Jane Falconer is a medical librarian with over 20 years experience in medical charities, the NHS and Higher Education. She is currently the User Support Services Librarian at the London School of Hygiene & Tropical Medicine, responsible for all user-facing library services, including user training and support, membership, access and enquiry support, reading lists, interlibrary loans and liaison services. She also provides literature searching support for systematic reviews, she has created and run the searches on a number of projects including the Lancet Commission on Planetary Health and the WHO Guidelines on Heptatitis B and C Testing. ORCID: http://orcid.org/0000-0002-7329-0577. Jane got in touch via Twitter as she was frustrated by copyright laws that prevented her sharing medical articles with researchers around the world. Here’s what she told us…..
My name is Jane Falconer and I need advice from the copyright community.
In order to stop people dying I have been asked to upload a large number of academic articles to the cloud, so that six people situated in different institutions around the globe, can synthesise the information and recommend the best course of action. Do I do this, and break all sorts of copyright and licensing restrictions in the process?
This may sound melodramatic, but this situation happens all the time when medical researchers collaborate on a particular type of study called a systematic review.
A systematic review is a specific type of literature review, with a set methodology designed to provide an unbiased answer to a question, based on a summary of all the currently available studies. They were first developed in medicine in the 1970s, where doctors were being asked to make life-changing decisions based on scattered and contradictory evidence. The systematic review was intended to bring this evidence together to produce a more clinically and statistically significant conclusion. They are now used by health professionals around the world to ensure they implement best current practice. They are also used to produce documents such as medical guidelines which provide national or global medical practice advice.
Given the importance of reliable information to frontline health professionals, it is important that systematic reviews are conducted according to recognised and tested methodologies. Two of the best known are PRISMA and Cochrane. PRISMA is commonly used by authors to guide their methodology and reporting and they provide the evidence to back-up their methodology. Where it gets interesting from a copyright viewpoint, is when researchers are deciding which items to include in the systematic review, and which to exclude. They recommend that:
“Efforts to enhance objectivity and avoid mistakes in study selection are important. Thus authors should report whether each stage was carried out by one or several people, who these people were, and whether multiple independent investigators performed the selection, what the process was for resolving disagreements. The use of at least two investigators may reduce the possibility of rejecting relevant reports. The benefit may be greatest for topics where selection or rejection of an article requires difficult judgments.”[my emphasis][i]
The Cochrane Library (named after Archie Cochrane, the main advocate for systematic reviews) is generally regarded as producing the gold standard of systematic reviews. Their methodology has been adapted and is now used to produce systematic reviews in other subject areas including social sciences and ecology. When we look at their screening methodology we see that they are far more explicit than PRISMA:
“The methods section of both the protocol and the review should detail:
- Whether more than one author examines each title and abstract to exclude obviously irrelevant reports:
- Whether those who examine each full-text report to determine eligibility will do so independently (this should be done by at least two people) [my emphasis]”[ii]
Therefore, we need at least two people to have access to the same items. This is relatively easy when the researchers all belong to the same institution which has a well-funded library with subscriptions to all the relevant articles. In my experience, this has never happened.
The reality more closely matches the situation described at the beginning of this blog. Let me put some more detail on that, loosely based on a real study. An international group of 6 researchers have been commissioned by the World Health Organization to conduct a systematic review on a topic, let’s say the efficacy of low-cost diagnostic tests for sexually transmitted diseases. The production of these tests is important in resource poor countries where current tests requiring samples be sent to a lab for analysis are too expensive to roll out to the entire population. If the review can show that these new tests are sufficiently sensitive and specific, the review will form a key part in the production of new guidelines which will update global recommendations. This will lead to more people being successfully diagnosed and treated, with a subsequent reduction in infection rates. The researchers are based in institutions in China, Tanzania, Brazil, USA and UK, in order to ensure regional differences are reflected in the study. To facilitate communication, the lead author creates a Dropbox folder, to share project-specific documents. Given the complexity of the topic, I am commissioned to conduct the literature search and help source the required information. I conduct a search of the literature, searching 12 bibliographic databases and several sources of grey literature, being deliberately broad and comprehensive to limit the possibility that a study has been missed. Once the results are de-duplicated, 6348 items are identified for screening. I export the titles and abstracts to Excel and upload to the Dropbox folder where the team exclude any obviously irrelevant items. We are left with 301 items where the abstract does not provide sufficient information to include or exclude. Therefore, the team need to read the full-text of all 301 papers. The researcher in the USA is based at an Ivy league institution, the researcher in Tanzania does not have access to a library at all, the researchers in China and Brazil only have access to a limited number of western journals and some local titles. Therefore, the lead researcher asks me to source the full-text of these items and upload them to Dropbox.
How can I do this without breaking copyright and licensing laws?
- 25 items are available open access. These are the easy ones to deal with and everyone has access to these.
- 150 items are available through my institution’s e-journal subscriptions. Can I upload these to the cloud for others in the research group to access, even if they are not affiliated with my institution?
- 50 items are available through the subscriptions of the American institution. Are US licences and copyright restrictions different to those in the UK?
- 2 items are from the 1970s and are available in print in my library. Can I scan these and upload them to the cloud for others to view?
- 99 items have to be ordered through our inter-library loans system. This is where it gets really difficult. If I order these from the British Library, they are linked to my BL account. I either have to share my account details with the group, or print them, re-scan without DRM and upload the papers. The National Library of Medicine in the USA sends PDFs of items without DRM but is clear that the documents should not be shared, copied, or stored in electronic format.
It seems to me that copyright and licensing restrictions are at loggerheads with study methodology. I cannot see any way the study can continue without breaking the rules. However, not completing the research to the highest methodological standard is not an option. Compromising the quality of the study risks drawing the wrong conclusion, causing potential harm to a large number of people, as well as damaging the reputation of the researchers and the institution.
I was commissioned to write this blog as the result of a twitter conversation I had with UK Copyright Literacy over a blog post on the ACRLog. It details a discussion around why academics consistently break copyright and licensing rules. It’s a good post, which brings up many important points. However, nowhere does it discuss the realities of trying to do research in an interconnected, digital world where funding grants are awarded for groups who collaborate between institutions, and money is specifically available for collaborative projects in resource poor countries (particularly where they have a capacity building component). I believe that copyright and licensing rules have not kept pace with technology and the reason many academics break the rules is because it is impossible to complete their research without doing so.
As a librarian I am caught in the middle. Can we tell academics that in order to conduct a systematic review, the balance of risk is such that they can break the rules to complete their study? If so, how can we then tell the same academic that they can’t email a PDF to someone, or upload a chapter to their course site? Like many others, we are working with academics to encourage them to publish open access. However, this is a solution for newly published information, it doesn’t help people like myself and my research group who are trying our best to source already published information in order to improve health across the world.
I’d be really interested to read everyone’s advice and comments.
[i] Liberati A, Altman DG, Tetzlaff J, Mulrow C, Gøtzsche PC, et al. (2009) The PRISMA Statement for Reporting Systematic Reviews and Meta-Analyses of Studies That Evaluate Health Care Interventions: Explanation and Elaboration. PLOS Medicine 6(7): e1000100. https://doi.org/10.1371/journal.pmed.1000100. Pg 9.
What do you think? Are you a publisher or rightsholder and do you have any thoughts about the issues that Jane has raised? We’d love to hear from you and publish a response to this blog post, but also to ask other librarians how they deal with these sorts of tricky issues related to systematic reviews.