Researchers and scholarly publishers need to be transparent and engaged with the public to prevent a flood of misinformation and fakery from undermining scientific integrity and trust in science, say academics and publishers.

So-called paper mills and predatory journals are churning out hundreds of thousands of papers annually that include shoddy or fraudulent research, no rigorous peer review, and outright fakery purporting to be science, they said during a webinar by the Canadian Science Policy Centre on “Maintaining Trust in Published Science.”

This misinformation and disinformation is leading to thousands of retractions of papers in the scientific literature, said David Moher (photo at right) clinical epidemiologist and senior scientist at the Ottawa Hospital Research Institute, who moderated the webinar panel.

Moher said a recent paper in the highly regarded journal Nature indicated more than 400,000 articles were generated by paper mills, while other research reported more than 10,000 retractions of papers.

A paper mill is a business that produces and sells fabricated or plagiarized research papers that mimic genuine scientific publications, with the aim of generating revenue through fake authorship or publication.

Predatory journals profit by charging researchers a fee to publish, which many legitimate publications also do, especially journals that are open access. Sometimes the fee is more than $10,000.

“The commercialized production of science is happening at scale through paper mills,” said Sara Elaine Eaton (photo at left), professor and research chair at the Werklund School of Education at the University of Calgary.

“And then as a result, mass retractions [are made] as a way of trying to pull back or correct the scientific record when paper mill products make it through peer review,” she said.

However, not every retraction in the scientific literature is the result of research misconduct or fraudulent behaviour, said Kaia Motter (photo at right), head of academic affairs, North America, for Springer Nature, a Germany-headquartered global academic publisher whose products include journals, book, databases and platforms.

Last year, Springer Nature received more than 3.2 million article submissions, of which just under 500,000 were published, Motter said. During that time, Springer retracted 2,923 published articles.

So the number of retractions represents only 0.6 percent of total published articles, she noted.

The problem with paper mills and predatory journals shouldn’t be underplayed, but the scientific community must be careful to not project the image that science on the whole is broken, said Juan Pablo Alperin (photo at left), associate professor in the School of Publishing and co-director of the Scholarly Communications Laboratory at Simon Fraser University.

“I think there’s an emphasis and a clear need for us to avoid creating a moral panic and us internally starting to doubt whether the processes of science and the people working from within science are actually operating on the whole with good intentions and trying to create knowledge to the best of their ability,” he said.

The research community is pretty good at being able to identify what is good research and what isn’t, Alperin said.

Paper mills and predatory journals account for only a small proportion of published articles, he added. “It doesn’t cause as much of a problem for the scientific community as the headlines would have us believe.”

“A massive, harmful and largely fraudulent industry”

However, Timothy Caulfield (photo at right), in his new book, The Certainty Illusion: What You Don’t Know and Why It Matters, calls predatory journals “a massive, harmful and largely fraudulent industry that will do nothing to advance scientific knowledge. The industry is nothing more than a cynical money grab.”

Caulfield is a professor in the Faculty of Law and the School of Public Health, and research director of the Health Law Institute at the University of Alberta.

He makes the case that the harm of articles published by predatory journals often is inflicted on the public and, by extension, policymakers.

Caulfield cites an article by a predatory journal published during the COVID-19 pandemic, in the Journal of Biological Regulators and Homeostatic Agents, claiming that 5G technology causes the spontaneous “induction” of COVID-19 in human cells. (It doesn’t).

But the paper was quickly embraced by conspiracy theorists and spread through social media and COVID-denier websites, “where it still lives on is often cited as proof of a pandemic conspiracy,” Caufield wrote.

As for the scale of the problem, he said, studies have placed the number of predatory journals at between 11,000 and 16,000. Though estimates vary, the number of articles published each year by predatory journals is likely over 750,000.

In 2022, the group Predatory Publishing, an initiative aimed at fighting this problem, estimated that almost $400 million is spent each year on the publication of research in predatory journals. Other studies have suggested the size of the predatory industry may be in the billions of dollars.

Researchers often use some of their taxpayer-funded research grants to pay for publication in predatory journals, Caulfield pointed out.

“It is, no doubt, a worldwide problem,” he said. “Unfortunately, research has also found that academics, clinicians and students are often unaware of the threat posed by predatory publishing.”

A recent study by panel moderator David Moher, co-authored by Dr. Manoj Lalu at the Ottawa Hospital Research Institute, found that over 8,000 national and international clinical public health policies referenced papers from predatory journals. And their study only looked at research published by one predatory publisher.

