Dissemination of medical publications on social media - is it the new standard?

Abstract

The way we all access information has changed beyond recognition over the last decade 1. Most people now access some, if not all, their news through social media platforms, and instant messaging has superseded telephone and written communications 2. Medical knowledge dissemination is also evolving, and journals now signpost newly published research and opinion to an army of readers, followers and postpublication peer reviewers on social media. This army includes professors, experts, clinicians, healthcare workers, patients, carers and anyone with an interest and a social media account. Anaesthesia has an active social media presence, with initiatives including a regular ‘TweetChat’, where interested parties, including authors, are brought together on one platform to discuss a newly published paper 3. The recent #Hyperoxia TweetChat, on a paper about evidence for compromised data integrity in studies of liberal peri-operative inspired oxygen 4, generated ~60,000 impressions (the number of times Tweets from @Anaes_Journal were seen) on the day of publication. Such initiatives must surely increase attention, citations and the speed with which the benefits from scientific research yield improvements in clinical care, but social media analytics is a new science and it is not yet clear how these metrics should be interpreted. In this month's issue of Anaesthesia, Ng et al. describe how the National Tracheostomy Safety Project (NTSP) was publicised to front-line healthcare professionals through commissioning a multimedia marketing and production company to create content and promote it using paid social media advertising 5. The company created high-quality multimedia content, including videos, and promoted these to a carefully selected target audience. The authors spent a total of £4140 on social media advertising, £0.02 for each view and £0.67 for each interaction. They also spent money on other areas, such as filming and editing, and redesigning their smartphone application and website. Although they provide useful data on the impact of their targeted paid social media dissemination strategy, there are several areas we believe require further discussion. Data on whether the content reached the intended audience do not appear to have been captured by Ng et al. Engagement statistics, which the authors use to demonstrate a successful promotional campaign, is only helpful if those viewing the content are those caring for patients with a tracheostomy in their day-to-day practice. To extend this reach, users, often described as ‘influencers’, with established credibility and access to a large audience, could have been targeted. We do not recall seeing the NTSP material on our social media feeds, which was a surprise to us, given that we are active and engaged on various social media platforms and probably in the middle of the target audience. Regardless, analytics such as ‘reach’ and ‘impressions’ might not reflect the true impact of content disseminated through social media. How best to measure the ‘true’ reach of the NTSP campaign has no easy answers, as there were no sales figures or tests of knowledge before and after exposure to NTSP content. Word-of-mouth strongly influences the attitudes and behaviours of patients and providers when making healthcare choices 6. During the paid marketing period, two-thirds of content sharing was peer-to-peer among networks of individuals. We may, therefore, be cynical and less receptive to paid attempts from third parties to attract our attention. Professional medical associations and bodies have strong trust, collegiality and networks. They might nevertheless be reluctant to click or share ‘promoted’ material, fearing the storage of personal interests by social media companies or due to a distrust of such promotions. Ng et al. pro-actively dealt with this by requesting stakeholders’ participation in content dissemination, which seems to be an innovative and powerful solution. Promotion using paid social media advertising may act as a catalyst when encouraging others to share content and increase reach. The targets of paid medical promotions are probably those individuals who may not use social media to access medical content, but have shared enough information to identify themselves as a medical professional. They may see a video about tracheostomy care, while scrolling through a timeline with photos from, for example, a friend's wedding. Regardless, individuals from the online medical social media community will usually always see important information freely shared through their network. Promotion with paid advertising may, therefore, be a useful tool to reach the places that peer-to-peer channels cannot. Perhaps the most alarming critique of medical knowledge dissemination through social media is a perceived lack of formal peer review and the risk of ‘fake news’. Vosoughi et al. fact checked 126,000 Twitter stories and found that falsehoods reach people faster than the truth 7. Among the stories analysed, and although the truth rarely reached 1000 people, the top 1% of false stories were expected to reach between 1000 and 100,000 people. Additionally, the truth took six times as long as false stories to reach 1500 people. Twitter ‘robots’ were found to spread fake news and truth equally, suggesting that it is we, the humans, who are responsible for the preferential like, share and retweeting of fake news over the truth. False stories tended to be novel, and this novel information was more likely to be spread amongst peer-to-peer networks. Twitter accounts spreading false stories had fewer followers, followed fewer people, were less active and were more likely to have been registered more recently. At the same time, and as we have seen in our own specialty 8, 9, academic publishing is not free of scandal where the spread of fake news is concerned. Concerns around the measles, mumps and rubella (MMR) vaccine persist, and have recently been recirculated on social media, despite countless widespread scientific rebuttals of Andrew Wakefield's now retracted paper 10. More recently, the World Health Organization came under pressure to review their surgical site infection guideline following concerns about data integrity 4. Doubts were previously expressed in an online blog by HJ many years before these concerns emerged, and the accompanying comments provide some interesting historical context (https://traumagasdoc.wordpress.com/2016/11/13/perioperative-oxygen-and-surgical-site-infection). Medical publishers and journals are already engaging their readers on social media with co-ordinated unpaid campaigns. Active promotion of this content is still in its infancy. It is likely that, in the future, journals may look to specialist social media consultants to increase their impact. There will, however, always be a requirement for a ‘content expert’ overseeing these activities, especially in specialties such as ours. As the sophistication of these strategies increases, poor quality research may receive undeserved publicity and attention simply through a successful social media strategy. The tongue in cheek ‘Kardashian Index’ quantifies the discrepancy between a scientist's social media following and academic citations 11. One possible solution is to make research promoted on social media free to access on the day of promotion (as does Anaesthesia), as this enables users to engage more deeply with and fact-check the manuscript content. Participation in social media does carry risk of reputational damage, for both individuals and organisations. A poorly thought-out tweet, an attempt at humour or a misplaced selfie can all become problematic. Communications professional Justine Sacco tweeted, boarded her flight to Africa, and by the time she landed, had become international news and lost her job. Regardless of privacy settings, every social media post can be captured using a simple screenshot and shared (https://www.nytimes.com/2015/02/15/magazine/how-one-stupid-tweet-ruined-justine-saccos-life.html). Despite this, with thoughtful use, we argue that the educational benefits of engagement in social media outweigh these risks and recommend individuals and organisations follow guidance from employers and regulatory bodies. There has long been a disconnect between the publication of research findings and their incorporation into clinical practice 12. For example, we have not yet eliminated the term ‘nil by mouth from midnight’, despite guidance to the contrary published in 2011 13. Intravenous gelatins are still widely used in clinical practice, despite yet more evidence of risks outweighing benefits 14. One possible explanation is that those delivering day-to-day patient care are not engaged with the academic literature. Another, is that this literature may fail to address the problems with the way in which care is delivered 15. We agree with Ng et al., that the way in which clinical staff access information at the bed-side is much more amenable to videos, infographics and short headlines viewed on smartphones as compared with a full paper published in a journal and viewed on a hospital computer. These computers are usually unreliable, slow, inaccessible and with badly designed interfaces and slow log-in processes. Once in the system, there are organisation-imposed restrictions limiting access to social media platforms. These problems may take the clinician or carer away from the patient and simply make working life more difficult 16. Ng et al. suggest clinical staff facing clinical problems prefer to abandon hospital computers and rely on their own portable devices 5. We suggest an important finding from their work is that hospital managers consider removing these barriers and allow clinical staff to access the information they require in whatever format they wish. The mainstream media has adapted to the need for their customers to access information through social media platforms – medical publishers and healthcare employers must now do the same. Here lies an enormous opportunity, as social media channels are equally accessible by the healthcare assistant, nurse, physiotherapist and patient as they are by the professor. Their social media timelines are probably very different, and the professor will probably see the publications anyway, but with the promotion of succinct educational material, important messages for clinical practice can be delivered to those who care for patients, day to day. We argue it is inevitable that there will be more rapid improvements in patient care, more so when medical publishers better understand how to target audiences with paid social media advertising. There is now a need to think about how this can be achieved and how analytical outputs such as reach, engagements, clicks and views should be measured and compared, along with judging value for money. For getting us started in this endeavour, Ng et al. should be congratulated. HJ is a council member for the Royal College of Anaesthetists. TS is a member of the International Advisory Panel for Anaesthesia. No other competing interests declared.