Grey literature and the gaps in the scholarly record
Grey literature is the reports, working papers and other material that organisations issue outside the publishers, supply chains and discovery services that build the scholarly record. Much of it is missing from the databases and catalogues a writer searches, its supply has grown over the past decade, and some of it drops out of reach. Some of it can be searched in repositories its issuers keep, such as the WHO's.
What the scholarly record is and leaves out
The scholarly record is an ever-evolving network, or graph, of truth assertions on which each discipline bases its discussions and against which each discipline measures its progress. Some go further and suggest that the scholarly record can even frame the identity of an institution. In one grey-literature specialist's image, the scholarly record is the giants' shoulders on which, as Newton wrote to his rival Hooke in 1675 and as Bernard of Chartres is quoted as saying five hundred years earlier, later scholars stand to see further than their predecessors. If a giant's shoulder is worth standing on, the question is how to ensure that the giants' truth assertions are collected and preserved, so that later scholars can see further and measure progress, and that all giants are included in the record.
Dougherty proposed that to qualify, an item must advance or summarise knowledge, have an identifiable author, be issued through an academic publisher, be catalogued by a university library, appear in curated research databases and belong to a recognised discipline. OCLC, a global library organisation, defines the record as the "published outcomes of scholarly enquiry" such as "journal articles and monographs", although others recognise that it has become much more diverse and encompasses protocols, code and data.
The first published outputs on COVID-19 fail both definitions. On 31 December 2019 the Wuhan Municipal Health Commission published a briefing on "a pneumonia epidemic situation" and informed the WHO China Country Office, and on 5 January 2020 the WHO published a briefing on its website. These publications marked the start of the pandemic, and future scholars and students might well want to study them, yet they fail to qualify because they were not issued by a scholarly publisher and did not appear as a journal article or monograph. By 2022 the link to the Wuhan briefing returned "404 – page not found", and it was unclear whether the content was simply offline or lost. Either way it was no longer easily accessible, and if it is lost, its loss shows why maintaining the scholarly record matters.
The beginning of the record of HIV/AIDS fails in the same way. On 5 June 1981 the US CDC published an article in its Morbidity and Mortality Weekly Report (MMWR) describing rare lung infections in five young men in Los Angeles, but the CDC is not an academic publisher and MMWR is not a journal, so in theory the report does not qualify for the scholarly record.
Two working papers by economists from the Bank of England, University College London, Cambridge University, the London School of Economics and the University of Warwick fail in the same way. In 2019 two professors from University College London and King's College London discussed them in a podcast. The papers made headlines in the UK press, including the Financial Times, and were cited in a blog, run by a professor at Royal Holloway, University of London, that has a larger following than most journals. None of these items passed through the hands of an academic publisher, and hunting for them in journals, subject databases and library catalogues is in vain, because the content was released without any thought as to how it might be captured for the scholarly record. The same specialist concludes from these examples that Dougherty's and OCLC's criteria need updating, not least to take account of how digital and Web 2.0 tools are changing the ways in which knowledge is published.
Why grey literature escapes discovery
In the analogue era authors had little choice but to find a publisher, because the cost of self-publishing and of dissemination in print was beyond most of them, and the cost of organising and maintaining archives meant that only institutional libraries could offer meaningful collections of published material, so publishers and libraries were central to the scholarly record. Publishers worked with booksellers and agents on an efficient, near-global supply chain that carried their publications to libraries, and to reduce administration costs and speed delivery, publishers, booksellers, agents and librarians co-developed processes such as ICEDIS and metadata standards with unique identifiers, ISBNs in 1969 and ISSNs in 1975. In parallel, secondary services and catalogue systems emerged to tackle the challenge of discoverability. Since 2000, with the transition to the digital era, both the supply chain and the discovery services have been re-engineered, and today's standards include persistent identifiers for content (DOIs), for authors (ORCIDs) and for their institutions (Ringgold).
Not every scholarly publication exists inside this complex of publishers, secondary services and libraries. Some institutions choose to self-publish for advantages such as control over branding, timing and pricing. Some of them, such as the OECD and the Brookings Institution, mimic mainstream publishers and use the same metadata standards and supply chains to channel their publications to libraries, but others, especially smaller organisations, do not, so their content is hard to source and missing from secondary discovery services, which frustrates librarians and readers alike. That content is known as grey literature. The term covers a wide range of information produced outside traditional publishing and distribution channels, and a lack of indexing or tagging with uniform metadata typically makes it hard to discover.
