Improving and spreading plain language summaries of peer-reviewed medical journal publications

View correction statement:

Correction

This article is related to:

Open Pharma recommendations for plain language summaries of peer-reviewed medical journal publications

Science has been the main means of progress in overcoming the COVID-19 pandemic, and the results of scientific studies have been on the front page of newspapers day after day. People from both general and professional audiences want to know what is happening with the pandemic, whether new treatments and vaccines work, how the virus spreads, and if wearing masks is beneficial. Increasingly, people access the peer-reviewed medical journal publications themselves rather than simply trusting media reports. This is a trend that antedates the pandemic and has been compared with the Reformation, when people wanted to communicate directly with God and not through the church¹. The publishing community has responded by producing plain language summaries, which have come in many forms, at different lengths, with variable amounts of detail, and sometimes including infographics and videos. The summaries have also been processed in different ways—sometimes they are peer reviewed, sometimes they are not. It made sense for somebody to standardize the requirements for plain language summaries. Open Pharma, a collaboration to drive positive change in the communication of pharma-sponsored research, has attempted to do just that². The recommendations and the process used to arrive at them are described in Rosenberg A. et al. https://doi.org/10.1080/03007995.2021.1971185.

Plain language summaries must be inclusive. They are intended for everybody—not just patients but also health professionals (many of whom do not have the time to read full papers or do not have access to them), policy makers, journalists, and many others. For some people, they may be enough, whereas for others, they might simply be a step to the full manuscript. It is clearly essential that, in the words of the recommendations, the plain language summaries are “understandable and readable, free of technical jargon, unbiased, [and] non-promotional.” The recommendations also state that they should be “accessible,” and everybody, including publishers, will surely benefit from them being free to access, even if the journal publications they are summarizing are behind paywalls.

I became an editor at The BMJ in 1979, and I was the editor in chief of the journal and the chief executive of the BMJ Publishing Group from 1991 to 2004. I subsequently spent 9 years on the board of the Public Library of Science, and I continue to blog for The BMJ and have co-authored a major report for the Lancet that is still to be published. Much of what I have seen in more than 40 years of being associated with journals has depressed me^3,4, but one development that has unquestionably been beneficial has been the standardization of how studies are reported. This standardization has been led by the Equator Network and started in 2010 with CONSORT for reporting randomized trials^5,6. As a result, the reporting of trials has improved⁷. CONSORT has been steadily updated, and the Equator Network now identifies 175 reporting guidelines⁵. The recommendations on plain language summaries join an important tradition and will no doubt be developed over time.

Importantly, Open Pharma included all stakeholders in producing the recommendations: patients, researchers, editors, publishers, and funders of research, particularly pharmaceutical companies. The recommendations start with the recognition that, to be indexed on PubMed, plain language summaries need to be text based and formatted as a single paragraph. PubMed will be the way that most people access summaries, and a 250-word summary (as recommended by Open Pharma) that is easily discoverable will be more valuable than a longer, more complex one that is hard to find. The group of stakeholders have also decided that the recommendations should set minimum requirements.

The recommendations should not be hard to follow, but they should increase the usefulness and consistency of plain language summaries. They should be based on individual journal publications (not collections of publications) with a link to the manuscript for those who want to access the full study. Unfortunately, it will not always be possible to access the full publications for free, but free access is surely something that funders should aim for—as is recommended by Plan S, a coalition of research funders that aims to make all research open access on the day of publication⁸. The summaries should have the same conclusion as the journal publications they are summarizing. This sounds obvious, but we know that abstracts of research articles commonly exaggerate the conclusions of studies⁹. The plain language summaries should be peer reviewed along with the manuscripts themselves. Ideally, reviewers should include a non-expert to ensure that the summary is clear and understandable.

The language of the plain language summaries matters hugely, and, although it might seem straightforward to use language that is “understandable and readable, free of technical jargon, unbiased, [and] non-promotional,” many researchers find it difficult. They are steeped in their subject, unable to recognize that jargon is jargon, and used to academic writing that was memorably described by Michael O’Donnell, editor of World Medicine, as “decorated, municipal gothic”⁹. I had the experience of reviewing case studies for a university submitted as part of the Research Evaluation Framework. These were supposed to describe the impact of the research in plain language and would be accessible to non-experts, but most of these case studies had to be rewritten. Publishers and editors may need to help authors, particularly in the early days, before authors become more familiar with writing the summaries.

A big challenge for all guidelines is implementation. Studies showed that long after the introduction of CONSORT, for example, many randomized trials still lacked information that should have been included⁷. If journals, particularly the leading ones that researchers long to publish in, require plain language summaries and require them to follow the recommendations then they will begin to be implemented. Perhaps the International Committee of Medical Journal Editors will endorse them, and networks such as the World Association of Medical Editors and the Committee on Publication Ethics might encourage them. Funders, including pharmaceutical companies, are in a powerful position to require researchers to produce summaries and follow the recommendations—and they clearly have an interest in the research they fund, reaching as many people as possible.

My expectation is that the recommendations will be increasingly used—just as other reporting guidelines have been—and that the audience for scientific literature will steadily grow as a result, increasing scientific literacy and making us better able to respond to the next pandemic.

Transparency

Declaration of funding

No funding of this work is to be declared.

Declaration of financial/other relationships

As stated in the editorial, RS has a long association with journals and has a pension from the British Medical Association, which owns The BMJ. He has chaired meetings of Open Pharma, including one that discussed the recommendations on plain language summaries, and has been paid for his time. He has not been paid to write this piece.

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Acknowledgements

Editorial assistance was provided by Velissaria Vanna of Oxford PharmaGenesis Ltd. No further writing assistance is to be declared.