INTRODUCTION
The term ‘predatory journals’ was coined by Jeffrey Beall in 2008, who also created a list of potential predatory journals and publishers.1 Although there is not an agreed definition of the term ‘predatory journal’,2 it could be assumed that these are open access journals that publish poor quality articles, with poor or no peer-review process, owned by publishers providing no transparent editorial services. Their main objective is financial gain by article processing charges to authors.
The total number of articles published by some 8,000 predatory journals rose from 53,000 in 2010 to 400,000 in 2014.3 This was accompanied by an increasing interest on this subject. From 2012 to 2017, the number of articles mentioning predatory journals in five bibliographic databases rose from 5 to 140, respectively, totaling 324.2 Although most predatory journals are located in developing countries, notably India and Turkey, many are edited in the USA and other western countries.4 The use of predatory journals has spread all over the world: researchers from 146 countries (out of 193 countries belonging to United Nations) have published in predatory journals.4 This is particularly important in Europe where the implementation of Plan S in 2020 will increase the percentage of research published to be immediately open access:5 investigators must know how to distinguish scholarly journals from predatory journals.
Although there are organizations dealing with the ethics and quality of scholarly publishing, such as COPE (Committee of Publication Ethics), ICMJE (International Committee of Medical Journal Editors) or WAME (World Association of Medical Editors), predatory journals pose serious issues to academic journals.6 In their aim to gain more prestige among researchers, many predatory journals claim to be members (or followers) of respectful organizations such as ICMJE. This is why the ICMJE has expressed its concerns about predatory journals using the list of ICMJE Recommendations (ICJME-R) followers to “gain the appearance of legitimacy.”7
The aim of this study was to assess the current presence of potential predatory journals on the ICJME-R list and their theoretical adherence to ICMJE-R.
MATERIALS AND METHODS
We chose a random sample of 350 journals from the estimated 3,100-3,200 biomedical or health-care journals listed as ICJME-R followers in February 2018 (a journal listed as an ICJME-R follower claims to adhere to the ICJME recommendations).8 Data collected from the ICMJE and journal websites in English included: adherence to the six main ICJME-R polices/requirements, year of journal’s listing as ICJME-R follower, discipline covered, publisher and its country of origin, and existence of article processing charge. The ICJME-R policies (or statements) were those referring to authorship, author’s conflict of interest and plagiarism; whereas the ICJME-R requirements were on participant’s informed consent, research ethics committee approval and clinical trial registration. Following the well-respected educational initiative ‘ThinkCheckSubmit’, potential predatory journals were considered those not being members of a recognized industry initiative, such as COPE, DOAJ (Directory of Open Access Journals), OASPA (Open Access Scholarly Publishers’ Association), AJOL (African Journals Online) or INASP (International Network for the Availability of Scientific Publications).9 As others have done before,10,11 we checked the inclusion of both the journal and publisher on the updated Beall lists.12
RESULTS
This analysis revealed that 31% (108/350) of journals had characteristics of potential predatory journals. Table 1 shows that most of them were included in the ICJME-R list of followers in the last four years (94%; 102/108). In four years, the annual number of new followers increased 120% from 15 (2014) to 33 (2017). Half (54/108) were published by publishers in the USA and 62% (67/108) were devoted to medicine. Adherence to five of the main policies and requirements considered was scarce, ranging from 51% (plagiarism) to 7% (trial registration). The policy on authors’ conflicts of interest was the only commonly (72%) mentioned policy. Only three journals stated that they followed all six policies and requirements, and 11 (10%) had no public evidence of following these policies. Information on an article processing charge was publicly available for 82 (76%) journals, could not be found for 24 (22%) and two journals specifically stated that there was no article processing charge.
Table 2 shows the comparison between American and Indian journals. Authorship policies (or instructions) were significantly more present in journals with publishers from India than from USA (53% vs 30%; p = 0.047), with no differences in the other five policies and requirements. Eighty percent (86/108) of potential predatory journals were included in the up-dated Beaall’s lists of potential predatory publishers or journals.12
DISCUSSION
Our study provides evidence that many potential predatory journals may indeed be gaining legitimacy by being included as ICJME-R followers and that this is a recent phenomenon. Although Beall considered 2012 to be the year when predatory publishers exploded,1 our results show that potential predatory journals needed two more years to start the race to list themselves as followers of the ICJME-R, reaching a maximum of 31% (33 of 108) of new followers in 2017. Potential predatory journals are also colonizing other databases to gain respectfulness. Hence, PubMed includes articles published by potential predatory journals and the percentage of potential predatory journals increased significantly in only one year. Thus, in 2016, between 11% and 20% of PubMed journals in rehabilitation, neuroscience and neurology were potentially predatory journals, whereas in 2017 these percentages rose to 16%-25%.13
There were two limitations to our study. The first is that among the elements that ‘ThinkCheckSubmit’ advises to check to assess if a journal could be potentially predatory, we checked only the three that were objective and feasible – the article processing charge, easily identifiable publisher and journal being a member of a recognized industry initiative – and we left out those being subjective and non-feasible, such as knowledge of colleagues about the journal, having a recognized editorial board or having articles indexed. For 22 journals, we were not able to identify the article processing charges. However, it is well known that many predatory journals only inform on the fees to be paid once the article has been accepted for publication.1,2 Finally, two journals explicitly stated that they will not charge any article processing fee; however, both journals and publisher were not included in any of the five recognized industry initiatives9 and both journals belonged to a publisher (AME Publishing Company, Hong Kong) that was included in the Beall list of potential predatory publishers.12 The second limitation was that we did not check the accuracy of the six policies and requirements since all, except that referring to authorship policy, can only be checked by submitting a manuscript. This is why we always refer to ‘potential’ predatory journals.
Publishing in predatory journals is unethical.11 Potential predatory journals on the list of ICMJE-R followers do not provide public evidence that they actually adhere to ICMJE-R, so it is questionable whether ICMJE should keep this list. They should be deleted from the ICMJE-R list of followers to prevent misleading authors. ICMJE-R followers need to be reevaluated with pre-defined published criteria, similar to the procedure undertaken by DOAJ and OASPA, and these quality checks should be applied to all future applications. A similar approach has been suggested to ensure that PubMed is free of predatory journal articles: journal candidates should satisfy the three MEDLINE preapplication requirements and should be a member of DOAJ, OASPA, COPE or WAME.14 Finally, a third way to address this scientific publishing problem – of special relevance to biomedicine, the topic of interest to most predatory journals10– is to generate a list of respectful journals. This has been the approach taken by urologists who are creating a ‘green list’ of reputable journals within their specialty.15 As of December 2018 there were 57 journals included in the ‘Urology green list’, all of them complying with several criteria such as, for instance, being a member of a professional organization, having a reputable publisher and editorial board, transparent manuscript submission and peer review process or membership or affiliation with COPE.16
SUPPLEMENTAL FILE
All the data collected for this study are available from the corresponding author, and is available upon request from the corresponding author.
DISCLOSURES
All authors declare no conflicts of interest. No funding or financial support was received.
REFERENCES