A public search engine such as Google does not close the gap. Public search engines deprecate content with poor metadata on low-traffic websites, so most grey literature is crowded out by content from websites optimised by digital marketers. They also seek to tailor results to each user's bubble of preferences, attitudes and even location, and their results can change from day to day as their algorithms evolve. So most scholars and students still turn to specialist search engines, where grey literature is largely absent.
What grey literature is
Wood coined the term grey literature in 1984 for material "which is not available through normal bookselling channels", which leads to problems for the producers of secondary services, for librarians who wish to collect it, and for end users. He noted its other distinguishing characteristics, "variable standards of editing and production, poor publicity, poor bibliographic control, and poor availability in libraries", but rejected as "mistaken" the belief that it was "essentially ephemeral and of local interest only", because "it contains information likely to be of use to a considerable number of people". His definition captures the challenge that grey literature poses information professionals and readers: the content is hard to find, capture and use.
The Prague definition of 2010 describes grey literature as manifold document types produced on all levels of government, academics, business and industry in print and electronic formats, protected by intellectual property rights, of sufficient quality to be collected and preserved by library holdings or institutional repositories, but not controlled by commercial publishers, that is, where publishing is not the primary activity of the producing body. It attempted to build on Wood, but in the same specialist's view its additions muddy the waters. On that view, the Prague definition lists some producers but excludes others, such as third-sector organisations and NGOs; its qualifier "commercial" fails to understand that any publisher, for profit or not, must behave in a commercial manner; and it is wrong to suggest that bodies for which publishing is not the primary activity necessarily produce grey literature, because many universities and some intergovernmental and non-governmental organisations run professional presses that publish in a manner identical to houses like Elsevier and Springer, and their publications are as easily found in secondary services and obtained through standard supply chains. On the same view, Wood's definition is just as valid in the digital era.
Material can also move from the formal record into grey literature. The Intergovernmental Panel on Climate Change switched from working with publishers to self-publishing on its websites, so what was formally published is now grey literature.
Grey literature is often thought not to have been peer reviewed, but this is a big misunderstanding: at least in policy, more than 60% of it is reviewed by experts before release. Wood reckoned it "a costly public asset going largely to waste", and one estimate puts the cost at $33 billion a year.
Growth in supply
The core problem is the one Wood identified in 1984, poor bibliographic control, and it is compounded by a significant increase in the supply of grey literature over the past decade. The supply of scholarly content strongly correlates with the number of researchers. In the 1980s just under a third of those emerging from education systems in OECD countries had first degrees, and roughly a quarter of these went on to a master's or doctorate, so around 8% emerged research capable. In the 2000s up to half left with degrees and half of those went on to postgraduate qualifications, so 25% emerged research capable. Yet the number of jobs in academia barely changed: in the 1980s around 15% of new PhDs in the UK could expect to work in academia, and by the 2000s around 3%.
Some of these researchers went into industry and government, and some must have joined the booming services and third sectors. The third sector is the part of an economy or society comprising non-governmental and non-profit-making organisations or associations, including charities, voluntary and community groups and cooperatives. The number of new third-sector organisations has grown strongly since the end of the Second World War, many of them employing researchers to support their mission. Unlike academic researchers, researchers in government, industry and the third and service sectors do not have to publish in books and journals to further their careers, so they are free to self-publish their research with their employers as reports, working papers and other digital-first formats, and they increasingly do so.
The Policy Commons database, which indexes grey literature from over 8,500 intergovernmental and non-governmental organisations, think tanks and research centres around the world, shows 55% more grey literature released in 2020 than in 2010 (287,545 items against 184,514). By one estimate, the field of policy alone sees around 400,000 newly published items of grey literature each year, 10% of the world's journal output.
Poor bibliographic control and lost content
Desktop publishing, Web 2.0 tools and websites make it easy for anyone to self-publish. In a 2012 interview Clay Shirky said that publishing was going away: the word had meant a cadre of professionals taking on the difficulty, complexity and expense of making something public, and that was no longer a job but a button that says "publish", which, when pressed, is done. In the same specialist's view he was half right: pressing the button is easy, but most people who press it do not know how to wrap the content in the metadata that makes it discoverable and easily and reliably citable, or how to get it included in specialist discovery services, and do not understand that it is not done until the work has been safely preserved in the scholarly record. The interview was published in the blog Findings, and within months links to it returned "404 – page not found" because the blog had closed and gone offline. Like the publisher of Findings, most organisations have no strategy to prevent link rot, and 75% of links in scholarly journals to items on the web at large lead to the wrong content.
