Report on the ONC 2017 Technical Interoperability Forum

Noam H. Arzt, Ph.D.Last week I attended with my colleague Mike Berry the ONC 2017 Technical Interoperability Forum. This meeting was convened under the 21st Century Cures Act passed by Congress in late 2016. Several hundred participants attended a series of panel presentations and discussions. The forum took place over one-and-a-half-days. The forum covered a variety of topics related to interoperability, including discussion of the business case for interoperability, semantics, national networks, and application programming interfaces (APIs). In many ways the speakers were “the usual suspects” involved in national networks, standards development, and HIE planning and implementation.

Nearly two years ago I wrote an essay, The Interoperability of Things, based on the collection of comments received by ONC on the draft Nationwide Interoperability Roadmap. In a previous meeting I asked the new National Coordinator for Health Information Technology, Dr. Don Rucker,  if the Roadmap was still relevant and he said it was. During the Forum, however, there was absolutely no mention of this document. In fact, it seemed that the Roadmap was the operative guide for ONC activities or thinking. My own essay drew out a number of themes in interoperability I perceived at the time, including:

  • Lack of consensus on definition and scope.
  • Ambiguity over the role of HIEs, especially at the state level.
  • Disagreement over whether the pace of change was too fast or too slow, too general or too specific.
  • The complex state of consent and privacy laws across the country that really put a crimp in cross-state data sharing.

I found the 2017 Forum to be no more conclusive. Here are some of the themes participants were talking about the most:

  • The world is on FHIR. Or so it seems. While everyone was talking about this emerging standard, and the SMART application environment (the growth chart app seems to be the most prominent example, mentioned several times during the Forum), its use in the field is still somewhat limited. And it represents the tension between those who believe we need to allow standards to be used and proven in the field at a pace consistent with the ability of organizations to implement them and those who see the improvement in standards as reason enough to push ahead with new versions.
  • Semantics. I concluded The Interoperability of Things by positing that if we did nothing else but work for several years on improving coding and shared meaning in source systems like EHRs we would be much closer to where we need to be by the time technical standards and implementation caught up. Without shared meaning interoperability has little chance of positively affecting patient care or anything else (and has some risk of negative effect).
  • Measurement. I have always believed that we need to measure the impact of interoperability, not the process of conducting interoperability transactions. There was a lot of discussion about this at the Forum, with some sentiment that you can’t improve what you can’t measure. I’m still not sure we know what to measure or how.
  • Patient-centered interoperability. Folks continue to beat the drum about patient-centered control of data and interoperability, but we still live in the US in a provider-centered world. It does not seem like much has changed in ten years, despite those who promote Direct as the answer to engaging patients.
  • Participation in Standards Development. There was some concern that enough organizations and individuals were not participating in the standards development process only to complain about the results later. I’m not sure this is fair criticism, since participation in standards development is very costly in terms of time, money, and talent. It is perfectly appropriate for users to raise concerns about the results of standards development whether they participated or not.
  • Public Health. Once again, public health barely got a mention at this meeting. Public health interoperability is wide-spread and measureable, even innovative (like our Service-oriented ICE Open Source Forecaster).

As I concluded in 2015, governance is still a major issue. We still can’t agree on the problem, let alone the ways to solve it. We have no real way to develop consensus across such a complex healthcare ecosystem, especially when it is very hard to overcome self-interest and even conflicts of interest in the process. We tend to ignore what is going on outside of the US, and have not yet found the “sweet spot” between what we have already implemented and what is still on the horizon. And finally, this is more about the pace of change than change itself.

But there is reason for some optimism. The CMS EHR Incentive Programs have certainly spurred the adoption of EHR systems which is a major enabler of interoperability. And more and more patients are gaining access to their own records through patient portals and, slowly but surely, various types of Personal health Record (PHR) applications. I hope at least that this month’s Forum will engage the community more broadly and encourage everyone to keep plugging away at developing a more interoperable healthcare system.