Frontiers’ journals saw large scale retractions—where does that leave the publisher’s reputation with researchers?

Brian Owens

BMJ 2024;384:q659

http://dx.doi.org/10.1136/bmj.q659

Frontiers, like other for-profit open access publishers, has struggled with a perception it has low editorial standards and weak peer review. But is that true? Brian Owens reports

In September 2023 Frontiers—one of the world’s largest open access scientific publishers, with a stable of 230 journals covering just about every field of science—retracted 38 papers. All had been linked to the “unethical practice of buying or selling authorship on research papers,” known as “authorship for sale,” in which authors, during the review process, sell coauthorship to people who have not contributed to the research.

In response to the scandal the company changed its policy on requests for changes to authorship after a paper has been accepted. The new policy states that requests for changes to the author list will be accepted only “under exceptional circumstances and after in-depth assessment by the Frontiers’ research integrity unit.” Frontiers also said it would maintain a record of all such requests to identify suspicious patterns and trends. And it pledged to monitor websites where authorships are known to be put up for sale for any mention of Frontiers content.

Anastasia Long, the company’s public relations manager, says that authorship for sale is an industry-wide problem, potentially affecting more than 10 000 papers. “As with all attempts to systematically breach codes of ethical conduct, publishers must work together to put a stop to this practice,” she tells The BMJ. “This is why we also share details of our investigations and the methods we use to identify authorship-for-sale papers, and ‘paper mill’ papers, with publishing organisations.”

Helen Macdonald, The BMJ’s publication ethics and content integrity editor, says: “When changes in authorship are requested we need to understand what change is being proposed and why. We also need to know that all of the other authors are aware of any additions, removals, or changes in key authorship positions and that there is no dispute about authorship. Journals consider whether author change requests seem genuine and credible. At that time it is worth considering whether there are any irregularities or concerns about the content or the interactions with the authors. It can be a challenge, because we cannot typically verify what each of the authors did or did not do.

“BMJ considers that the sale of authorship is unethical and unacceptable. Were there evidence of this before publication then the content would not be published, and we would consider the issues in more detail. If concerns arose after publication then we would consider whether the content required post-publication changes, such as a correction, expression of concern, or retraction. In all instances we would consider further action, such as contacting the authors’ institution.”

This is not the first time Frontiers has been involved in controversy. In 2015 the publisher was added to the now defunct “Beall’s list” of predatory publishers after the editors of three of the company’s medical journals complained that it was interfering with editorial independence and had “unacceptable peer review procedures,” such as allowing authors to pick which associate editor handles their paper and sometimes allowing Frontiers staff to override editorial decisions. In response Frontiers removed 31 editors and strongly denied any editorial interference.

Long says, “Even a superficial fact checking of our editorial programme makes clear that we are a recognised, world leading open access publisher, with clear and transparent editorial processes alongside dedicated and extensive research integrity resource and investment. It’s frankly baffling that this defamatory term [“predatory”] persists in some circles.”

Although some websites still list Frontiers as a predatory publisher, to which Frontiers has publicly objected, many in the industry say the reality is more complex.

Not so simple

“Frontiers is a really interesting case,” says John Dupuis, a librarian in the science and engineering library at York University in Toronto, Canada. “Some of its journals are quite good and have a decent reputation in their field, others not so much.”

Ivan Oransky, cofounder of Retraction Watch, is not a fan of the “predatory” label either, for Frontiers or any other publisher. “We should eschew false binaries,” he says. “Everything is a continuum, and the quality of publishers is no exception.”

In fact, he says, there is probably a bigger difference in quality between journals at any given publisher than between different publishers.

That reflects the experiences of researchers who have written, reviewed, or edited for Frontiers’ journals. Keith Brunt, who studies translational medicine at Dalhousie Medicine New Brunswick in Saint John, Canada, has been an associate editor at Frontiers in Physiology for 10 years. He says he has had both good and bad experiences working with the publisher. He disagrees that it is predatory and says the fact that editors and reviewers are named on each paper improves accountability. But he says it can be more difficult, when peer reviewing for Frontiers, to reject papers that receive poor peer reviews than for other journals. Associate editors do not have enough power to set minimal criteria or to reject on the basis of suspected fraud, and there is no limit on the number of times authors can re-submit, sometimes leaving reviewers frustrated and overwhelmed. “It is not impossible to reject; it just takes a bit more effort and being direct—for example, by saying ‘address this or it’s a no-go,’” he says.

Long says the company does not get involved in peer review and editorial decisions. She tells The BMJ, “Frontiers’ approach to publishing is to put the editorial responsibility squarely in the hands of the editorial board, who collectively make acceptance or rejection decisions and have the independence to shape the direction of research in their journal.”

The company’s data show that, after authors’ revisions during the review process, handling editors follow the recommendations of the first two reviewers about 90% of the time, says Long. And rejection rates have been rising as Frontiers enhances its quality controls, she adds. In 2023 the desk rejection rate was 33% and the overall rejection rate was 56%. A survey of visitors to the Frontiers website found that 92% of researchers rated the company’s peer review as excellent or good.

Impossible scale

Failure to detect scientific fraud is hardly unique to Frontiers, Dupuis points out. All the big publishers, such as Elsevier and Springer Nature, have faced huge retractions in recent years involving hundreds of papers. “The issue is rampant throughout the entire publishing ecosystem,” he says.

And that is largely because the industry has grown so much in recent decades, probably beyond its ability actually to review all research to the desired level, says Oransky. “The elephant in the room is the impossible scale,” he says. “There are not enough people to do all the peer review that is needed. The industry is setting itself up to fail.”

The huge number of journals out there is part of the reason predatory publisher blacklists, or alternative whitelists of trustworthy journals, are so popular, to take some of the guesswork out of deciding where to submit a paper. But even the people who run those lists warn that they are not a substitute for an author’s own critical thinking.

For example, most of the Frontiers titles indexed in the Directory of Open Access Journals (DOAJ), an independent, community run listing, have been awarded the “DOAJ seal,” something that less than 10% of journals in the directory achieve. The seal indicates that the journals are adhering to best practices in open access publishing, says Judith Barnsbury, head of editorial at DOAJ, such as using a Creative Commons licence, authors retaining copyright, and providing article metadata to DOAJ. But, she says, the seal isn’t intended to imply anything about editorial quality.

Because it is often interpreted that way, DOAJ spends a lot of time reviewing the journals that apply for the seal, to make it as much of a whitelist as possible. “We don’t want to include predatory publishers,” says Barnsbury. Journals can have the seal removed if people report serious concerns, but that has not happened with any Frontiers titles recently, she says.

No list, black or white, is going to be infallible, so resources such as the Think Check Submit website (https://thinkchecksubmit.org) provide a good reference guide when thinking about where to publish your research, instead of relying on lists that are by design binary, says Dupuis. “I understand that the lists save time and mental effort, but they can lead you down the wrong road,” he says.

“It’s more mental effort to do your due diligence, but it’s worth it.” That due diligence might need to be more stringent when considering a Frontiers journal rather than a big commercial or scholarly society journal that you already know well, says Dupuis, but it still needs to be done even for journals from high profile publishers, such as Elsevier and Springer Nature, whose publications can very a lot in quality.

“I don’t have a lot of sympathy for the idea that Frontiers is any worse,” he says.