It’s About Time? Mapping Physician Twitter Networks

Mishori, Singh, Levy, et. al. writing in JMIR offer the first I’ve see mapping physician Twitter networks. The article is titled “Mapping Physician Twitter Networks: Describing How They Work as a First Step in Understanding Connectivity, Information Flow, and Message Diffusion”

The objectives were to describe the characteristics of four medical networks, analyzes their theoretical dissemination potential, their actual dissemination, and the propagation and distribution of tweets.

The four networks were, The AMA, the AAPF, AAP, and ACP. Visualization was used to determine overlap between the groups, actual flow of tweets for each group was assessed and examined using a Twitter data aggregator Topsy. 

Results showed that overlap across groups is small limiting the community cohesion and cross-fertilization. AMA followers’ network is not as active as the others though the AMA posted the largest number of tweets while the AAP posted the fewest. Retweets were low showing sharing of information was well below potential. 

The authors make an important observation, dissemination of information has a huge potential in these networks. The more individuals within a group Tweet the larger the dissemination. The following is striking 

…the percentage of followers that have not sent any tweets is 6.92% for AAFP (522/7546), 8.17% (962/11,768) for AAP, 7.22% (430/5955) for ACP, and 18.43% (39,275/213,122) for AMA.

It shows that when those who have not sent any tweets are removed from the dissemination network the potential only decreases by 1%. Yet when all followers who have sent only 10 or fewer tweets are removed then the AMA group information dissemination potential falls by over 35%. The other professional groups impact remains at less than 1%. The AMA network is not as active as the others and the AMA has a larger portion of individuals who are not tweeting. 

Retweeting is key in the sharing and dissemination of information and the potential of networks to drive information. The authors examined tweets sorted by number of retweets. Since the authors only examined dissemination up to Level 2 followers and not beyond retweets may be disseminated to a great extent. 

The number of retweets and the number of individuals who received the tweet is less than 0.2% of the total potential. 

Conclusions: To increase the dissemination potential, medical groups should develop a more cohesive community of shared followers. Tweet content must be engaging to provide a hook for retweeting and reaching potential audience. Next steps call for content analysis, assessment of the behavior and actions of the messengers and the recipients, and a larger-scale study that considers other medical groups using Twitter.

This examination of physician networks and Twitter is fascinating and rich in information. It points to more study and hints that content is key. Why aren’t I surprised. I know for me on Twitter it takes time to get comfortable with it and what I retweet is driven by its value to me or to others who I think it will have value for. So that may be part and parcel of this study the need to look at content analysis and beyond. 

I see this study demonstrating how similar Tweeter is to Communities of Practice. CoP can be defined as 

Communities of practice are groups of people who share a concern or a passion for something they do and learn how to do it better as they interact regularly.

This fits with the goal of the authors to look at content analysis. You and I are only interested in what we are interested in. How can someone in the AMA network with an interest in say diabetes management in patients with CHF find others with the same interest in the AMA and across other networks? They seem not to look across networks and are not keying content to need. 

I think the goal is to create mini Twitter networks that mimic CoP? That in my mind is the real potential for Twitter and other social media as information and learning platforms. Thus CoP behavior may occur outside Twitter occuring within institutions and among colleagues. Twitter, FB, G+ etc. are perhaps where physicians and learners go to find information specific to the problems they want to solve and from there they join smaller networks or communities.Yet I think we are a long way from that. Either online or off line. When you talk to me about problems I want to solve I am interested.

And there is this study which was referenced in this article “The Role of Social Networks in Information Diffusion”  The PDF is available on the hyperlink. No paywall. 

Online social networking technologies enable individuals to  simultaneously share information with any number of peers. Quantifying the causal effect of these mediums on the dissemination of information requires not only identification of who influences whom, but also of whether individuals would still propagate information in the absence of social signals about that information.

This Facebook study demonstrated the following:

We show that, although stronger ties are individually more influential, it is the more abundant weak ties who are responsible for the propagation of novel information. This suggests that weak ties may play a more dominant role in the dissemination of information online than currently believed.

This has been said in may places and studied. It is worth noting strong ties are our echo chamber, weak ties are those on the edge who will introduce us to something we may not have considered or known. And since we are a weak tie in return information we share may not known therefore they share what is new and different.