Convert your FHIR JSON -> XML and back here. The CDA Book is sometimes listed for Kindle here and it is also SHIPPING from Amazon! See here for Errata.

Friday, August 9, 2013

They got me started

OK, so ONC got me started yesterday on this topic with their call for participation.

This project arises from the idea that it would be a way to execute on a memorandum of cooperation between the European Commission and the US signed by Secretary Sebelius back in 2010.  For some time nothing really happened.  Then there were some high level meetings between ONC and EU, and still, nothing really has happened.  And now we have an ONC S&I Framework project that wants to do the following.
  1. Create initial set of use cases, based on community and stakeholder input
  2. Compare the data/document structures used in the US and EU by comparing the consolidated CDA (C-CDA) and the exchange standards used in epSOS
  3. Compare US and EU vocabularies, terminologies and clinical models relevant to selected use cases to identify areas of overlap and commonality
  4. Create best-effort Template and Terminology Mappings for selected use cases based on an agreed methodology
  5. Create strategy briefs on long term adoption and sustainability of EU/US health record exchange focusing on 7 areas: cross-vendor integration; education; security and privacy; incentives; standardization; innovation; research
  6. Identify available resources and opportunities for aligning them (technology and standards to support ongoing collaboration with vocabularies, modeling, and interoperability)
  7. Create a common interoperability testing strategy including testing plan and tools
  8. Agree on specifications, standards, architecture, tests for the validation pilot
  9. Validate selected use cases in the transatlantic setting
  10. Release feasibility analysis document for EU/US health record exchange

#1: Use cases is how S&I works, and I don't recall a project that started without use cases.  In this case, though, I'd skip right to the next items, and save use cases for phase 2.  There's enough work to keep an army of experts busy for the next three years.  We already have the use cases in the US and in the EU.  If you must have a use case, call it medical summary be done.  Let's not waste time much more time on it.  It is after all, the most commonly chosen first use case in any regional program, and it's pretty clear from the rest that that is where we need some alignment.

#2: On comparing the data structures, I'm so glad you asked.  epSOS is principally based on IHE PCC Technical Framework plus extensions  mostly in support of e-prescribing.  CCDA comes from HITSP C32 which is additional constraints on the PCC Technical Framework.  They both originated from HL7 CCD.  IHE has been in the process of harmonizing its technical framework with CCDA for the last two years, as led by yours truly.  We are readying content for public comment in a few weeks, and are considering taking some of it through the HL7 ballot process.  Please use that process to your advantage in this activity.

#3: Comparing US and EU vocabularies and terminologies: My advice would be to focus on how to harmonize medication vocabularies.  That is the biggest problem internationally.  The second biggest problem is finding a vocabulary or value set that supports allergies. Problems and Labs are best addressed through international efforts like SNOMED and LOINC, especially given their recent announcement of cooperation.  Get out of their way, and I mean it.  Don't throw up any roadblocks or create any distractions; it has already been too long in coming. Vocabulary alignment is workgroup #4.  Which of those are important is Workgroup #3 (see #5 below).

#4: For template and terminology mappings, I really hope you'll take a look at the existing IHE work which also has been too long in coming.  It will address quite a bit of what you are trying to deal with here.  My advice: Don't redo, rather reuse.

#5: This is actually a really good one to do to start with. This should be a third workgroup though.  It's neither tactical nor technical, but rather strategic.  Let's not confuse them.

#6:  This is strategic and gets assigned to workgroup #3 again.  And if you want to align resources, it would be good to start talking with them before just haring off and doing, and expecting them to follow your lead.  ONC and the EC could quite readily have approached IHE, ISO, HL7 and other SDOs to create a global summit addressing standards alignment.  But it seems that ONC thinks that S&I Framework is the right place to start that.  I'm thoroughly unconvinced, and mostly because of how my International colleagues view ONC. At the doer level, ONC doesn't have the reputation or standing to pull it off.  They have done more to annoy International standards developers than they have to recruit them to their cause.  ONC stumbles around the international front like a clumsy American tourist in Paris who thinks that people will simply understand them  if they speak slowly and loudly enough.  These might be my words, but I'm echoing what I'm getting back from the International community.

#7: I can think of at least on organization that does testing internationally and is already quite involved in EU projects.  Have you considered working with them? Hint: They do this thing called a Connectathon.

#8. Save this for phase 2, after we've done some aligning, and identified a common use case.

#9. Again, save this for phase 2.  And do you mean test it and prove out the concept, or do you mean pilot something, or is there some expectation that something useful and lasting would come out of this?  I want to see the latter, not the former.  Please save your pilot for something that you are pretty sure someone would pony up for, rather than hoping that if we build it, they will come.

#10.  Phase 2.  Like I said, there's enough work for multiple years in this proposal.  Let's work first on alignment around a particular topic (e.g., templates for a patient summary), and then move on to actually doing it once we have it.

So, overall, recommendations, break this into some doable phases, coordinate with international SDOs, and add a workgroup dealing with strategy and one on vocabulary, and finally, pay attention to what others are already doing in the space.


  1. Keith, Thanks for your insightful and cautious advice. You rightly say that "enough work to keep an army of experts busy for the next three years." I am struck in particular by your emphasis on medication vocabularies and allergies.

    There seem to be two issues: the role ONC should play in international affairs and the broad range of urgent tasks on the domestic front.

    Few people have your breadth of experience and perspective, so I ask: If ONC were charged to address only 25% of the activities in which they are currently engaged, what should their priorities be?

    Thanks as always from a reader who has learned so very much from your book and your postings.

  2. Actually I would start with the use cases. And maybe a cost/benefit analysis. Just how often does a patient have healthcare data in both the US and the EU? How much are we going to spend to support how many transactions?

    1. Ability to share medical records would obviously be one benefit. However, a second benefit (and likely the larger one) would be the ability for vendors to work to a common standard on both sides of the pond and the ability to combine development efforts. That would likely both produce savings and increase adoption speed. Those benefits would need to be considered in any cost benefit (and are significant enough benefits that I doubt little analysis would be required).