Over the last three years we at MDigitalLife have been evolving the way we analyze online conversations about health. Initially, our focus was primarily on topical analysis – and that’s still an important part of what we do. But now we’ve begun to take a much deeper look at the interpersonal dynamics of the people involved in those conversations and the underlying networks and communities where content flourishes. Now that we’ve identified online accounts according to stakeholder groups (what we refer to as the online health ecosystem) to cut through the noise of online conversations, how do we quantitatively evaluate these networks to understand which healthcare stakeholders are engaged and with whom? How do we evaluate the diversity of the networks? How does the volume of content combined with the quality of the network create a lasting discussion? And finally how can we move beyond crude and overly simplistic measurements like impressions to truly measure effective content spread in a network? Medical conferences provide the perfect opportunity to think through some of these questions, and with the ASH Annual Meeting only days away, we wanted to dig into our global health ecosystem data to use this as a test case to think through these questions and evaluate the evolution of the online conversations associated with ASH over the last 3 years (2012-2014).
While the volume of posts have exploded (increasing nearly tenfold between 2012 and 2014), it appears that stakeholder participation across the ecosystem has remained more or less constant over the past 3 years. Even though ASH is primarily an event for physicians to learn about and discuss the latest research in hematology, other stakeholders are participating as well. Surprisingly, physicians (globally) only account for 55% of the online conversation making this a very diverse network of stakeholders for a physician-targeted conference. In comparison, the American Urology Association (AUA) Annual Meeting (for one example) was 71%-84% physician participants over the same three years. The second biggest stakeholder group participating in ASH was advocacy organizations – both patient advocacy and physician organizations – at 10%-12% of conversation over this time frame. It may be surprising that with the increasing comfort of healthcare industry on online channels, industry participation is small at about 6%. It is also surprising that patients contributed as much to the online conversations over this time period as did industry. ASH’s patient participation at ~6% is significantly higher in comparison to AUA at 1%-2%.
Although the balance of stakeholder participation has remained constant, conversation volume has doubled every year since 2012. Not only does online conversation for ASH14 produce a much larger network in comparison to ASH12 (in addition to the significant growth in the number of posts, the number of individual authors rose accordingly), the difference between the two goes beyond just volume. It also produces a dense central network characterized by a high level of interaction among stakeholders. 2012 saw separate and isolated clusters of stakeholders whereas 2014 saw more connections across groups. Physicians are clearly in the forefront of this connection but smaller groups like patients are also embedded in that central node, representing influence that goes beyond number of participants and volume of posts. (*The same legend in chart “Stakeholder Participation in Online ASH Conversation” applies to network visualizations)
Interestingly, in 2012 conference participants created twice as much original content than they retweeted of others (in other words, they had a high post-to-share ratio). As tweet volume went up, retweets went up and the post-to-share ratio receded. In 2014, participants retweeted almost as often as they posted. Likes and retweets of health stakeholder content also increased, nearly doubling every year. Furthermore, the average participant was mentioned 19 times in ASH14 conversation and at least 75% of all accounts were mentioned at least 2 times. The average mentions per account has also doubled every year since 2012. This highlights the increasingly effective back channel that twitter has become for creating discussion before, during and after the conference for participants at the event and at home. This growth in engagement creates the expanding and densely connected network that we see in the ASH2014 conversation.
This is only the start of an algorithm we’re building to quantify and evaluate the strength of online networks. By better understanding the functioning of online networks, all of the stakeholders in the online health ecosystem should be able to better understand which conversations they should be participating in, and how to do so most effectively. At ASH, the signs are all positive … in addition to the obvious factors like the increasing volume of posts, ASH conference participants (in-person and afar) are following one another more, conversing with one another more, engaging one another more, and extending the conversation farther and farther beyond the timing of the conference – leading to a diverse network of stakeholders that are increasingly connected. We’re excited to see if this trend continues into the 2015 conference. Stay tuned.