LSE - Small Logo
LSE - Small Logo

Vincent Larivière

Fei Shu

Cassidy Sugimoto

March 5th, 2020

The Coronavirus (COVID-19) outbreak highlights serious deficiencies in scholarly communication

17 comments | 193 shares

Estimated reading time: 7 minutes

Vincent Larivière

Fei Shu

Cassidy Sugimoto

March 5th, 2020

The Coronavirus (COVID-19) outbreak highlights serious deficiencies in scholarly communication

17 comments | 193 shares

Estimated reading time: 7 minutes

As research and government responses to the COVID-19 outbreak escalate in the face of a global public health crisis, Vincent Larivière, Fei Shu and Cassidy R. Sugimoto reflect on efforts to make research on this subject more widely available. Arguing that a narrow focus on research published in high ranking journals predominantly in English has impeded research efforts, they suggest that the renewed emphasis on carrying out open research on the virus presents an opportunity to reassess how research and scholarly communication systems serve the public good.


Major crises often reveal the hidden norms of the scientific system, making public well-known practices within science. The coronavirus (COVID-19) outbreak exposes an inconvenient truth about science: the current scholarly communication system does not serve the needs of science and society. More specifically, the crisis makes manifest two inefficiencies in the research system: the default to closed science and the overemphasis on elite, English-only publishing, irrespective of the context and consequences of the research.

On 31 January 2020, the Wellcome Trust called the coronavirus a “significant and urgent threat to global health” and called on “researchers, journals and funders to ensure that the research findings and data relevant to this outbreak are shared rapidly and openly to inform the public health response and help save lives.” Signatories to this statement included dominant publishers—such as Elsevier, Springer Nature, and Taylor & Francis—as well as several funders and scholarly societies. The joint signatories on this statement committed to making all research and data on the outbreak open immediately: on preprint repositories for those items that have not been peer-reviewed and on journal platforms for those articles which have already been reviewed.

The papers and book chapters that have been liberated by this measure represent only a tiny proportion of the available literature on coronavirus.

This is a positive step, but it does not go far enough to address public needs. The papers and book chapters that have been liberated by this measure represent only a tiny proportion of the available literature on coronavirus. According to the Web of Science (WOS), 13,818 articles have been published on the topic of coronaviruses since the late 1960s. More than half (51.5%) of these articles remain closed to access. The coronavirus is admittedly a large family of viruses and one might question relevance of older works to the current outbreak. However, as an example, the three papers on COVID-19 published in the February 15th issue of the Lancet relied on 69 distinct WOS-indexed papers, of which 73.2% are in the set of 13,818 coronaviruses papers. The oldest reference in these papers is to 1988, underscoring the fact that although the corona virus may be novel, research on the corona virus in fact draws on a long tail of often closed research literature.

The embeddedness of this scientific literature within much wider streams of research also highlights the limitations of this approach. The 13,818 coronavirus articles cite more than 200,000 articles—from virology to cancer and from public health, to genetics and heredity (Figure 1). Less than one third of the cited articles from which the “coronavirus articles” drew information and inspiration were other “coronavirus articles”. Even if all articles on the topic of coronaviruses were made available, this would still be insufficient to address the crisis, given the inherently interdisciplinary nature of biomedical research. The knowledge base of science is simply much broader than a single topic. Viewing the literature through the narrow lens of coronavirus articles directly relevant to COVID-19 alone blinds the research effort to other work that could prove crucial. Cures for diseases often come from novel combinations and insights from several areas of research. If the goal of opening research is to advance science and serve society, all research should be open, not just a portion of it.

Figure 1. Percentage of references cited by the coronavirus papers, by speciality of the cited journals. NSF field and subfield classification 1988-2018.

 

Publishing incentives are the other controversial element revealed by the current outbreak. Over the last decade, Chinese authorities and institutions —like those of many other countries—have offered direct financial rewards based on the journal in which researchers publish, with an implicit objective of improving their institutions’ position in international rankings. Invariably, publishing in these journals implies conforming to the lingua franca—English—and publishing on topics of significance to the gatekeepers of these journals, who are disproportionately from Western countries. While dissemination to the wider scientific community is an important objective, it should not come at the expense of dissemination to local communities, particularly those with direct connection to the topic of study. Due both to paywalls, and to the use of English, international journals are often inaccessible to those on the global front line of providing medical care and crafting health policies, especially in times of crisis.

The current coronavirus outbreak exemplifies this deficiency. At the end of 2019, the Chinese Center for Disease Control and Prevention (CCDCP) sent a group of experts to Wuhan to retrieve data on the virus. This was nearly three weeks after the first patient had presented with symptoms and immediately following the news of human-human transmission on social media by eight Wuhan doctors (who were subsequently charged by the police). The researchers analysed the data and submitted the results—including a verification of the human-to-human transmission of the virus—to the high ranking western journals, The Lancet and NEJM, published on January 24 and 29, respectively. On 20 January a public statement was released, acknowledging the human-to-human transmission of the virus.