Publications in predatory journals can be used strategically to legitimize fringe and harmful ideas, Caufield noted.

For example, there have been predatory publications on “truly off-the-rails beliefs” like chemicals (such as the conspiracy theory that the contrails behind jet planes are part of a government plot to drug the population) and the denial of climate change and HIV/AIDS.

“Alternative medicine proponents love predatory journals as they can be used to make their science-free beliefs appear to be supported by evidence,” Caulfield said.

Robert F. Kennedy Jr., the U.S.’s new health secretary, twice during his Senate confirmation hearings mentioned an article by Anthony Mawson, an epidemiologist and a former academic who has published several papers alleging a connection between childhood vaccines and autism. (Any such connection has been thoroughly debunked).

As an article in The Atlantic magazine, by Adam Marcus and Ivan Oransky pointed out, Mawson’s most recent paper on the subject and the one to which Kennedy was referring, appeared in a journal that is not indexed by the National Library of Medicine or by any other organization that might provide it with some scientific credibility.

One leading member of the journal’s editorial board, Peter A. McCullough, an advocate for using hydroxychloroquine and ivermectin to treat COVID-19, has lost five papers to retraction. Another member is Didier Raoult, a presence on the Retraction Watch leaderboard with 31 retractions. A third, and the journal’s editor in chief, is James Lyons-Weiler, who has one retraction of his own and has called himself, in a since-deleted post on X, a friend and “close adviser to Bobby Kennedy.” 

“The scientific literature is an essential ocean of knowledge, in which floats an alarming amount of junk,” wrote Marcus and Oransky.

Publishers are increasingly using technology to help ensure research integrity

Motter said Springer Nature has more than 174,000 external editors working on the publishers’ journals and other products and 1.2 million peer reviewers.

Springer Nature has implemented new AI-powered tools to help with technical checks and quality control screening of submitted articles, as well as tools that help detect image manipulation and text generated by large language models such as ChatGPT.

“But ultimately those tools are there to provide support to our editors,” Motter said. “They are not there to make decisions on papers, so the editors still need to make a judgment call when they’ve reviewed that analysis to see if a manuscript should move forward into peer review.”

Springer Nature also recently made it mandatory that every research paper author publish a data-availability statement. A statement accompanying each article makes it clear whether or not there is data publicly available and how to link from the article to the data in a data repository.

Similarly, Springer Nature is continuously looking for ways to link the final article – the version of record – to other research artifacts that come out of a study, such as software, protocols, code and pre-prints of papers, Motter said.

Springer Nature also is using technology to provide a feedback mechanism so that if there are changes made to the version of record on the publisher’s website, then those changes are also communicated to other platforms or repositories where the content might be hosted, she said.

Of course, such feedback mechanisms don’t prevent any private websites – especially those operated by conspiracy theorists – from keeping the initial article posted, without alerting visitors to changes to the article or retractions.

Scholarly publishers are responsible for helping to mitigate known flaws in the system – which is not always transparent or equitable – of conducting research and reviewing and disseminating this information, said Natasha McDonald (photo at left), director, peer review at Ottawa-based Canadian Science Publishing.

“I think we’re also at a point where a lot of resources and technology are needed to combat fraud and manipulation of data and resources and publications,” she said.

She pointed to the STM Integrity Hub that offers services for publishers to detect research integrity-offending manuscripts.

Alperin, who’s the scientific director of the Public Knowledge Project at Simon Fraser University, said the project has been championing the use of a “publication facts label” as a way for journals to be more transparent about how they operate.

The aim of the label, which in its current form is only available for journals that use Open Journal Systems as their publishing platform, is to help readers learn more about an article’s and a journal’s adherence to the scholarly standards “that set research apart from other sources of knowledge.”

“It really is [about] helping to open up and having the people who are not from within science really understand how things work and be involved and engaged and see value in what we do,” Alperin said.

McDonald said Canadian Science Publishing does a lot of training programs, webinars and workshops for early-career researchers on topics such as best research practices, scientific integrity and the need for peer review and how the process works.

Springer Nature offers online training modules and has done early-career researcher outreach, Motter said.

However, she noted that research institutions and research funders also have a role to play in ensuring researcher and scientific integrity.

Springer Nature did a survey in 2023 asking researchers in the U.S. about access to researcher integrity training. A large percentage – 43 percent – either didn’t know there was any researcher integrity training available to them or they were pretty sure there wasn’t any training available.

