ABUSE OF ONLINE SUBMISSION SYSTEMS

One of the most serious and recent reasons for retractions pertains to the abuse of online submission systems (OSSs). OSSs exist, in fact, to facilitate the management of manuscripts, authors, peers and editors by journals and publishers. In theory, the system is based on the assumption that it will not be abused. Yet, some high profile cases of some main-stream science, technology and medicine (STM) publishers (Informa Health, SAGE, Landes Bioscience, Elsevier, Wolters Kluwer, BioMed Central / SpringerNature), exemplified in more detail below, indicate that OSSs are not only highly fallible, but also apparently easily subject to abuse. Some of the ways in which the system can be abused are through the creation of false accounts, using either false names or aliases, false e-mails or emails that do not reflect those of the actual authors or the use of false identities to feign peer status and, thus, manipulate the result of the peer review for a favorable outcome. These would be the most obvious cases of abuse of OSSs to derive academic advantage over competing authors.

Alarm bells initially began to ring in 2012 in the plant academic community when it was learned that a South Korean researcher, Hyung-In Moon (Department of Medicinal Biotechnology, College of Nature Resources and Life Science, Dongguk University, Busan), had created fake e-mail addresses to complete his own favorable "peer" reviews and, thus, favor the publication of his papers ¹. This led to at least 35 retractions, all in Informa Health journals (Journal of Enzyme Inhibition and Medicinal Chemistry, Pharmaceutical Biology, International Journal of Food Sciences and Nutrition, Enzyme Inhibition and Medicinal Chemistry, Immunopharmacology and Immunotoxicology) ^2,3. The Editor in Chief of Immunopharmacology and Immunotoxicology, Prof. Emilio Jirillo, subsequently resigned ⁴. That case had been preceded by a case of abuse of Elsevier's OSS, Elsevier Editorial System (EES), by a Chinese researcher (Guang-Zhi He, Guiyang College of Traditional Chinese Medicine) who had "fabricated information during the review process to obtain a favorable review" in Experimental Parasitology^5,6. Prior to that case, China had come under the spotlight when researchers at Huazhong University of Science and Technology in Hubei created, among other fraudulent actions, a fake email address for one of the authors. It was meant specifically to be "used by the authors to intercept any information that would be sent to the corresponding author," thus gaining a favorable and unfair advantage for a paper submitted to Cancer Biology & Therapy, published by now-defunct Landes Bioscience ⁷.

SAGE, a UK-based publisher, became a victim of major author abuse of its OSS in 2014, retracting 60 papers from volumes 16, 17, 18, 19 and 20 of the Journal of Vibration and Control^8,9 after its OSS, SAGE Track powered by ScholarOne Manuscripts(tm), was abused by one author, Peter Chen, formerly of the National Pingtung University of Education, Taiwan. In this case, covered more extensively at Retraction Watch ¹⁰, not only had assumed and fabricated identities been used to manipulate the OSS, using as many as 130 fake e-mail addresses, so too was this burst of self-reviewed and self-approved manuscripts used to manipulate the author's own citations; i.e., the establishment of a citation ring. In addition to the retraction of these 60 papers, the then Minister of Education of Taiwan, Wei-ling Chiang, and one of the authors of several of those papers, resigned, as did Peter Chen ¹¹.

Elsevier once again tasted the bitterness of author-based submission abuse when Khalid Zaman, of COMSATS Information Technology Center in Abbottabad, Pakistan, abused EES to submit his own faked peer reviews for three journals: Economic Modelling, Renewable Energy, and Renewable and Sustainable Energy Reviews, resulting in 16 retractions ¹².

Wolters Kluwer's Medicine became the next victim of abusive Chinese scientists from the Third Military Medical University, who created a fictitious account and submitted a review "under the name of a known scientist without his/her knowledge," leading to the retraction of two papers ¹³.

In another twist emerging from the use of fake e-mail addresses or OSS accounts to complete fake peer reviews, 43 papers were retracted from BioMed Central (BMC) (i.e., at that time Springer Science + Business Medium, now SpringerNature), following the use of third-party companies that sell such services ¹⁴. In addition, BMC temporarily shut down the author-suggested reviewer option of the OSSs of their 250+ journal fleet. Even though 43 papers were officially reported, entering the term "because the peer-review process was inappropriately influenced and compromised" into BMC's search function reveals 86 items, while "The Publisher and Editor regretfully retract this article" reveals 83 items, suggesting the number of retracted papers related to the abuse of the OSS may have exceeded the original tally advertised.