Figure 2. Number of papers on disease surveillance published by Chinese researchers in international journals (WOS) and in national journals (CNKI). Following the 2003 SARS epidemics, the importance of research on disease surveillance in China has increased exponentially.

 

In response, the Chinese government stipulated that funded projects on coronaviruses—including those from the new 1.5M initiative of the National Science Foundation of China (NSFC)—should be published in local Chinese journals rather than international journals, and that the emphasis should be to control the virus and save lives. This suggests a recognition on the part of the Chinese government that the focus on publishing in elite journals did not provide the most expedient mode for disseminating results. Furthermore, Ministry of Education (MoE) and Ministry of Science and Technology (MoST) issued a joint statement that requires universities and research institutions to limit the use of SCI papers as well as related indicators (e.g., JIF, ESI, etc.) in research evaluation. MoST has also stipulated that the number of papers cannot be used as the key criterion for assessing research performance and has banned the use of cash-per-publication policies. All of these initiatives point to an underlying truth: prioritising indicators over the timely delivery of research findings to relevant communities is not in the best interest of society.

Is there not a public health benefit to making public research on this and all research that could accelerate discovery in biomedicine and save lives?

The signatories on the Wellcome Trust statement agree to follow these principles not only for the present outbreak, but for all situations in the future “where there is a significant public health benefit to ensuring data is shared widely and rapidly.” This statement makes a direct link between public health and the sharing of research results: implicitly making the case that journal paywalls and embargoes hinder the advance of science and as a result are a threat to public health. However, it also raises the question, where does one draw the line as to what constitutes a ‘public health benefit’? In the last five months, the Center for Disease Control in the United States estimated that there were between 18,000 and 46,000 flu-related deaths. Is there not a public health benefit to making public research on this and all research that could accelerate discovery in biomedicine and save lives?

We call on the scientific community—publishers, funders, and societies—to stay true to their word. The Wellcome Trust statement is unequivocal: the rapid sharing of research is necessary to inform the public and save lives. While we applaud the work that is being done amidst this crisis, we hope that this moment serves as a catalyst for change. The Trump administration in the United States, for example, is considering an executive order that would make all federally funded studies free to read on publication. Similarly, the Plan S coalition of funders require all funded research to be published in open access journals. While many funding agencies have adopted open access policies, compliance is variable and embargoes currently limit immediate access to biomedical research. Both the potential executive order and Plan S have been opposed by many of the signatories on the Wellcome Trust statement. This is a blatant contradiction.

Signatories of the Wellcome Trust statement should extend the principles to encompass all their practices: making research available immediately and incentivizing science communication to all stakeholders. The scientific response to COVID-19 has demonstrated some of the benefits of opening the scientific system: including the torrent of papers being immediately shared on preprint servers, the open collaboration and discussion of scientists using social media platforms, and the accelerated modelling of the viral genomes. However, this will have been in vain if the scientific system does not change. It is essential to recognise what is made clear in this moment of crisis: a robust scientific system and an informed citizenry requires immediate and public access to research.

 

Note: This article gives the views of the author, and not the position of the LSE Impact Blog, nor of the London School of Economics. Please review our comments policy if you have any concerns on posting a comment below.

Featured image credit: Orda Noel, Deep Breath Respiratory Tissue, via the Wellcome Trust, (CC BY-NC 4.0)

Print Friendly, PDF & Email

About the author

Vincent Larivière

Vincent Larivière is professor of information science at the Université de Montréal, where he holds the Canada Research Chair on the Transformations of Scholarly Communication. He is also scientific director of the Érudit journal platform and associate scientific director of the Observatoire des sciences et des technologies (OST-CIRST).

Fei Shu

Dr Shu is a senior researcher in the Chinese Academy of Science and Education Evaluation at Hangzhou Dianzi University; he also works in École de bibliothéconomie et des sciences de l’information, Université de Montréal as an associate researcher. His research focuses on bibliometrics and scholarly communication, especially on the impact of science policies on research activities.

Cassidy Sugimoto

Cassidy Sugimoto is Professor of Informatics at the School of Informatics, Computing, and Engineering at Indiana University Bloomington. She is also currently serving a rotation with the National Science Foundation as the Program Director for the Science of Science and Innovation Policy (SciSIP) program. Sugimoto's research expertise is broadly situated in the domains of science policy, scholarly communication, and scientometrics.

Posted In: COVID 19 | Research policy

17 Comments

This work by LSE Impact of Social Sciences blog is licensed under a Creative Commons Attribution 3.0 Unported.