Publishing a publication summary and getting it found
Whether a plain language summary of a publication reaches its readers depends on the sponsors and journals that commission and accept it, how it is reviewed and indexed, and the routes by which patients and clinicians find it.
Uptake by publishers and sponsors
Plain language summaries of publications (PLSPs) are increasingly considered part of the publication plan, and more journals include them in their requirements and their guidelines for authors. Journals such as PLOS Medicine had produced them since 2004 and Autism since 2011, yet before 2016 only about 100 were available on PubMed. The number grew to more than 400 in 2018 and doubled to about 800 by 2020. The number of summaries associated with peer-reviewed publications was still relatively low, and they varied widely in content, format, quality and where people could access them.
Journal publishers are increasingly supportive of accessible summaries as article enhancements. The rapid growth of accessible summaries may have been sparked by author and reader demand, but it was fuelled by regulatory mandates and further accelerated by publishers' analytics. The EU requirement for lay summaries of trial results is accelerating the need to write more documents in plain language, such as plain language protocol synopses and summaries of publications. Complementing the mandated lay summaries of trial results, some journal publishers make plain language summaries of journal articles available to the general public.
More journals encourage summaries of publications, submitted as a supplement to a manuscript or as a standalone publication, and some publishers, such as Future Medicine, provide a dedicated platform for them along with resources to help scientists and medical writers write high-quality summaries. In 2023, Future Science Group journals published standalone summaries of publications. Such articles have quickly gained popularity in the last few years. Several publishers, including Taylor & Francis, Sage, Becaris Publishing and Adis, offer the option to publish them in different styles, giving authors a choice of journals across a wide range of healthcare topics. By increasing the understanding of scientific content, such articles have the potential to increase the readership and downloads of the original article. Although many publishers have responded and are making great strides towards summaries that are fit for purpose, more can be done to help industry, authors and, ultimately, the non-specialist audience.
Some industry sponsors of clinical trials have made plain language summaries of journal articles available to the public for some time and others are starting similar initiatives, with commitments to make them accessible to the public for all clinical trials. From July 2022, Ipsen, one of the stakeholders who developed the Open Pharma recommendations, committed to publishing a plain language summary for every company-sponsored journal publication that includes data from human studies, with, as a minimum, a 250-word summary accompanying each. The mandate is a minimal commitment to accessibility and does not preclude other formats. Ipsen's implementation of the mandate included these steps:
- Define the mandate with clear minimum requirements.
- Communicate it internally and externally.
- Develop a plain language summary lexicon to help keep language consistent.
- Develop a review process for summaries.
- Prepare briefing materials and checklists.
- Identify non-expert or patient reviewers, patients being the gold standard.
- Build summary development into the publication standard operating procedures.
Some sponsors make plain language documents available as PDFs, infographics, comics and videos, and keep improving how they deliver them. Their impact on patients and the public still needs measuring, and awareness that the documents exist needs raising. A coordinated effort to raise that awareness and to make the documents available on shared platforms would advance open access, and doing it effectively may need metrics on whether the information is useful and understandable to the public, whether and how the data are shared and reused, and whether patients and the public can contribute to and track the scientific value of trial data.
Journal acceptance, review and indexing
Publishers and journals vary dramatically in how they approach PLSPs. Some journals include a PLSP in the submission process or ask for one on acceptance, while for others it is optional or not required at all. Some companies have their own guidelines and some publishers have short guides, such as Elsevier's "In a nutshell: how to write a lay summary" and Wiley's "How to write a lay summary for your research", but most give authors little or no guidance. Journals most often specify only the length and the target audience, and the guidance varies widely. Length ranges from 60–80 words to 250. Target audiences range from "understandable by media and educated patients", "someone in high school" and "an interested person without a scientific background" to a distant family member who works in retail, fashion or hospitality, and most journals target a higher reading age or ability than is expected in the general population. Very few give guidance on language, and where they do it varies dramatically; one recommends a readability analyser to indicate the reading age of the text. In general the content and structure follow the manuscript's abstract, and authors are advised simply to summarise the study's impact, importance, relevance and key findings.
Concise text-based summaries can help a nonexpert and time-constrained audience find and use published research. Concise text-only summaries can be indexed on PubMed. PubMed is one of the most widely used platforms for accessing biomedical research, and since 2019 a text-based summary hosted there can be discovered when it carries the plain-language-summary metatag. An automated search by Open Pharma of how the tag was used found a need for explicit guidance on both the indexing process and the tag's correct use, which could help improve uptake and correct tagging. Everyone, medical writers included, can widen a summary's reach by making sure it is tagged correctly on PubMed.
Several of Open Pharma's recommendations fall under the responsibility of journals: linking the summary explicitly to its source article, peer reviewing it fully alongside the manuscript, and tagging it with appropriate metadata and keywords to improve its discoverability. In late 2022 and early 2023, a 16-question Open Pharma survey asked journal editors and publishers about summaries hosted with their associated publication and whether their practices aligned with the recommendations.
