Research integrity: trending topic – paper mills
This section draws on:
the relevant web pages of COPE concerning the paper mill phenomenon. https://publicationethics.org/publishers-perspective-paper-mills + related resources as mentioned on this page.
Current topics are often new and knowledge is subject to further development or deepening. This tip was prepared on the basis of the relevant knowledge as at February 2022.
What is a paper mill?
Paper mills are profit oriented, unofficial and potentially illegal organisations that produce and sell products ranging from research data through to ghost written fraudulent or fabricated manuscripts and submission services for review. They may also sell authorship to researchers once the article is accepted for publication.
What integrity issues arise with paper mills?
In essence, there are several issues related to paper mills. One is that 'products' are delivered to people who did not do the work. You buy an article as if it were your own, but you did not write it yourself; you get a place on the author's list for which you did not contribute.
Although the work itself isn’t always a fake (but it can be), the personal attribution to an individual (the buyer) most certainly is fake.
Some of the paper mills might have actual laboratories that perform experiments and produce actual data and images. Data might then be real, it is still wrongfully used, for example by using the images in different non related studies.
But in many cases also parts of the work, images used, etc. appear to be reasonably real but are in fact created fraudulently or manipulated.
There can also be issues of plagiarism, often not detected because the lines come from a translated version of another article.
How to recognise a paper coming from paper mills – red flags?
There are some features that seem to recur when it comes to paper mills. Of course, the presence of (one of) these characteristics is no guarantee of its fraudulent origin, but caution and extra scrutiny may be required.
Issues with data and its visualisation:
• Often underlying data is being send when asked, but the data consist of broad datasets, almost always badly managed, labelled or incomplete and the data do not completely match the types of experiments described in the methods.
• The data requires specialised knowledge or skills, either to open the datasets (specialised software) or to read (specialised knowledge or expertise to make sense of it).
• Often about small non-coding RNAs (micro RNAs or circular RNAs)
• usually have a particular set of photos and types of experiments such as western blots, flow cytometry and microscopy.
Issues with authorship:
• Often authors with private or non-institutional email addresses, or email addresses that do not appear to be linked to any of the names of the authors – which makes it hard to contact and check the actual situation as you never know who is at the other end of the mail. It could be a paper mill agent using all and different addresses.
• Often no reference is being made to funders
• Extremely high number of changes in authorship, often over several rounds of peer review, even after the article has been accepted for publication
Issues with content:
• Incorrect nucleotide sequences and often the reagents described in these papers are wrong
• Description of highly generic study hypotheses and experimental approaches
• Follow a set template having unusual similarity of text
• Often in the cancer field
Individually, these factors may not be problematic but taken together they often raise concerns and could be part of a pattern. This is also the reason why paper mills are difficult to spot, the pattern only becomes visible once there is a bulk of published papers that can be compared, often across publishers.
How to deal with paper mills as a peer reviewer or editor?
It goes without saying that everyone involved in the process of publishing scientific articles should perform their task in good conscience, meaning that they not only take the necessary time and care to approach the content and data of a manuscript scientifically, but they can and must also play a role in reducing harmful practices, in whatever form.
This means that both editors and peer reviewers have a responsibility in trying to prevent and expose paper mills.
Editors and peer reviewers have the responsibility to inform themselves about the phenomenon of paper mills and to keep this knowledge up-to-date or deepen it. This should ideally happen in a more structural way e.g. by explicitly mentioning paper mills in the journal policy or in the guidelines for peer review. Smaller magazines can make use of expertise and provisions developed by the bigger players in the field.
It can also be valuable to contribute to the knowledge generation on this phenomena and its approach, e.g. by putting the topic on the agenda of journal meetings, asking questions to the publisher board, sharing experiences on (disciplinary) conferences, etc. In this way, a community of practice is established that allows for faster and more informal learning from each other, also when being confronted with individual cases. This approach is even more useful as paper mills are hard to detect when one only sees an individual manuscript. Sharing experience can help detect patterns quicker, which may inhibit the wide-ranging operation of these companies.
We’ve already pointed to some red flags (see above ‘how to recognise a paper coming from paper mills’) that might indicate an origin coming from paper mills. Editors and peer reviewers have a responsibility to be attentive to these signals and to take proactive action to further investigate doubtful cases or to call for help where one is insufficiently skilled to do so oneself. Again, a more structural approach can be of help to ensure attention is being given to the topic of paper mills, e.g. explicitly mentioning possible red flags in the evaluation sheet of a manuscript.
To guide editors and peer reviewers in a good understanding of systematic manipulation of the publication process and help them deal with it in practice, the Committee on Publication Ethics (COPE) has made a guidance document, including step-by-step flowcharts to investigate: https://publicationethics.org/files/publication-process-manipulation-cope-flowchart.pdf. We strongly recommend this document.
Submitted manuscripts are processed according to the due diligence principle, which prescribes that due care must be taken. But even with all these precautions in place, the possibility remains that editors and peer reviewers fail to notice the fraudulent nature of a contribution, or do so too late. It is therefore important to remain alert even after a manuscript has been accepted for publication, in recognising patterns with later submitted manuscripts but also in taking responsibility when questions are raised or late signs of fraud or deceit become apparent. Restoring the scientific record must always be the highest priority.
Last modified Feb. 22, 2022, 6:53 p.m.