Open Health Data

Open health data encourages innovation & entrepreneurship, improves transparency in our healthcare system and most importantly turns healthcare into health for our digital citizens.

Posts by @HealthEugene & krabczuk

The 2009 Digital Britain Report described data as ‘an innovation currency’ and ‘the lifeblood of the knowledge economy.’  We are now in 2013 and while there is tremendous buzz around open data in general, open health data is definitely lagging behind.

I have been a great proponent of the movement for a number of years after being inspired by Todd Park at a Health 2.0 NYC Chapter event. But it really clicked with me when I saw three young entrepreneurs mashup various environmental and health data, create an MVP app in 6 hours and win two prizes at an open data hackathon. These three students are on their way to starting a company and making a difference in this world while helping healthcare consumers make better decisions in their everyday lives. This is the power of open health data! We, the citizens, ultimately own the data, not our governments and while there is certainly a need to preserve our privacy, there is a lot of “innovation currency” locked up in vaults, desperately waiting to be unleashed.

Below, you will find a brief report (50 slides, but don’t get scared!) that Katarzyna Rabczuk and I put together.  It showcases how nascent this movement really is, while showing samples of social and economic impacts of these initiatives across the US, UK and a select few Western European countries picked at random.

The United States is undoubtedly leading the way with and almost 400 valuable datasets published, ranging from Medicare data to epidemiology. Health Datapallooza is already turning 4 with the next event taking place in June of this year.

The United Kingdom is right behind (or ahead, depending which side of the pond you are on) with Tim Kelsey pushing forward and “unleashing the power of the people to save the NHS from a crisis”. The next NHS Hack day will take place on January 26th-27th in Oxford and some of the recent initiatives to open up prescription data generated a tremendous amount of buzz after a team that included two startups,Mastodon C & Open Healthcare UK as well as Ben Goldacre, published a report that showed how to save the NHS ~ £200M – this news reached even The Economist.

Unfortunately, the rest is very much of a long tail story and as you will see, the economic and social impacts dwindle as we travel outside the US & UK. In some cases, like Germany, we really needed to stretch to find an example of water quality (well, it could impact health) of swimming places around Berlin!

While we have spent the last few months compiling the data, this overview is not meant to be a comprehensive report on all the global initiatives, funding models, health outcomes or economic activity surrounding this movement. We strongly believe that open health data is one of the major keys to bridging the gap between digital citizens and governments and a great way to engage with grassroot communities of evangelists, private enterprises and not-for-profit organizations. We would love your feedback, additional examples and honest and open (pun intended) conversation on the newly created Google+ Community and LinkedIN group.

Healthcare is complex, intertwined, and touches every single individual on this planet. The average per capita spent on healthcare costs has been tremendously increasing over the past few years. Governments seem to focus on increasing premiums, changing tax rates and augmenting efficiency gains. A great article by Clay Christensen briefly illustrates that the focus for dormant capital has shifted to efficiency gains instead of innovation. And as we all know… innovation is what this world needs.

During November’s Health 2.0 conference in Berlin, Tim Kelsey -Patients and Information Director at the NHS- stressed the importance of open data. Tim wears many hats, including a role as a CIO, CTO and pseudo-CMO. No matter what the title states, his speech was inspiring and while he made some bold statements about giving all patients in UK their own health record by 2015, the one thing that resonated was his passion and dedication to open health data. As he puts it: “open data will unleash the power of people and save the NHS from a crisis”.

There are numerous reforms being pushed across the world. Whether a country has completely socialized healthcare system, a completely privatized system or anything in between, the issues are all the same: an utter lack of transparency.

Cost transparency is being viewed as a no brainer in many industries, except in healthcare. Do you know the true costs for procedures, tests and treatments whether you are in the ER, visiting your GP or heading over to see your dentist? Let’s not even talk about quality metrics that are associated with any of the above visits. What’s the discharge rate for a particular hospital and especially: what are re-admission rates and the root causes for these?

What is the solution?

I would argue that open health data plays a big part in the solution.

  1. First and foremost: by opening healthcare data, government organizations will start bridging the divide between governments and the digital citizens. A great portion of our taxes goes towards healthcare; the argument that it is “our data” still holds true and is even more relevant now.
  2. Public, private, NGO and civic society collaboration is key to inspiring and fostering innovation and entrepreneurship. Again, groundbreaking innovation is what is needed across the globe. Large corporations and governments are simply not suited to provide this level of progress. However, “staying in the race” can also lead to policy changes that inspire innovators; lowering tax brackets (unlike the French), opening up data and engaging with the community are all fundamental points that need to be addressed.
  3. Scientific research, grants and incentive systems for the aforementioned points need to move towards opening and sharing data across universities and other academic institutions. By the virtue of rewarding a large grant and publishing a paper, the incentives for the researchers are flawed. This leads to a negative mentality and a disincentive to opening “their” data.
  4. Finally: transparency is needed. It’s needed now and in real time; it’s needed to understand the money flows and inefficiencies of healthcare organizations; it’s needed to drive quality into these institutions and it’s needed in order for the civil society to have a valid input into the future of health and not just healthcare.

All of the above points are difficult to achieve. It takes time and perseverance, but I will voice some challenges for the future of open health data and also provide suggestions on how to mitigate these issues:

  1. Resistance of government entities to publish the data. Yes, resistance and concerns are still hampering the data from being used. It’s thought to be employed against politicians. It’s thought to be incomprehensive for those who ask for it. It’s thought to be mangled before public consumption. These outlooks will require government agencies to spend the scarce resources to “defend” themselves instead of working towards the goal of open data. My own rebuttal is that for every single report that will be wrong, there will be multitude-fold reports that will argue the contrary. Digital citizens, who understand the data with the goals of the social good, will consequently help these agencies squash the nay-sayers.
  2. Lack of access to data. Government agencies may have no data (shocking). This could be the result of a completely unwired system or a completely privatized system. I will focus on the latter though would welcome discussions on the prior. Care providers or insurers are approving, adjudicating and paying for healthcare claims. They therefore control the data flow. The quality metrics bestowed upon these entities are yet to be available and incentives to open the data are not present. If I am an insurer, what is my incentive to open up claims data and potentially lose my differentiation in the clinical analysis, risk premium calculations and more? I love capitalism and hence my proposal is for these private entities to open up the data, let the entrepreneurs come up with solutions, fund these entrepreneurs and refocus on innovation and truly value-added service. Oh, and not to mention, we ARE the end consumers, thereby being the main ones needing to be informed about the details of the services offered.
  3. Tracking the results of the open health data initiatives is also challenging. There is a general feeling that all the code-a-thons, hackathons, and other open data challenges are producing great results but by the nature of “open” -tracking the results is very difficult. My proposal here is to identify and structure a Crunchbase type of system, call it with an open API to list the data sets and events. This tool will furthermore track the developers and entrepreneurs who make use of this data. A tremendous effort would be needed to implement the above-mentioned tool and it would have to be highly crowd sourced. However, if you are passionate about succeeding in the open health data movement, this would be the perfect platform to contribute and gain additional exposure for your newly generated idea.
  4. Finally, I will leave you with the word “data dictionary”. Yes, things are different everywhere, form ICD-10 codes that are modified to healthcare systems that don’t even use them at all. Having said that, the local entrepreneurs can focus in the local market, make sense of the data, normalize it and map it to global standards.

None of us truly can predict where open health data will be in the next decade, but I strongly believe that the “Golden Age” of healthcare is rapidly emerging and open health data may soon become a common reality.

“What marked the Golden Age was a sense of wonder and curiosity” – said a well-known Dutch historian Eric Jorink.

I would add that the times we are in, are much more then just curiosity and include budgetary constraints, austerity measures and declining quality of care. So what are you waiting for?