Health Datapalooza IV (#HDPalooza) just ended last night as the fourth annual national conference born from government efforts to liberate health data. Two conference days concluded with 11,141 tweets created! An amazing feat! In the same spirit of liberating health data, we have taken the data from Twitter and visualized it in the network centrality analysis shown above.

The larger the node, the more influential or central to the graph. Links between nodes signifies one of more direct conversations or mentions between two Twitter Users.

Download the PDF version to zoom in on the graph or even search for specific Twitter Users.

What you see above a scatter plot of all #hcsm tweet chats and their tweet volume that we’ve got in our database going back to 2010. These tweet chats have taken place every Sunday at 9PM EST. As you can see, #hcsm is still going strong so far in 2013 with an average of 800 tweets during the chats.

The most popular chat in terms of number of tweets took place Sunday 20th of March 2011 09:00 PM EST with 1,541 tweets. You can read the transcript of that historic chat here. Patients were the focus of that chat with topics including the epatient-physican relationship and patient education information online. From the bubble chart below you can see some of the actual issues being discussed.

http://www.conectedhealthasia.com

www.allnycem.org

Annual symposium of Thames Valley Faculty RCGP for established GPs and trainees across Oxfordshire, Berkshire, Buckinghamshire and Northamptonshire.

Venue: The Oxfordshire Hotel and Golf Club, Rycote Lane, Milton Common, Thame, Oxford, OX9 2PU

Programme: http://www.rcgp.org.uk/courses-and-events/south-england/thames-valley-faculty/~/media/Files/RCGP-Faculties/South-England/Thames-Valley/symposium-2013-flyer.ashx

Booking: http://www.rcgp.org.uk/courses-and-events/south-england/thames-valley-faculty/thames-valley-faculty-symposium-all-day.aspx

In the two previous posts in this series of Twitter visualizations, we’ve looked at the growth of patient communities and the dynamics and centrality of one of them. This time we will take a deeper look at the people who make up these communities.

From looking at the visualization below (PDF version), one can think of them as virtual communities set in space like constellations of stars but with people orbiting one another. The laws of physics still apply, but gravity has become a function of conversations. Conversation is gravity. Boundaries are broken down. Traditional barriers like location, profession, demographics, physical abilities and conditions will not stop you from joining in. The world is flattening. And because many traditional hierarchies and perceptions are eroding as a consequence of healthcare social media, new types of communication and communities are emerging. Twitter is the new melting pot of ideas, people, and disruptive innovation.

Is it this melting pot of people, experience and knowledge that makes Twitter so attractive for patients that they are willing to trade some privacy in return for the benefit of joining in? In our series we’ve categorized these communities as patient-centric, but as many participants can testify, the members are made up of more than just patients. Our focus this time is the #BCSM community, Breast Cancer Social Media. Below we have pointed out some of the patients represented (not all are highlighted).

In our first post we created a visual from about 100 million tweets over almost 2 years, while in our second post the time span was one month. This time we are looking at a snap shot of only 1 hour. But this is not just any random hour, it’s the hour of the weekly scheduled #BCSM tweet chat. These tweet chats are the epitome of a melting pot – stakeholders from many different professions exchange and discuss ideas relating to a few pre-defined topics compressed into an hour time frame. The energy can be explosive. This visual is a network centrality analysis. The larger the node, the more central or influential a member is. Links between nodes signify a direct interaction or conversation between the two. In the following visual we point out some of the healthcare providers represented (not all are highlighted).

Twitter provides a convenient and powerful place for patients to meet other patients where they can both receive information but also give information from their own experience – it’s that human need created in us to share knowledge which has powered much of social media. However, from the visual above where we’ve highlighted nurses, physicians, a pharmacist, etc., we can see a larger view of the attraction Twitter has for patients.

Many patients on social media are very engaged, many are living with chronic conditions, cancers, or rare diseases, and many have more than just knowledge to impart to others – they want to change the way healthcare is delivered. Change can only come about in a public space, you’ve got to step out from the echo chambers of private forums. The open and public nature of Twitter provides a platform where people can find a cross section of all the relevant stakeholders in healthcare. A birthplace for disruptive ideas. The perfect place for engaged patients, engaged providers, and the supporting third parties.

For over 2 years, Symplur has collected health conversations on Twitter. From a humble start, our infrastructure has grown and matured to scale with the tremendous growth of healthcare social media. We’re now approaching 150 million health tweets in our database which we analyze and categorize by thousands of variables. And similarly with other repositories of big data, our greatest challenge is to figure out how to present insights from such a vast dataset in a meaningful way. The answer is often found in visualizations.

We plan to start a small series of short posts displaying some of this data in a visual way.

What you see in the video below is a 22-month timeline of about 2,000 different health communities and topics each visualized as a bubble. In total, about 100 million healthcare tweets are represented in this visual. The green dots symbolize patient-centric topics, while pink encompasses more professional/provider topics. Larger bubbles signify larger volume of conversations within that community. The data is visualized dynamically over the this 22-month time period starting September 2010.

Why the growth of Patient Communities on Twitter?

What we discovered was somewhat of a surprise. From the start, many considered Twitter as a kind of virtual water cooler, mostly used by healthcare professionals. It’s been thought that the need for privacy would push patients to more closed platforms. Twitter as you know, is totally public and should not be considered private.

However, from what you can clearly observe, the green bubbles have grown in numbers and significance quite dramatically in past months. This indicates a strong growth of conversations in existing patient communities on Twitter and also a growth of new patient communities on Twitter within our dataset.

Why this growth? Why would patients choose to utilize a communication platform without privacy? We will continue this visual journey trying to give possible answers to that question.

Update: The second post in this series is now published –  The Dynamics of a Twitter Patient Community – Network Centrality Analysis