6) Maximize network effects - Does anyone remember how important the network was to making PCs useful? No one may want to fund separate Health Information Exchange organizations, but there is much that can be done to advance networking anyway. There must be interoperable directories, functional indexing, security infrastructure and application programming interfaces / standardized transactions beyond “push” email to support HIT nationwide. Build a named network that providers want to join because it provides tangible value for them to do so. If we are to get past data hoarding and business proprietary interests, we need to enable, not frustrate, network effects. Providers will want to join a network, or a network of networks, if they can go to it to pull down the information they really need.

5) Build up infrastructure - We need to build some of this HIT infrastructure “up,” architect it, instead of expecting that somehow business requirements will suddenly align to make a coherent architecture appear below them. If one only follows outcomes or business requirements “down” to technology implementation, inefficient silos of activities are developed instead of shared infrastructure. Are the technical needs of population health management really that different from registries, public surveillance systems, and quality reporting systems? By “architecting up” we can more readily build components to serve multiple functions.

4) Organize for technology leadership too - Rearranging the HIT Policy and HIT Standards committees was overdue. But it is also critical to address the relationship between them. As was stated at the last HIT Standards Committee meeting, the perception is that “the Policy Committee has high-level policy thinkers and the Standard Committee has implementers.” As per the JASON and PCAST reports, we need high-level architecture thinkers who can help design “up” the strategic interfaces and components to meet multiple business needs and make for a coherent complex system. Where is the strategic technology discussion?

3) Emphasize managed data - Providers want value in the information that they get. The paper-based medical record has frequently been a detailed recording of care. EHRs can compound this sometimes overwhelming information accumulation or they can help provide up-to-date, managed information like current and well-maintained problem, medication, care plan, allergy, and immunization lists. Current, succinct information needed for care is valuable to providers.  A record of care is also important, which is why providers will always “own” certain aspects of the record need, but it is not the only one.

2) Make quality reporting help HIT - HIT has been pitched as critical for quality reporting, but why has quality reporting taken such a heavy portion of the limited HTECH leverage? Quality reporting can always be incented through differential reimbursement. With the HITECH leverage evaporating, let’s think about how quality reporting can help the more general HIT agenda.  Instead of having a quality reporting “silo,” have quality reporting standards serve multiple purposes including supporting health information exchange, continuity of care and population health needs.

And the number one recommendation for rebooting HITECH and advancing health IT nationally is…..

1) Stop talking only about EHRs! - EHRs are part of a much bigger HIT ecosystem. They are like the leaves on a tree. There must also be branches, a trunk and roots. There are networks and hundreds of other HIT systems that support ancillary organizations and activities, population health and healthcare.
 
Progress has been made. We may be now entering the post-HITECH period, but it is not the time to regress. Clear leadership and resolve can build on the HITECH investment and put broader HIT outcomes into reach.

John Loonsk, MD, served as director of interoperability and standards in the Office of the National Coordinator for Health Information Technology from January 2006 to December 2009.