Twenty-nine surveys came back, from 22 publishers or publishing imprints. Eighteen reported on a single journal, 4 on two journals each and 7 on a portfolio of journals; if each portfolio answer applied to the full portfolio, those 7 would cover between 6 and 418 journals (mean 100, median 30). Twenty-one of the 29 represented medical or health journals, and two represented journals publishing basic science and some medical science. Four of the seven portfolios let authors submit a summary to some or all of their journals, two did not, and one respondent was unsure. Of the 26 individual journals, only 7 accepted summaries and 19 did not, although two of the 19 wrote their own summaries in more accessible language for selected articles. One survey covered two journals with different policies and is counted in both groups. Journals could give more than one reason, and the commonest for not offering summaries were lack of reader demand (6 of 19), lack of author demand (5), lack of relevance to the journal's content (5) and lack of infrastructure such as costs, time and resource (3). Other reasons were that the journal produces such content itself, that its subject matter is not related to patients, that summaries were not yet a priority, and that its audience is very technical.
Among the seven journals that accepted summaries, practice varied. All took a text-based, abstract-style summary; three also took single-page infographics, and one took multipage infographics or video. Three placed the summary directly after the scientific abstract, one in the supplementary material and two in a text box within the article, and one placed it after the abstract, in the supplement or both, so four of the seven could carry it directly after the abstract. Respondents could name more than one intended audience: patients, their organisations and advocacy groups were the commonest (5 of 7), followed by healthcare and research professionals and students (4 each), then policy and governance professionals (3), educators and trainers (2), and, in free-text answers, funders, people with lived experience, media and social media, and anyone (1 each). Most of the journals that published summaries believed that summaries enable them to reach various non-specialist audiences. Six of the seven agreed that publishing summaries alongside the scientific abstract and article may increase or diversify their readership, and the seventh was unsure. Only three sent the summary for peer review with the manuscript; three did not and one was unsure. Three involved lay or non-expert reviewers, three did not and one was unsure, and two of those that did not explained that their non-expert reviewers took part in other journal processes but not in reviewing summaries. Two gave peer reviewers specific guidance on summaries, and one journal that said it offered authors no summary option said it gave such guidance. Only one used the PubMed metatag specific to plain language summaries; three did not and three were unsure, one of them saying it "probably" did not. The results also indicated that publishers and editors were not always aware of their own policies and practices, and needed better internal information sharing and training.
That so few journals use the tag may lead to incorrect indexing. A 2022 Open Pharma analysis found that 14.6% of PubMed records carrying the plain-language-summary tag used it wrongly, on a non-English abstract, on other content that is not a summary, or on a duplicate of the scientific abstract. Publishing summaries alongside scientific articles was increasing but had yet to become widespread practice across journals: another 2022 analysis found just 10 journals responsible for 73.5% of the text-based summaries indexed in PubMed. Earlier research had found great variability in the content, format and visibility of published summaries, and a need for consistent, standardised instructions to authors. Standardised tagging would improve indexing, but it is unclear whether it would make summaries more visible and easier to find for general readers, including patients and caregivers. For that, general readers must know where to look and be able to read summaries free of charge, perhaps through other medical information sources that link to PubMed.
The survey suggested that many journals had yet to adopt summaries and that those that had often departed from the recommendations, but that all of them offered at least the recommended minimum, a text-based, abstract-style summary. Taken with other studies of how summaries are published and with growing demand from readers and study sponsors, the survey points to a need for change in journal practices, for more journals to accept summaries and to follow best-practice recommendations so that summaries are peer reviewed and readers can easily find them. The survey's authors hold that the change should start with every journal letting authors submit a text-based summary of 250 words or fewer with any manuscript, and that peer reviewing the summary alongside the manuscript and tagging it with metadata for PubMed indexing would improve its accuracy and discoverability for specialist and non-specialist readers alike. Peer review alongside the manuscript is imperative if the summary is to be a trusted resource for both, because review ensures that it is scientifically accurate, true to the source article and free from bias.
Journals and publishers should require lay versions of publications. A requirement would strengthen scientists' efforts to communicate their science to lay audiences. If journals require PLSPs and insist on high-quality, fit-for-purpose ones, this will drive uptake and quality, and the requirement may also ease companies' compliance issues and the danger of their being accused of cherry-picking journals with no requirement. Whatever publishers decide, PLSPs should be free of charge and a routine part of publication planning, with medical writers involved in producing them from the start of that planning. One view, offered as perhaps simplistic, holds that many of the challenges of providing fit-for-purpose summaries could be solved by producing the abstract itself in plain language. Non-specialists would then have an easily accessible summary of the paper in language they understand, which time-poor healthcare practitioners would also appreciate. The detail would stay directly attached to the summary, with no risk of decoupling the scientific information from it and a much lower chance of misunderstanding and confusion.
One journal's editor-in-chief agreed with a requirement but worried that an ineffective lay version would only be ridiculed by scientists. A truly good lay version should still communicate the relevance of the research, which may require additional writing support and therefore cost. To support the usefulness of lay summaries, a cost-benefit assessment is needed, and with it further evidence about the public's attitude towards lay information.
