Two years ago I wrote Without Profiler-Enforcers, Healthcare IT Standards Cannot Enable Interoperability for Gartner clients. It argues that achieving interoperability for clinical data requires a single entity with three characteristics: (a) the ability to synthesize the work of standards bodies into a coherent, purpose-specific specification; (b) end-to-end responsibility that runs from concept formation through actual, measured, large-scale adoption of the standards; and (c) economic clout.
The ARRA has put us closer to that position than I imagined was possible. Working backward, item (c) is to be achieved by various HITECH programs and the meaningful use incentives; and item (b), the end-to-end responsibility is at the level of the ONC, not a subordinate entity tasked with that purpose. Achieving (a) is the topic of this blog post.
ONC has managed to work miracles (when compared to the normal course of government work) to meet Congressionally mandated deadlines for other programs, but specifyin standards required a more ad hoc approach. At the same time ONC (and particularly Doug Fridsma) have been working in the background to adapt an approach to (a) that has worked in the Department of Homeland Security. This is called the National Information Exchange Model.
Sidebar: to avoid adding wind to any Web-based firestorm here, I would like to say that there is no reason to believe that the program will be run by DHS, will force DHS developed standards on healthcare or well give DHS any legal or pragmatic access to an individual’s healthcare data other than what it already has. ONC is simply adopting and adaption parts of the DHS standard-developing methodology. Ideally the methodology comes with standards-developing tools to implement it.
The final contract for NIEM in healthcare was completed last Saturday. Doug gave a summary presentation at the HIT Standards Commitee meeting that included this slide:
These preparations have proceeded without the transparency that we have become accustomed to from ONC. As we begin to learn about the program I will be looking for the answers to these issues.
- To what extent has the NIEM methodology and tool set had to adapt to the ginormous complexity of codes and value sets that is table stakes for the healthcare standards game?
- Can the tools and methodology adapt to XML representations that were not created specifically for the purpose of representing standards under the toolset?
- I can understand why ONC did not want to give a single contractor exclusive knowledge of and influence over the development of these standards. I can also understand why ONC did not want to use potentially vulnerable internal positions to fund the work. But I still have to wonder if using five contractors is a manegable proposition.
- The program calls for trial implementations. This is good and similar to parts of the unusual approach being used in the NHIN direct program. However, what we really need is an active feedback loop that extends through early real-life implementations. It is hard to see how that level of feedback can be achieved in this approach … but I have been surprised before.
As of now this particular work is being met with a good deal of skepticism, including my own.
That being said, I have been impressed with many examples of innovation that have come from ONC.
In a time when red meat passes for a formal gown and “caterwaulablogging” is as routine as waving howdy, I pledge to express my concerns openly and with as little verbal plumage as I can muster, in hopes that I am contributing to the evolution of these ideas towards the vision I set out two years ago (and the evolution of my own views).