Exclusive: PLOS ONE to retract more than 100 papers for manipulated peer review

In March, an editor at PLOS ONE noticed something odd among a stack of agriculture manuscripts he was handling. One author had submitted at least 40 manuscripts over a 10-month period, much more than expected from any one person. 

The editor told the ethics team at the journal about the anomaly, and they started an investigation. Looking at the author lists and academic editors who managed peer review for the papers, the team found that some names kept popping up repeatedly. 

Within a month, the initial list of 50 papers under investigation expanded to more than 300 submissions received since 2020 – about 100 of them already published – with concerns about improper authorship and conflicts of interest that compromised peer review. 

“It definitely shot up big red flags for us when we started to see the number of names and their publication volumes,” Renee Hoch, managing editor of PLOS’s publication ethics team, told Retraction Watch. “This is probably our biggest case that we’ve seen in several years.”

The journal’s action on the published papers begins today, Retraction Watch has learned, with the retraction of 20 articles. Action on the rest will follow in batches about every two weeks as the editors finish their follow up work on specific papers. Corresponding authors on the papers to be retracted today who responded to our request for comment said they disagreed with the retractions, and disputed that they had relationships with the editors who handled their papers, among other protests.

While 100 retractions over a short period of time may be eye-popping, it’s also not surprising, and is a reminder that PLOS ONE has invested in expanding its research integrity team in recent years. It began issuing more retractions around 2018 as its team worked through hundreds of reports from Elisabeth Bik about papers with duplicated images, at least some of which are clearly linked to paper mills.

The case differs from typical paper mills, Hoch said. “We’re not talking about the content of the articles at all” – no duplicated images or plagiarized text – “we’re talking about how that article was processed and reviewed,” about the people involved. 

Manipulations of the peer review process, including peer review rings and fake review, which we’ve reported on since 2011, are not uncommon. Searching our database for “fake peer review” as a reason for retraction yields more than 3,000 results. 

In the present case, Hoch said, the author who submitted a paper often would request to work with a particular academic editor. At PLOS ONE, academic editors are external experts in given areas who manage peer review and decisions about manuscripts with only as much input from staff editors as they request. 

Hoch and her colleagues realized that the authors and academic editors didn’t disclose potential conflicts of interest, such as recent collaborations or working at the same institution. They also saw that the editors invited peer reviewers who also had some type of undisclosed relationship with one or more of the authors. 

“Altogether, the peer review process was compromised,” Hoch said. 

The publication ethics team’s investigation identified more than 1,700 authors involved with the submissions. The majority, about 75%, were involved only once, Hoch said. Those people may have collaborated on the research in the articles and were “likely not primarily responsible for the integrity concerns as far as we can tell.” 

A group of 41 authors and academic editors were involved with 10 or more of the submissions. One author was listed on more than 30 papers.  A single academic editor was requested to manage a quarter of the submissions, and ended up managing peer review for more than 30. 

About half of the people involved with at least 10 submissions were affiliated with institutions in Pakistan, the investigation found, and the next most-represented countries were China, Turkey, Saudi Arabia and India. 

About 75% of the academic editors the journal identified as having been involved in the series had joined the editorial board since 2020, when the implicated submissions began coming in, and had co-published with each other, sometimes extensively, Hoch said. 

The investigation also identified authorship concerns, including cases in which authors’ declared contributions did not meet ICMJE criteria, and several authors with suspiciously high publication volumes. 

Although the journal doesn’t know exactly what happened behind the scenes with the submissions, the relationships between authors and editors the investigation found “paints a picture of a sort of network working together,” Hoch said. She speculated that a group of people may have offered services to help authors with publishing papers, including shepherding articles through peer review. 

“If you look at any single article, even any two articles, the issue might not seem as severe, but when you look at a group of 100 or 300 it’s really very concerning,” she said. 

