By ADRIAN GROPPER, MD

What do the President’s Council of Advisors on Science and Technology (PCAST), the recent Institute of Medicine (IOM) report and the New England Journal of Medicine article on value in health care by Porter have in common?

They call for increasing aggregation of information around individual patients even as they seek to avoid complications for and interference with providers.

Even so, the debate between provider-centered and patient-centered health information exchange is still very much with us. Recent progress with the Direct Project, Blue Button and Kaiser’s donation of terminology suggest a trend toward simplicity and open-source collaboration as essential catalysts for health information exchange.

The next logical step is a PCAST-inspired patient portal.

The essential elements of a PCAST Patient Portal are already available:

• Blue Button success proves that patients appreciate and will adopt enhanced portal features and that providers can readily connect their databases to the portal if the technical and policy requirements are truly minimal.
• The Direct Project shows that combined federal and industry support can produce accessible open source code for bidirectional health information exchange in record time.

The idea is to allow any practice or vendor to participate in automated health information exchange through the incremental enhancement of a simple patient portal. The portal would:

• Allow the patient to authenticate and access the portal (Blue Button and Direct Project both already support this capability)
• Display information available for exchange (Blue Button and Direct Project both already have this capability)
• Allow the patient to approve defaults and selectively modify metadata describing the information available for exchange. In PCAST terms, metadata must include provenance, person and privacy:
  • Provenance is simply the name of the portal operator or document author and is not subject to modification by the patient. The Direct Project address of the institution and/or author would be a simple default to start with.
  • Person might be the name and date of birth of the patient along with their Direct Project email address. Modifications to this information by the patient would be allowed subject to verification steps to prevent fraud or abuse.
  • Privacy would start with very clear defaults consistent with Blue Button and Direct Project practice such as a whitelist of specific practitioners, family supports and other Direct Project endpoints. If the portal makes an anonymous version of the information available for research they would display corresponding metadata and suggested privacy electives.
• Automatically process an incoming Direct Project message to release metadata by reply Direct Project message.
• Automatically process an incoming Direct Project message to release information by reply Direct Project message.

The proposed interpretation of metadata in terms already defined by the Direct Project and consistent with directed exchange as defined by the Privacy and Security Tiger Team make a PCAST-inspired patient portal practical in very short order. The infrastructure and some of the code for the Direct Project could be adapted to add patient portal features to the Health Information Service Provider (HISP). Over time, portals will add standardized terminology and document formats. Information that is inappropriate for the patient to view would be added to the portal as more sophisticated standards and policies for consent became accepted. Finally, the common infrastructure for locating and assembling individual elements of a patient’s records via secure “data element access services” (DEAS), as envisioned in PCAST, will evolve in parallel with the simple Direct Project approach that gets the whole process started.

ARRA Stage 1 meaningful use requirements did not directly address bidirectional health information exchange that will be required for accountable and patient-centered care. Stage 2 now has an opportunity to mandate the simplest and most essential elements of exchange that will engage consumers, providers and vendors as participants in a value-driven health care framework. The marriage of Direct Project specifics with Blue Button-inspired patient portal principles is arguably the simplest approach to a rapid consensus on bidirectional health information exchange.

Adrian Gropper, MD is a founder of MedCommons and driven by the vision of doctors and patients collaborating around shared health records on the Web.