There are several loop-holes in the OSSs of publishers, including in the security functions of ScholarOne, allowing OSSs to be abused ¹⁵. In part, the request by journals for authors to suggest potential reviewers is, in itself, open to bias and abuse, even though it is ultimately the authors who must assume collective responsibility for those individuals they have suggested as peers. There are also serious concerns that third party companies might be selling such services, including fake peer reviews, false e-mail accounts and other services that fraudulently help authors to increase their chances of having a paper published, as appears to be taking place in China ¹⁶. These cases further accentuate the weaknesses of the traditional publishing system used by many publishers ¹⁷. This paper covers cases reported by Retraction Watch until March 2015, and newer cases will be covered separately elsewhere.

NEW AND NUANCED ABUSES OF ONLINE SUBMISSION SYSTEMS

Even though the Bohannon "sting" ¹⁸, in which Bohannon submitted dozens of open access journals to unsuspecting journals, proved that peer review in many of them was lax, or even non-existent, accepting nonsensical manuscripts, there were some underlying ethical abuses of that sting: a) false names and e-mail addresses were used; b) false institutional addresses were created in real countries; c) false declarations of originality were made upon submission; d) guarantees that all information was accurate were false and misleading; and e) false e-mails using false dialogue were used to mislead journal editors. Even though, ultimately, the "sting" proved what was already known about the predatory operations of many open access journals, and even though it helped to widely spread consciousness among academics, the Bohannon sting was deeply entrenched in lack of publishing ethics, with wide-spread abuses of OSSs. What remains unclear is why such widespread lack of publishing ethics protocol has not merited the retraction of that paper ¹⁹.

There is one more issue that is not being discussed or criticized anywhere, not in the literature, nor on blogs; namely, the forced creation of accounts on OSSs by editors or publishers, without the explicit permission of the individual(s) for whom an account has been created. There is no doubt that such a topic will elicit a wide range of responses, from the "this is a non-issue" to the other end of the opinion spectrum, at which concern is raised about the lack of ethics or protocol in the creation of such OSS-based accounts. Appendix 1 shows two examples of journals that created OSS accounts on my behalf during April 2015. This is a phenomenon that has happened to me about a dozen times, but only two recent examples are highlighted, since complaints to the same publishers regarding the previous unapproved creation of OSS accounts fell on deaf ears. The core question here is: should editors and publishers create accounts on behalf of authors? In most cases, such accounts are created to establish a platform on which the invitee then proceeds with the peer review of a manuscript he/she has been invited to review. The principle, in itself, is acceptable, but the order of events is not. An editor or publisher has the responsibility of first inviting an author to a peer review, indicating that to do so they will need to create an OSS account. Secondly, explicit permission must be obtained from that individual to create that account. Once formal approval has been obtained, both to review a manuscript and to create an OSS account, the publisher is then welcome to create it. Once such an OSS account has been created, then all subsequent invitations to peer review can be automatic. The current problem, as exemplified in Appendix 1, is that no permission is sought before an OSS account is created. As I see it, this is both unethical and a breach of privacy. Such over-reach by editors is, in addition to the instances indicated above, contributing to a gradual corruption of the submission process and, thus, to the erosion of trust in the peer review system overall.

ARE THERE ANY REALISTIC SOLUTIONS?

I propose three simple, yet practical and most likely effective solutions, if implemented:

a. Sting publications must end. There cannot be a dichotomy of publishing ethics, allowing some to use unethical means to achieve success, or show a point (e.g., Bohannon ¹⁸⁾⁾, while the remainder of the scientific base has to conform to another set of publishing ethics. There are more examples of stings in ^20,21.

b. Editors and journals/publishers must not request authors to recommend peer reviewers. This task must be the exclusive responsibility of the editorial board.

c. Editors and publishers should be allowed to create an OSS account only following explicit approval from the invitee.