Findability and trust
Although publishers increasingly publish PLSPs, their availability, accessibility, visibility and discoverability are still challenging. Most journals have tried to make them easily accessible and open access, but some place them behind a paywall or do not make them available at all. For members of the public, who may not be aware of publishers' websites, PLSPs are generally not publicly available or easy to find; journals and databases rarely have a dedicated category for them, and a 2020 study found only 2 of 11 PLSPs visible on PubMed. The lack of standard terminology may make searching harder too: lay summary, plain language summary, plain English summary, patient summary, author summary, general scientific summary, non-technical abstract, significance statement, highlights and blog are all in use, some journals use more than one, and the terms mean different things to different people. The lack of visibility and discoverability is a huge challenge and frustrates the general public, and patients feel that there is not enough open-access material online to be useful and that it is hard to find.
The introduction of plain language summary of publication articles has added to publishers' growing need to reach and engage patient audiences, and has brought them new challenges in discoverability, trust and awareness. Because the main audience of medical journals is healthcare professionals, the summaries are not always discoverable by lay readers; with so much health information online, it can be hard to tell legitimate information from misleading information, and hard for people without medical knowledge or awareness of medical journals to trust that what they read is accurate and complete; and, perhaps most fundamentally, patients must know such summaries exist before they can search for one.
A June 2022 Open Pharma roundtable of the publishing and pharmaceutical sectors identified three unmet needs for summaries: methods for evaluating how much easier a summary makes a publication to find and understand for non-experts, quality standards for summaries, and ways to reduce the extra work they cause publishers. Among the possible solutions it identified were promoting text-based summaries fully incorporated in the main manuscript alongside the abstract, so that they sit in front of any paywall and are indexed on PubMed, and discouraging hosting them in supplementary material or on third-party sites, which are less likely to be read unless the summary is distributed to readers some other way. Other possible solutions were to continue research and thought leadership on the value and readership of summaries, to convey their value to stakeholders at conferences and in the literature, and to educate using Open Pharma's best-practice recommendations article.
Plain language summaries should always be freely available, even when the article is paywalled, and some journals ensure this through social media, PubMed abstracts or dedicated hosting sites. They should be peer reviewed alongside the article, to ensure scientific integrity and clarity. They should be as visible as the article's other key components, and when hosted externally they must link directly to the full article, so as not to be classed as duplicate publications. Databases such as Datavision have begun flagging journals that publish them, which helps authors choose where to submit for a broader audience, and wider adoption of this by other databases would greatly benefit accessibility.
How patients and clinicians find summaries
In a 2023 survey of how people discover summaries of publications, 41% of 17 responding patients had found them through Google, while 41% of 32 healthcare professionals had found them through PubMed, against 12% of patients. This may be because patients without publication experience do not use PubMed or publishers' websites, the main routes to traditional journal articles. If publishers apply best practices for search engine optimisation to each summary, such as optimising its metadata with keywords like "plain language summary" in its title, they can improve its chances of being found on Google, though this relies on readers typing those words, and many patients may not know the term. Translations of summaries need hosting where patients who may not be confident using a journal website can discover them: many publishers have moved from supplementary files to signposting on the article page and hosting on platforms such as Figshare, which makes supplementary content easier to discover but is not optimised for readers with no experience of publishing.
Potentially the best way to increase the discoverability of summaries, and awareness of and trust in them, is to share them with patient advocacy groups. The groups represent and support patients and their caregivers living with a specific condition, and patients rely on them for credible information and support, such as plain language resources about their condition and treatment options. Publishers should be encouraged to share summaries with these groups, which can disseminate them directly to patients and on their social media channels, whereas pharmaceutical companies cannot do so without the sharing being considered promotional. In the 2023 survey, 18% of summaries were also found through individual recommendation and 12% through social media. Other research found that the commonest reason patients used social media for health was social support, such as sharing resources, and that social media is an important way for patient organisations to discover research. This suggests the power of sharing summaries with patients who have relevant networks and are more likely to recommend material to others. Reaching and engaging patients needs many stakeholders to work together, among them patients and caregivers, translators, advocacy groups and publication professionals. To optimise engagement with patients, publishers need to focus on improving discoverability and awareness and on increasing trust in publishers, researchers and their publications.
Few articles are published with easy-to-find plain language summaries, although healthcare professionals and other audiences value them and pharmaceutical companies increasingly write them. Between 24 April and 17 June 2024, an online survey was sent to 5141 people who had previously contributed to articles sponsored by AstraZeneca, Ipsen or GSK. Of the 185 healthcare professionals who responded, 62% (115/185) had more than 20 years of clinical practice and 60% (111/185) did not have English as their first language. Of the 185, 133 (72%, 133/185) had read or contributed to a summary, yet seven of them (5%, 7/133) had never read or used a summary provided with an article. Of the 126 (95%, 126/133) who had read or used summaries, most (71%, 89/126) found them by chance alongside articles. Publishing more summaries and improving how they are found will help broaden the impact of scientific research.