This is an open-access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in the Journal of Medical Internet Research, is properly cited. The complete bibliographic information, a link to the original publication on http://www.jmir.org/, as well as this copyright and license information must be included.
Multiple sclerosis (MS), a progressive demyelinating disease of the brain and spinal cord, is the leading cause of nontraumatic neurological damage in young adults. Canada has one of the highest reported incidents of MS, with estimates between 55 and 240 per 100,000 individuals. Between 2009 and 2014, the MS Society of Canada provided over Can $90 million to researchers and, since 2013, has encouraged researchers to make both current and previous research products openly available.
The goal of the study was to determine the open access (OA) cost implications and repository policies of journals frequently used by a sample of MS researchers. This study benchmarked current publishing preferences by MS Society of Canada researchers by examining the OA full-text availability of journal articles written by researchers funded between 2009 and 2014.
Researchers were identified from the 2009 to 2014 annual MS Society of Canada Research Summaries. Articles were identified through searches in Web of Science, Scopus, Medline and Embase (both via OVID). Journal level analysis included comparison of OA policies, including article processing charges (APCs) and repository policies. Data were analyzed using descriptive statistics.
There were 758 articles analyzed in this study, of which 288 (38.0%) were OA articles. The majority of authors were still relying on journal policies for deposit in PubMed Central or availability on publisher websites for OA. Gold OA journals accounted for 10.2% of the journals in this study and were associated with significantly lower APCs (US $1900) than in hybrid journals (US $3000). Review of the journal self-archiving options highlighted the complexity of stipulations that authors would have to navigate to legally deposit a version of their article.
This study found that there are currently researcher- and publisher-imposed barriers to both the gold and green roads to OA. These results provide a current benchmark against which efforts to enhance openness can be measured and can serve as a reference point in future assessments of the impact of OA policies within this field. With funding agencies worldwide releasing OA mandates, future success in compliance will require changes to how researchers and publishers approach production and dissemination of research.
Multiple sclerosis (MS), a progressive demyelinating disease of the brain and spinal cord, is the leading cause of nontraumatic neurological damage in young adults [
The MS Society of Canada has strongly encouraged open access (OA) and the broad dissemination of research. The MS Society of Canada’s OA policy, which came into effect in July 2013, requires grant recipients to make every effort to ensure their peer-reviewed publications are available OA within 6 months of publication [
The information needs and information-seeking behaviors of clinicians and patients with MS have been well documented [
Using the MS Society of Canada as a case study, this paper examines the OA full-text availability of scientific articles by Canadian researchers funded between 2009 and 2014, including the publishing venues and associated costs of making work openly available. The goal of the study was to determine the extent of OA, the cost implications, and repository policies of journals frequently used by MS researchers as an assessment of the current context which can be used in future evaluations of the extent of openness and the effectiveness of mandates in making research OA.
Researchers were identified from the 2009 to 2014 annual MS Society of Canada Research Summaries [
The results from the database searches were merged into a single, deduplicated file of articles. A secondary quality check was performed by manually reviewing the article details to ensure relevance and accuracy. The article title was searched in Google and PubMed to determine if the full-text was openly accessible (open access) either through the publisher website or PubMed Central. The journal policy for access through PubMed Central and the version of the article on PubMed Central (author manuscript to meet compliance with a funder policy or a journal OA policy) were recorded. Full-text access through other sources such as Academia.edu or ResearchGate was not included because these sites do not necessarily guarantee that the full-text option is a legitimate copy.
Subscription policies were used to identify the title as a pure gold journal or a hybrid journal. The self-archiving policies for each journal were obtained from SHERPA RoMEO, a UK academic supported database that provides information regarding copyright policies and rights retained by the authors when publishing in specific academic journals. SHERPA RoMEO uses colors to categorize publishers’ archiving policies: white (archiving is not formally supported), yellow (authors can archive preprint, that is, prerefereeing), blue (authors can archive postprint, that is, final draft postrefereeing, or publisher's version), or green (authors can archive preprint and postprint, or publisher's version). Each journal in the study was assigned one of the 4 colors. The OA policies were analyzed first by publisher, per journal, and then at the article level.
