue and cost Valley more than $1 million through lost sales and damage to its reputation.
Charles Avrith, one of Valley’s lawyers, said that the company saw a significant dip in sales after Polis started publicly commenting on the device, and that sales picked back up after she stopped posting in 2020.
While Valley aims to prove that Polis’s comments were defamatory, experts said courts generally have recognized disputes over science as an issue of opinion.
“Things can be proved through scientific discourse,” said David Schulz, a director of the Media Freedom & Information Access Clinic at Yale University. “But the give and take of that is not the sort of thing that should be litigated in the courts. It should be decided through scientific inquiry.”
Schulz said this general rule holds even in circumstances when debates get heated, leading to accusations of shoddy science and unethical behavior. “You can only have a libel claim for something that’s provably false,” he said. “And opinions don’t lend themselves to factual proof. They’re opinions.”
The Supreme Court found in a landmark 1990 case, Milkovich v. Lorain Journal Co., that opinions can imply the assertion of an objective fact, and therefore are not universally shielded from defamation claims. But lower courts have applied the ruling narrowly, finding that heated rhetoric and loose, figurative language should not be conflated with demonstrable falsehood.
“To describe something as junk science or to say that they have bad ethics — typically that would be protected as pure opinion,” Schulz said. “And even if you give an opinion based on fully disclosed facts, you can reach wrong conclusions. Scientists do it all the time. That’s the process.”
The legal system generally sees things in black and white, but the field of science has to accommodate far more fluidity. New discoveries can upend long-held dogmas, scientists can disagree on methodology or the interpretation of results, and further research can undercut findings once thought to be firm.
But for some companies relying on scientific research as part of their marketing machine — especially in sectors like digital health, where products are often already being sold directly to consumers when studies are published — that kind of uncertainty can pose problems.
In 2013, a company that made lung surfactants for premature babies sued another researcher for publishing a paper that found its products were less effective than a competitor’s. Two years later, a dietary supplement company sued a Harvard researcher for publishing a paper that concluded the company’s products contained an unapproved ingredient. Even more recently, a pharmaceutical company sued a medical journal itself for publishing research with an unflattering conclusion; that case was dismissed in February.
In all these cases, people were sued for materials published in a scientific journal. And in all three of those cases, the judges ruled that these papers weren’t defamatory, with one judge even ruling that scientific disagreements published in journals should be specifically protected from defamation claims.
Elisabeth Bik, a microbiologist and consultant who researches errors and misrepresentations in scientific publications, said in some circumstances, the purpose of these threats is not to correct the scientific record, but quell criticism. Her own scientific critiques have drawn personal insults, online harassment such as doxxing, and legal threats.
Polis’s case is somewhat unique, in that Valley is only suing Polis for the statements she made on social media, to reporters and on her blog — not for what she published in a journal.
Social media in particular has become a forum for scientific discussions. The editor in chief of Science recently told the tech news site TechCrunch that Twitter has become a significant place for scientists to “bat things around and openly air ideas, support them or shoot them down — the things they used to do standing around a blackboard, or at a conference.”
It’s also a tool scientists can use to talk about their field and their research findings to a broader audience. “Most people who are interested in fertility-awareness-based methods, they’re not reading scientific journals,” Polis said. “Folks are kind of on their own if they want to use these FABMs — so of course they’re turning to online communities for support.”
Valley argues that regardless of the medium, her remarks don’t reflect scientific debate. “I don’t think her remarks are scientific in nature,” Avrith, the company’s lawyer, said.
“I don’t think it will have any chilling effect on legitimate scientific debate. In fact, I’m worried about the opposite effect — that it will give people the idea that they can defame somebody by claiming that it has something to do with scientific debate. That’s what worries me,” he added.
Polis, whose legal team is representing her at no cost, said even without the financial impact, the lawsuit has cost her professional opportunities; a project about FABMs that she had been working on had to be shelved, she said. Nevertheless, she said she wouldn’t have withheld her criticism if she had known a lawsuit would follow.
“I would definitely still have written the commentary, I would definitely still have reached out to the FDA,” Polis said. “I’m happy to have the knowledge that this kind of thing can happen — it would have been helpful to know that beforehand. But I don’t think that any of my actions should have or necessarily would have been different.”