Over the past two decades, publishers and librarians have focused on capturing research findings from the academy, mainly in books and journals, to build a digital scholarly record overlaid with sophisticated discovery systems for the academy's use. At the same time they are attempting to pivot this $25 billion industry to open access, so that the scholarly record becomes an asset for society at large as well as for the academy. In parallel, and largely ignored, a growing number of researchers at non-academic institutions and organisations have used digital publishing tools to post their findings openly on their websites as reports and papers. This grey literature is also a $25 billion information industry, but it is missing both from specialist discovery systems and from library collections and is still woefully under-used, a costly asset going to waste.
What it means for a writer's literature review
In 1990 a professor running a laboratory in France kept its library open around the clock but gave the laboratory key only to those who had first used the library to complete a thorough literature review of the topic they wished to investigate. At a time when practical and financial hurdles kept grey literature scarce, the policy made sense, because the professor could be confident that the library's access to the scholarly record was such that laboratory time would be spent only on looking further than was possible from what had gone before. Now that a button has removed the practical and financial barriers to posting findings on employers' websites, the policy would be increasingly undermined: laboratory time might be wasted because a growing volume of truth assertions is missing from the library's collection. Preparing content and metadata to the standards that make it discoverable by users and easily available to librarians is still incredibly difficult and complex, and despite the advances in digital publishing, gathering the scholarly record is still as challenging as herding cats.
Being free to access is of no value if content is not discoverable. The same specialist calls on medical writers and communicators to engage with grey literature as a rich and diverse source of information.
The WHO's Institutional Repository for Information Sharing
The WHO's Institutional Repository for Information Sharing (IRIS) is a free-to-access digital library of the WHO's published scientific material and technical information, and its full content is freely accessible and searchable in the six official languages of the WHO: Arabic, Chinese, English, French, Russian and Spanish. WHO country offices, regional offices and headquarters add publications and articles, and the uploaded material is stored in a central location with unique document links. The WHO also publishes through the IRIS of the Pan American Health Organization (PAHO), the international health agency for the Americas, which gives free access to PAHO publications in digital format.
The WHO publications available include these:
- Reporting on the world's biggest killers: a journalist's guide to covering noncommunicable diseases (8 October 2025), developed with and for journalists from real-life experience of reporting on noncommunicable diseases, with tip sheets and advice for covering them.
- Advancing the responsible use of digital technologies in global health (24 September 2025), a report in which the WHO Science Council examined the state of digital health across countries, looked at obstacles to and goals for the digital transformation of health systems, and made specific recommendations.
- Guidance for best practices for clinical trials (10 September 2025), for those whose work relates to clinical trials in any way, including the planning, conduct, analysis, oversight, interpretation and funding of all clinical trials assessing the effects of any health intervention for any purpose in any setting.
- Combating false information on vaccines: A guide for health workers (30 June 2025), in PAHO's IRIS, a resource for health workers to combat vaccine misinformation, support informed decision-making and promote trust in vaccines within communities, with related guides for Expanded Program on Immunization managers, for educators, and for risk communication and community engagement teams.
- Evidence-informed policy-making: a glossary of key terms (29 March 2025), 133 terms commonly used in evidence-informed policy-making, which addresses confusion and offers a shared vocabulary by clarifying existing definitions, and for which, according to the WHO, a glossary had been identified as critical in supporting the WHO "to promote a common understanding and interpretation of the key elements of evidence-informed policy making".
- ICD-11: International Classification of Diseases for Mortality and Morbidity Statistics, Eleventh Revision (February 2025), a standardised language for health professionals worldwide to record and report diseases, injuries and causes of death, used for recording and reporting mortality and morbidity statistics nationally and internationally.
- Glossary of health data, statistics, and public health indicators (22 January 2025), an alphabetical stand-alone glossary with a background and introduction that gives standardised definitions for public health data, statistics and health indicators, intended according to the WHO for data managers, health analysts, monitoring, evaluation and surveillance specialists and others who deal with them across the data cycle, from collection to use.
- WHO Drug Information, a quarterly journal giving an overview of topics in medicines development and regulation for a wide audience of health professionals and policy makers, its latest 2025 edition being volume 39, number 3.
- International Standards for Clinical Trial Registries, version 3.0 (27 September 2018), which presents the minimum standards that must be attained to meet the requirements of a primary registry in the WHO Registry Network, on the principle that the registration of all interventional trials is a scientific, ethical and moral responsibility.
- Basic epidemiology, second edition (2006), which introduces the role of epidemiology and presents information aimed at strengthening education, training and research in public health, and which as of 7 October 2025 was one of the top downloaded publications according to the IRIS site statistics, within the 50 items most downloaded and viewed.