“This highlights an issue at the institutional level, or maybe even the funder level, where there are some gaps in resources,” Motter said.

Institutions should make sure, with support from research funders, that researchers have access to researcher integrity training earlier in their careers and don’t just rely on their principal investigator or mentor to provide that training, she said.

Elaine Eaton agreed, noting that Canada has a “terrific blueprint” and requirements to ensure research ethics. “The [research] funding bodies could have a real crucial role to play in ensuring that researchers are trained on research integrity in the same way that we’re trained on research ethics.”

More incentives needed for peer reviewers and less focus on paper citations

Another issue for ensuring research scientific integrity is the lack of sufficient support and incentives to encourage researchers – especially early-career researchers with multiple responsibilities and limited time – to voluntarily take on peer reviewing manuscripts.

Moher said when he’s asked to peer review a manuscript, the publisher almost always asks him to turn around the review in two weeks.

In June 2023, Moher was asked to peer review 88 articles. He and colleague Anna Catharina Viera Armond at the Ottawa Hospital Research Institute published a paper about the experience in the journal Accountability in Research. Little if any of the revenue that publishers make from papers is shared directly or indirectly with peer reviewers, according to their paper.

The paper concluded that: “With almost no reciprocity in the peer review process, journals and their publishers need to promote and establish more reciprocity in a system that currently largely favours them disproportionately.”

Peer review becomes this kind of “perpetual motion machine” where researchers don’t have enough time for their reviews and they feel there’s no reward for doing them, McDonald said. “If we want that to change, we have to prioritize and incentivize peer review, and make it equally important to conducting your own research and writing your own paper.”

Caulfield pointed out in his book The Certainty Illusion that a big reason predatory journals thrive is because they exploit the intense pressure on academics to publish.

He cited a 2021 study that found that predatory publications are “authored by scholars from all fields and levels of academic experience” and that the driving force behind the industry is the “research evaluation policies and publication pressure that emerge from the research environment in which scholars operate.”

Moher referred to his own recent annual performance review where Ottawa Hospital Research Institute, like many other Canada research and academic institutions, is very focused on data. Specifically these “bibliometrics” include the number of scientific papers published, the impact factor of these papers, and the number of papers where the researcher was the senior author.

Alperin said he has served on many Social Sciences and Humanities Research Council (SSHRC) research evaluation committees where the committee is still focused on the impact factor of journals, yet SSHRC’s evaluation guidelines don’t require having such a focus. “There’s a culture that still needs to be shifted,” he said.

There’s a growing recognition by universities that research and researchers’ impact involves more than just publishing scientific papers read by academic peers, the panelists agreed.

Elaine Eaton said the University of Calgary (U of C) has an office of knowledge translation and knowledge dissemination that encourages scholars to make the outputs of their research available in different formats, such as TED talks, blogs, webinar panels and other formats aimed at popularizing science with the public.

Some Canadian universities, including the U of C, have signed the Declaration of Research Assessment (DORA) and are exploring alternative or complementary ways to assess research and its impact on community in faculty members’ annual performance reviews.

Research funding agencies in Europe and the U.K. are shifting away from using only publication-based metrics to assess the impact of research and researchers, Elaine Eaton said.

“Current research assessment methods rely heavily on publication-metrics such as citation counts, and often fail to recognize the wide array of contributions made by researchers,” according to the international Coalition for Advancing Research Assessment.

The coalition, which includes over 700 research organizations, funders, assessment authorities, professional societies and their associations, has agreed on guiding principles to implement reform in the assessment of research and research organizations.

Motter cited a survey by the Pew Research Centre that found only 43 percent of those surveyed said that scientists are good at communicating their research results.

Both the lay public and policymakers need to understand how science is conducted, including that science is iterative and that new findings can sometimes overturn assumptions of the past, she said. “We really do need to think about the way that we communicate science and change the way we communicate about science.”

Alperin said Simon Fraser University – like many other Canadian universities – wants its faculty to be engaged in responding to public policy concerns, connecting with communities, and addressing social challenges.

“Those things naturally lead to less publications, because they require a different kind of engagement and output,” he said. Yet the uptake and the impact of such initiatives “can be much larger than any number of citations on a paper.”

Added Alperin: “If we don’t start collectively understanding that our roles are more broad, then we’re not going to create a culture where publishing or working and doing different kinds of things is valued.”

R$