For each journal, APC policies were searched on the publisher website. APC costs were also converted to US dollars. The maximum APC fee quoted by the journal was recorded where there were variable fees listed. APCs fees were variable depending on the type of article, memberships, institutional affiliations, funding bodies, or which Creative Commons license was selected.
The MS Society of Canada funded 77 Canadian researchers between 2009 and 2014 in one of their four major research areas. This study identified a total of 758 MS related articles in 211 journals produced by 61 publishers. Of the 758 articles, 288 (38.0%) were OA articles produced in 93 journals by 42 publishers.
When article distribution per journal was analyzed to identify the preferred or most popular publishing venues, there was a noticeable difference in the dominance of publishers. Three major publishers (Elsevier, Wiley, and Springer Nature) dominated the research output, with 105 of 211 journals (49.7%) and 374 of 758 articles (49.3%). However, these publishers only accounted for 93 of the 288 OA articles (32.2%) in the study (
A total of 19 journals from 13 publishers accounted for 388 of the 758 articles (51.2%) and 146 of the 288 OA articles (50.7%) in the study (
The most frequently used journals, accounting for 51.2% of the articles. Number of articles, number of open access (OA) articles, article processing charges (APC), and SHERPA RoMEO color are provided.
Journal title | Articles (n) | OA articles (n) | Maximum APC (US $) | SHERPA RoMEO color |
76 | 15 | 3000 | Green | |
66 | 35 | 3100 | Yellow | |
23 | 0 | 2500 | Green | |
23 | 4 | 3000 | Yellow | |
19 | 2 | 3000 | Yellow | |
18 | 2 | 2500 | Green | |
18 | 2 | 2500 | Green | |
17 | 7 | 3000 | Green | |
16 | 14 | 3000 | Blue | |
16 | 0 | 5000 | Green | |
14 | 14 | 1350 | Green | |
13 | 0 | 2000 | Green | |
13 | 2 | 2200 | Green | |
10 | 7 | 2822 | Green | |
10 | 10 | 2820 | Yellow | |
9 | 9 | n/a | Ungraded | |
9 | 6 | 3250 | Green | |
9 | 7 | 3200 | Yellow | |
9 | 8 | n/a | White |
The top three publishers by journal and article count.
A closer examination of the reason for the availability of OA revealed limited use of APC payments or deposit of author manuscripts in PubMed Central, even though most journals offered these as options (see
APCs for pure gold and hybrid journals revealed significant differences in potential costs to authors. Of the 211 journals in this study, 22 were pure gold, with each of the 60 articles produced by these journals freely available full-text upon publication. Pure gold journals not only use APCs to cover publication costs but also use institutional memberships or other funding options to provide all content openly accessible upon publications. The pure gold journals in this study had variable APCs depending on memberships, government support, and the type of article published. APCs were listed on publisher sites for 19 of the 22 journals. One of these unlisted journals (
APCs were found for 137 of the 189 hybrid journals. The other 52 journals either required authors to contact the publisher for fee information or it was not evident if the option existed. Fees within a journal could vary depending on memberships, institutional affiliations, funding bodies, or which Creative Commons license was selected. The APCs ranged from US $600 to US $5000. The APC fees were notably higher for hybrid journals than for pure gold journals. The average cost was US $2800, and the mode was US $3000. The APCs for the hybrid journals with the highest number of articles in this study ranged from US $1000 for members in the
Snapshot of maximum article processing charges (APCs) for articles published in 2014.
Journal Type | N | APCa Information | |||
Journals | Articles | Range | Average | Mode | |
Journals without APC information | 18 | 22 | N/A | N/A | N/A |
Gold journals with APCs | 9 | 16 | US $1350 to US $2310 | US $1735 | US $1500 |
Hybrid journals with APCs | 54 | 108 | US $600 to US $5000 | US $2979 | US $3000 |
aAPC: article processing charge.
The number of final articles available as open access due to article processing charge (APC) payments, article manuscript deposit in PubMed Central, and final articles available without evidence of APC payment through deposit of published articles to PubMed Central or availability from the publisher website.
