Is a Universal Exchange Language in Our Future? | [node:field-byline] | Healthcare Blogs Skip to content Skip to navigation

Is a Universal Exchange Language in Our Future?

January 25, 2011
by David Raths
| Reprints
This controversial PCAST suggestion raises the hackles of hospital organizations and software vendors alike

The report on health IT released in December by the President’s Council of Advisors on Science and Technology (PCAST) has drawn a fairly cool response from the industry establishment. One suggestion, in particular, has raised the hackles of hospital organizations and software vendors alike: Universal exchange using metadata-tagged data elements. What does that mean, you ask? PCAST basically proposes scrapping a lot of the work the industry has already done on interoperability in favor of an entirely new approach: breaking medical records into data elements. Each element would have attached information describing it and allowing it to be searched the way a search engine searches web site data.

The PCAST report states: “We believe that the natural syntax for such a universal exchange language will be some kind of extensible markup language (an XML variant, for example) capable of exchanging data from an unspecified number of (not necessarily harmonized) semantic realms. Such languages are structured as individual data elements, together with metadata that provide an annotation for each data element.”

As Vince Kuraitis principal and founder of Better Health Technologies, LLC, notes on his blog, that aspect of the report has become something of a “political piñata” with groups ranging from HIMSS to American Hospital Association to the Radiological Society of North America telling the Office of the National Coordinator that the radical suggestion is an attack on existing health IT stakeholders. HIMSS’ comment to the Office of the National Coordinator said, “We believe that the PCAST approach could lead to substantial and negative disruptions that will impose clinical and financial costs that are not offset by reasonably foreseeable benefits.”

I saw this dynamic at work at last week’s eHealth Initiative conference in Washington, D.C. A few people mentioned the PCAST metadata approach in a positive light; others asked questions about how seriously it would be taken, while others dismissed it.

Speaking from the payer perspective, Charles Kennedy, M.D., vice president for health information technology for WellPoint, said that attempts to get value from advances from health IT in the utilization management process always run into problems with the heterogeneity of systems. “This is what the PCAST report stressed,” he said. As an industry we are challenged by a lack of semantic interoperability. Until we get to that point, we will be stuck in the equivalent of the early generation of PCs before the Internet really took off with e-commerce.

During a panel discussion at the eHealth Initiative conference, Charles Jarvis, vice president of healthcare services and government relations, for NextGen Healthcare, was asked about the PCAST approach and said it was an interesting idea but “not the right fit for this market.” Nevertheless, David Blumenthal, M.D., the national coordinator for health information technology, made sure to mention in his talk to the group that ONC is bringing in software experts to study the implications of the PCAST suggestion.

I’d be interested to hear what Healthcare Informatics readers think of the PCAST report. Please leave a comment below to share your thoughts.




Hi Joe,
I appreciate your comments. I agree with you that the Universal Exchange Language is an interesting concept, but one that does have the potential to disrupt the meaningful use momentum because it seems to lurch in a totally different direction. One analyst told me that the feds have taken this approach in other areas, such as with SEC documents, and she seemed to think ONC will take the PCAST report quite seriously. I have also read that at least one vendor is moving in this direction. Cerner is working on a feature called "Chart Search," which it says enables clinicians to perform structured, medical concept searches of unstructured patient charts. According to Cerner's web site, "clinicians can search using either industry standard terms or free text across all documents and discrete results such as labs and vital signs in an individual patient's chart. Chart Search improves productivity by limiting the amount of time needed to search a patient's medical record. Clinicians can semantically and securely search a patient's EMR by author, location, medication, document type or name, which reduces errors and improves patient safety."

David, