Secondary Logo

Journal Logo


Better safe than sorry

An observational study of #ESALondon tweets

Lukić, Anita; Novak Fumić, Nina; Gluncic, Vicko; Lukić, Ivan K.

Author Information
European Journal of Anaesthesiology: July 2017 - Volume 34 - Issue 7 - p 482-484
doi: 10.1097/EJA.0000000000000644
  • Free


Social media have recently emerged as an unavoidable mechanism for dissemination of conference content, and among those, Twitter, a micro-blogging platform, has a special place. It has been used for conference coverage by a number of medical fields, such as resuscitation and emergency medicine,1–3 but no data are so far available on the use of Twitter in the context of anaesthesia meetings.

With that in mind, we set out to learn about Twitter activity built around Euroanaesthesia 2016 ( To do so, we accessed Twitter's application programme interface searching for tweets containing the official conference hashtag (#ESALondon) and published from 25 May to 5 June 2016, using the R programing language.4

We retrieved 4117 tweets, created by 717 user accounts. The number of tweets on the London conference was approximately double that at the previous year's meeting in Berlin, whereas the attendance was roughly the same (∼7000; Combined, those figures can be seen as growing interest of ESA members in Twitter and imply an increasing impact of the conference on social media and, hence, society. However, one should note that only one-third of the tweets (n = 1527; posted by 417 users) were original content, whereas the others were re-tweets (broadcasts of previously published tweets).

Furthermore, it is insightful to consider tweeting on the #ESALondon in the context of other major meetings, such as the Society for Academic Emergency Medicine in 2011 (#SAEM11), the International Conference of Emergency Medicine in 2012 (#ICEM2012) and the European Resuscitation Council (ERC) meeting in 2015 (#erc15prague). The ratio of tweets per delegate in London was approximately 0.6 (4117 over 7000), which is twice as high as the tweet/delegate ratio of 0.3 at #SAEM11,3 but several times lower than the tweet to delegate ratios at #erc15prague (3.1)2 and #ICEM2012 (1.95).1 When considering the proportion of original tweets, #ESALondon with 36% was comparable with #erc15prague,2 whereas the fractions of original tweets at #SAEM11 (67%)3 and #ICEM2012 (69%)1 were twice as high. If we put together these numbers with the observation that many delegates were not involved in social media, we see ample opportunities for improvement. For instance, ESA could offer crash courses on Twitter at a pre-designated Twitter hotspot or include a session on social media. The congress hashtag could also be made more prominent on the congress webpage and marketing material and could be referred to during the opening ceremony.

By using the number of re-tweets and favourites per tweet as a measure of a tweet's popularity, we observed that the most popular tweet (21 re-tweets and 12 favourites) was posted by @Manu_Malbrain: ‘Well 50 bags of regular salt potato chips = 1 bag of [0.9%] saline #ESALondon’. That is an excellent example of an educational tweet: an informative and easy-to-grasp illustration of the actual amount of salt in a bag of saline (0.9% NaCl, which translates to 154 mmol of sodium per liter). The second most popular Tweeter entry (seven re-tweets and 15 favourites) was made by @traumagasdoc: ‘So …. Surgeons aren’t terribly good at assessing muscle relaxation according to the #ESALondon this morning’. This points to a very important topic, that is difference in assessment of muscle relaxation between an anaesthesiologist and a surgeon. Although of high clinical relevance (safety!), the issue is only sporadically addressed in the literature in the form of case reports.5,6

Next, we explored the text of the tweets by creating a word cloud of the most frequent words. As seen in Fig. 1a, those included ‘esa’, ‘esahq’, ‘safety’, ‘airway’, ‘management’ and ‘care’. Moreover, to learn more about the content, we classified tweets into topics by ‘manual’ revision of the original tweets by two independent reviewers, using previously published criteria.1 Most of the tweets in London (58%) were session-related, 19% social, 12% logistic, 9% advertising and 2% other (inter-rater agreement: Cohen's κ = 0.950). The most prominent session-related topic was safety, followed by airway management and sepsis. Combined, those results illustrate the attendees’ appreciation of, and preference for, certain topics, and, in turn, they give the conference organisers suggestions for future events.

Fig. 1
Fig. 1:
Analysis of tweets commenting on Euroanesthesia 2016 – official meeting of European Society of Anaesthesiology in London, 2016. In total, 4117 tweets containing hashtag #ESALondon and published between 25 May and 5 June 2016 were retrieved. (a) Word cloud showing the most frequent 200 words appearing in 1527 unique tweets. Word size is proportional to word frequency. (b) Network of Twitter users commenting on the conference (plotted by Gephi 0.9.1; The Gephi Consortium, Paris, France). Each node represents a Twitter user (node label), whereas the lines connect a user who tweeted and a user who re-tweeted that post. The nodes (for clarity, only nodes with degree range more than 7 are depicted) are sized by the betweenness centrality.

Switching focus to exchange of tweets, we matched original posts and their re-tweets and used that as a basis for network analysis. As shown in the network chart (Fig. 1b) the network of people tweeting about #ESALondon was a homogenous group, with a few users dominating the flow of messages. Therefore, it seems that the dissemination of information and knowledge on anaesthesiology can be augmented by directly addressing Twitterers with high network resonance (e.g. @Fluid_Academy, @ICU_Management and @eurtraumacourse).

In summary, this is the first study providing information on the use of Twitter as a social media platform accompanying anaesthesiology meetings. In addition to being an illustration of a colourful Twitter activity by #ESALondon delegates, our results can serve at least two important purposes. First, they highlight the topics of particular interest for the conference attendees and second, they could be helpful to the organisers as guidelines for the upcoming events, such as the 2017 ESA meeting in Geneva.

Acknowledgements relating to this article

Assistance with the study: we thank Dominik Dsazuta (@rubbadoc) for his comments on the conference content.

Financial support and sponsorship: none.

Conflicts of interest: none.


1. Neill A, Cronin JJ, Brannigan D, et al. The impact of social media on a major international emergency medicine conference. Emerg Med J 2014; 31:401–404.
2. Lukić A, Lulić I, Lukić IK. Education first: content analysis of #erc15prague tweets. Resuscitation 2015; 106:e21–e22.
3. Nomura JT, Genes N, Bollinger HR, et al. Twitter use during emergency medicine conferences. Am J Emerg Med 2012; 30:819–820.
4. R Core Team. R: A language and environment for statistical computing 2008. Available at: [Accessed 26 May 2016].
5. Lamberg JJ, Answine JF. A surgeon's assessment of inadequate neuromuscular antagonism in a case of prolonged neuromuscular blockade. J Anaesthesiol Clin Pharmacol 2013; 29:244–247.
6. Welliwer MD, Jones WH. Interdisciplinary intraoperative communication and collaboration needed for optimal neuromuscular blockade management. J Anaesthesiol Clin Pharmacol 2014; 30:442–443.
© 2017 European Society of Anaesthesiology