With the impracticability of the APC model for many researchers, the policies for self-deposit were reviewed for the 211 journals in the study. The SHERPA RoMEO website provided information on journal policies for 203 of 211 journals in the study. The ability for authors to self-archive a preprint or postprint of the article to an institutional or subject-based repository was offered by the majority of journals, but there was large variability in the version of the manuscript that could be deposited.
Of the 211 journals, 120 allowed deposit of the final draft (postprint) of the manuscript, 71 allowed the final draft but with additional restrictions, and 11 did not allow the final draft and information was not provided the other 9 journals. The SHERPA RoMEO analysis identified deposit embargoes on 138 of the journals that would have affected 564 of the 758 articles in this study (74%). In addition, the restrictive self-archiving policies (yellow and white SHERPA RoMEO categories) were applied to journals produced by 16 societies.
Of the 77 MS Society of Canada funded researchers included in this study, 62 had affiliations that would offer access to an institutional repository (IR). A secondary search of repositories for the 760 articles found that 8 of the 760 articles (1%) identified in this study were made available through an IR. Of these, all 8 were originally published in pure gold journals, namely in BioMed Central titles. IRs did not appear to function as a mechanism to make otherwise inaccessible content publicly available.
Despite encouragement from the public and the clinical community [
The hybrid gold model is fraught with cost implications to both the individual researcher and institutions. The monopoly of hybrid journal ownership by the big three commercial publishers (Elsevier, Wiley, and Springer Nature) has serious implications for the openness of research findings. In this study, these 3 publishers controlled 49.7% of the journals (105/211), 49.3% of the article output (374/758), and the majority (96/105, 91.4%) of their journals followed the hybrid model.
The hybrid model had been developed as a compromise between subscription publishers and OA advocates [
The mode APC cost for hybrid journals was US $3000, which was 45% higher than for pure gold journals, a finding which is consistent with other recent research [
The pure gold road to OA is not a road well-travelled by MS researchers. Pure gold journals accounted for 7.9% (60/758) of the articles and 10.4% (22/211) of the journals in this study. Although this study showed evidence of the use of pure gold journals by the researcher, the low number of articles in these venues may indicate that MS researchers feel that OA is poorly regulated, of poor quality, and lacking in peer review. This would support previous research that has found that despite awareness of OA and OA issues, faculty concerns about quality, reputation, copyright, plagiarism, and a perceived lack of peer review remained constant [
Self-archiving is the most cost-effective method of providing OA to research findings for researchers. Local self-archiving includes depositing a postprint or preprint of the article in an IR or self-archiving in a subject repository. However, rather than offering a simplified option, publishers have created barriers to OA through unsustainable complexities involved in self-archiving policies. Of the 35 professional society or association journals represented in this study, 15 had restrictive yellow self-archiving policies. Yellow self-archiving only allows authors to post the prerefereed preprint. With peer review considered the most important criteria in journal publishing, there is potentially little value to the open communication of research of an article without peer review [
Poltronieri et al [
The solution to resistance, either from journals or researchers, is sometimes assumed to be the requirements from funding sources or institutions [
Whether mandated OA will move MS research to a more open dissemination environment remains to be seen. The movement to OA requires incentives for involvement [
Surveys have shown that researchers have generally favorable views of OA and its benefits for both the public and the scientific community [
As this study relied on bibliographic analysis rather than contacting researchers directly to provide a list of articles, the 758 articles used in the analysis may not have been directly produced as a result of the MS Society of Canada funded research. Any participating authorship by the 77 researchers during the time of funding by the Society was included in the study as long as the article was related to the topic of MS.
The prevalence of OA literature produced by the MS Society of Canada researchers has remained consistently low between 2009 and 2014. Of the 758 articles, 288 are available OA. Most OA articles were made available without evidence of APC payment. Although APCs in hybrid journals was significantly higher than those associated with pure gold journals, the use of APCs for publication was low among this group.
The recent implementation of the MS Society of Canada’s OA policy may increase OA publication within this field. Future research should include prevalence of OA of newly funded research as a means of determining policy impact and effectiveness.
article processing charge
Canadian Knowledge Research Network
Institutional Repository
multiple sclerosis
National Institutes of Health
open access
This work was supported by the Research Support Fund of Wilfrid Laurier University and Alberta Health Innovates.
None declared.