GLOBAL
The research community around the world needs to come together and pool both material and intellectual resources to fight the growing menace of research paper mills, in order to protect the integrity and value of research, the World Conference on Research Integrity heard this week.
There are a number of international initiatives committed to fighting the vice, but there are not enough and they are insufficiently resourced to conduct a sustained global campaign against paper mills.
As a result, campaigns to create awareness about paper mills are inadequate and the technical capacity needed by the relevant actors – including journal editors, reviewers, funders, researchers, and higher learning and research institutions – is not enough to win this ‘war’.
The same situation persists with regard to the availability of modern tools to help detect research manuscripts emanating from paper mills, said speakers at the 8th World Conference on Research Integrity held in Athens, Greece, from 2 to 5 June.
The biennial, hybrid WCRIs bring together experts from across a range of fields to discuss developments in and challenges to research integrity and to produce a policy statement. The Athens Statement, to be finalised later this year, is on the theme of “catalysing the translation of research into trustworthy policy and innovation”.
Scale of the problem
Statistics on the size of the research sector’s fake paper problem is as murky as the paper mills themselves. Nature reported last November that an unpublished analysis suggested “there are hundreds of thousands of bogus ‘paper mill’ articles lurking in the literature”.
“Paper mills rely on the desperation of researchers – often young, often overworked, often on the peripheries of academia struggling to overcome the high obstacles to entry, to fuel their business model”, wrote Lex Bouter, professor of methodology and integrity at Vrije Universiteit Amsterdam, in an article in The Conversation recently.
“They are frighteningly successful,” said Bouter, who is also founding chair of the World Conferences on Research Integrity Foundation and co-chair of the 8th WCRI in Athens.
One paper mill in Latvia advertised the publication of more than 12,650 articles since its launch in 2012. “In an analysis of just two journals jointly conducted by the Committee on Publications Ethics [COPE] and the International Association of Scientific, Technical and Medical Publishers [STM], more than half of the 3,440 article submissions over a two-year period were found to be fake,” wrote Bouter.
“It is estimated that all journals, irrespective of discipline, experience a steeply rising number of fake paper submissions. Currently the rate is about 2%. That may sound small. But, given the large and growing amount of scholarly publications it means that a lot of fake papers are published. Each of these can seriously damage patients, society or nature when applied in practice,” said Bouter.
Paper mills are described as the process by which manufactured manuscripts are submitted to a journal for a fee on behalf of researchers with the purpose of providing easy publication for them, or to offer authorship for sale.
The COPE and STM paper defines a paper mill as “generally a commercial enterprise”, with some being both “sizeable and highly professional”.
Papers mills are organised international crime syndicates, and they tend to target non-English language speakers. Russian and Iranian speakers have emerged as some of the main consumers of papers published by the syndicates, said Dr Lisa Parker, a bioethicist and honorary senior lecturer at the University of Sydney in Australia, in a WCRI session titled “Addressing the challenge of paper mills through research and policy”.
Consumers of paper mills sometimes include senior researchers who unwittingly fall for the schemes of the scholarly fraudsters, after seeing advertisements by paper mills, Parker noted.
Paper mills usually go for popular subjects such as health and life sciences, knowing that these fields are likely to attract readers from far and wide. They identify journals that they approach to publish their work complete with fake peer reviewers and authors and get paying customers.
How editors can catch paper fraudsters
For these reasons, editors need to be more vigilant, acting as detectives and always being on lookout for “suspect patterns”, including for instance a spike in similar submissions including topics covered and the country of origin, Parker advised.
“In addition, they should be on the lookout for repeated papers appearing with different titles, as well as those that appear to be linked, while checking the peer review process,” she said.
This can be done by scrutinising submissions that have been reviewed by the same reviewers, checking the details of authors, including their email contacts, and so on. Equally important to confirm, besides repeated similar topics with same authors and reviewers, is the presence of similar formats as well as repeated keywords.
“In screening the signs of a paper mill, always look at the authors and process of authorship, did they for example offer multiple papers? Some fakers submit papers but do not respond to emails regarding the peer review process and provide no cover letters,” said Parker.
Research has shown that text anomalies, ‘tutored phrases’, image anomalies or manipulation, and ‘variable writing style’ are possible red flags for fraudulent works.
While investigating suspicious papers might take years to crack, whenever editors have reasons to question a paper, they should not hesitate to publish an editor’s note, and when sure the paper is suspect, issue a retraction notice.
Parker named some of the challenges to detecting and stopping submissions emanating from paper mills, including lack of training for editors, competition in the academic publishing industry, legal obstacles and fear of damaged reputation. The use of artificial intelligence also makes detection difficult.
Adopting open access review despite its pros and cons, educating editors, creating more awareness about funding mills, funding watchdogs and developing automated tools to help in detection are some of the solutions that Parker suggested should be explored. Institutional review of papers before researchers submit them for publication could also help.
“Paper mills are in a race to avoid detection, hence the need to increase support to editors to help them detect fake papers,” Parker stressed.
Paper mill watchdogs
One watchdog initiative is United2Act, created by COPE and STM to principally address and do research on paper mills, and facilitate and support dialogue between stakeholders about the manipulation of the publishing process.
It was formed in 2022 by stakeholders from 15 countries, representing research bodies, publishers, researchers and universities among others, with the funding ending in 2025.
United2Act consists of five working groups determined to fight the large-scale, manipulative research mills. Each working group is undertaking different activities, in recognition of the fact that a lot needs to be done to thwart the mills.
“We are united, and we are acting. We are not one of those groups of people who talk and do nothing,” co-chair of the United2Act steering group Deborah Kahn told the WCRI 2024 session on 4 June.
One of the activities involves developing communication content tailor-made for different audiences and conducting training for research industry actors.
United2Act is also doing research to fill the knowledge gap on paper mills. There is not much research on this pressing problem, and more should be encouraged to reveal the full extent of the phenomenon.
Another working group facilitates dialogue between various research sector actors, enhancing communication between stakeholders by fostering a shared understanding of the problem.
“Paper mills are a growing problem for the whole of the scholarly research world therefore a collaborative and sustained effort is needed,” Kahn noted.
The watchdog is also advocating developing ‘trust markers’, generally defined as explicit statements on research papers regarding aspects such as funding acknowledgement, data availability, conflict of interest statements, author contributions and ethical approval, and other measures to end research fraud.
This could help stop the papers before they get into the review process, Kahn added. “We will work to put in place strong measures so that excellent research is at all times valued and protected,” she said.
According to Thomas Stoeger, an assistant professor at Northwestern University in the United States, one major characteristic of academic publishing fraudsters is ‘journal hopping’. Perpetrators of fraud move from journal to journal to avoid detection.
He also warned that a rapid increase in the number of publications targeting a certain subject should raise the antennae of integrity detectives, and cited one as being the popular subject of ageing in the United States.
Another solution Stoeger suggested is ‘de-indexing’ publications – the practice of removing a webpage or website from search engine result pages. With paper mills not being easy to detect, other solutions may lie in training reviewers in detection and “throwing more money” at the effort, he said.