In addition to taking action on the published articles, the journal is changing some of its submission and pre-publication processes, such as additional screening for competing interests between authors and academic editors. Hoch declined to share specifics about some of the authorship issues the team identified and subsequent workflow changes, to prevent people from circumventing the new processes. The journal also will be evaluating the academic editors involved on a case-by-case basis, and may give them feedback or dismiss them from the editorial board. 

“We’re very concerned that our editorial board, whom the journal has such a strong reliance on, seems to have been involved in this case for peer review manipulation, whatever the mechanism,” Hoch said. A “miniscule fraction” of the overall editorial board was involved, but “we take that vulnerability very, very seriously. It has really important implications for the journal overall.”

We received a list of the 20 papers being retracted later today and a sample of the retraction notices, which will state: 

The PLOS ONE Editors retract this article because it was identified as one of a series of submissions for which we have concerns about authorship, competing interests, and peer review. We regret that the issues were not addressed prior to the article’s publication.

We emailed the corresponding authors and academic editors who were named on at least two articles to ask for their comments. The three who responded to our emails disputed the retractions. 

Yunzhou Li, a corresponding author on four of the papers who is affiliated with Guizhou University in China, sent us a comment addressed to “Plos One editor”: 

I strongly disagree with the retraction, first of all, I don’t know the reason for the retraction. Second, all data and review processes are conducted with strict security procedures without any violations. Third, it will affect the reputation and credibility of your journals, just like we have already paid, and you have returned the goods, which will seriously affect the credibility of your journals, I strongly disagree. If any journal could freely withdraw an author’s manuscript, that journal would also die.

Subhan Danish, a corresponding author for two papers who is affiliated with Bahauddin Zakariya University in Pakistan, told us: 

I totally disagreed with one side decision. We don’t suggest any reviewer. It was decision of editor. How can we be biased for reviewers.

Secondly we suggest reviewer from field of experts. My whole team in the article did not have a single paper with editor Farooq Shahid.

When we don’t know even editor and reviewer how can we be guilty. Career of many scientists are now on risk. Our paper was also good. But we are suffering from a crime which we have not done.

On humanity basis I think suggestion of any reviewer who even don’t Handel our article is not a crime.

I request you please be polite in your decision as neither I nor my team know editor or reviewer in this paper. 

It is journal who decide them. I shall appeal you please don’t spoil young scientists career at out early stages.

Muhammad Hamzah Saleem, a corresponding author of two papers affiliated with Huazhong Agricultural University in China, said: 

First of all, I want to mention it that this paper is published which everyone know.

Now, you have mentioned two issues in the last Email for retraction.

1) Academic editor is our friend

2) paper found to be similar with the other submitted paper.

I want to clarify, your two points one by one.

1). The academic editor (Saqib Bashir) is not our friend and non of our author from previous submitted paper. In addition, when we submitted the paper we have chosen two different academic editors. And you did not selected them. Even we did not know about the academic editor through the entire process and when paper was published then we can see his name on the website that this editor was handled the paper. In summary, We have selected a different academic editor for this paper, and Saqib Bashir was not chosen by us, it’s from PloS staff.

2). The paper is not similar with any other submitted paper. We have checked carefully and submitted our paper very attentionly. So, please do not erase this point. As this is the different paper from all other submitted paper.

Hope you understand. 

And we do not wish to retracted any of our paper.

From her experience working on various other cases involving multiple manuscripts, and reading about paper mills on Retraction Watch and other sources, Hoch said this case shows publishers need different types of assessments to detect systematic manipulation of the publishing process that doesn’t show up in the content of articles, and cross industry collaboration could help.

“We have seen them change and evolve over time,” she said of paper mills. “They’re getting more and more difficult to detect.”

Like Retraction Watch? You can make a tax-deductible contribution to support our work, follow us on Twitter, like us on Facebook, add us to your RSS reader, or subscribe to our daily digest. If you find a retraction that’s not in our database, you can let us know here. For comments or feedback, email us at [email protected].