Blue Button Drama
The recently launched Blue Button for All Americans Challenge has brought to the surface an interesting debate about the best path to health data interoperability. The Department of Veterans Affairs (VA) wants to encourage widespread use of the Blue Button personal health record (PHR) based on their success with the Veteran patient population. To do this, they’ve put forth a challenge that will award $50,000 to the PHR developer which gets 25,000 clinical professionals to install a Blue Button compatible PHR on their patient websites. However, what sounds like a promising step for clinical transparency and patient engagement, has been met with harsh public response.
Soon after the Challenge launched, Fred Trotter posted a public comment on the Challenge.gov Blue Button page accusing the Challenge of not being in line with meaningful use. The opening line of the comment simply read, “This contest is deeply problematic. Either you are promoting the notion of patient access, which is wonderful, or you are promoting @#$% health data standards, which is unethical.”
If you’re not familiar with Fred Trotter, he’s a self-identified “Hacktivist” who is probably best described as the Chuck Norris of participatory medicine. While very professional and well spoken, he has little tolerance for illogical ideas and does not hesitate in making his opinions known. It’s also important to note that O’Reilly Media is helping Trotter publish his most recent work Getting to Meaningful Use and Beyond.
Trotter’s main complaint is that the Blue Button data format encouraged by the requirements of this Challenge is neither readable as a summary printout nor parseable as XML (a format that would allow for much greater interoperability in the future). He would rather see any advanced format, like the existing CCR/CCD formats, included in the rules of this challenge and points out that the Challenge objectives are redundant given what is already required by Meaningful Use. According to Trotter, the Blue Button format seems to have no formal standard and is the worst compromise between human and computer readability. With a format not rich enough to transfer into another health record the VA is running the risk of rendering Blue Button data as nothing more than a patient “FYI.”
On his own blog Trotter writes: The blue button initiative was a good initiative because it allowed greater access. It made that possible by ensuring that access to patient data did not have to wait for the VA/DOD/Whatever to create a download that conformed to the still-forming XML standards that make true interoperability possible…The innovation of Blue Button is to not let standards compliance get in the way of access. For this reason the DOD and the VA use of the Blue Button format should be applauded! The Blue Button initiative was fully give-me-my-damn-data compliant! But promoting it as an alternative to a process that supports fundamental data reuse and is -already- required by meaningful use is unethical… I formally request that the VA withdraw this contest, or make it clear that a CCR/CCD/printable report download meets the requirements of the contest.
If that wasn’t enough to have posted on Blue Button’s public Challenge page, then, Adrian Gropper, founder of MedCommons and advocate of doctor-patient collaboration, responded with equal force. Gropper stated that neither CCR nor CCD formats are human readable and that the very XML formats Trotter praised are so complex, they poses significant barriers to innovators. Further more, Gropper claims that these XML formats have “yet to prove themselves as a solid foundation for interoperability.” He also points out that even PCAST has gotten to the point where it is focusing on metadata rather than the XML itself; leaving many to wonder if interoperability would be better served by JSON.
A diplomatic post from a VA organizer quickly backed Gropper citing something about the importance of interoperability and the thread has lay quiet ever since.
For an initiative that started off with a core goal being data portability for the express purpose of future clinical use, the VA’s lack of clarity for how this Challenge might alter the evolution of the blue button initiative is surprising. In not providing resources on how these standards compare or how they might affect the future functionality of PHRs, the VA is leaving a lot up to chance and in the hands of developers that might not be invested in these long-term issues.
Designing a challenge is no easy task. If managing our own Developer Challenges has taught Health 2.0 one thing, it’s that the devil is in the data. Regardless of how usable CCR or CCD specifications may be today, they should still be included in such initiatives if they truly offer a richer data set or more complete patient profile.
Technical-lock in is very real. While we could come up with a few EHR examples, we’ll play it safe and just site the ongoing influences of standards like MIDI and QWERTY. The Blue Button format may offer greater functionality and patient participation in the short-term but it’s unknown how it will serve us in the long run. Although there are the other feisty threads appearing on Blue Button’s Challenge page, we found the Trotter-Gropper debate particularly compelling because it highlighted this riff between the need to deliver data today and the importance of fostering the meaningful systems of tomorrow.
It’s fantastic that the VA, Trotter and Gropper are all in support of open source design and patient data-rights. But open source and patient access aren’t the issues here. Their divergent views capture an important debate about the role organizations should play in the evolution of health data interoperability and how best to integrate new talent who may not find such acronym riddled discussion threads (human or computer) readable.
Health 2.0 Developer Challenges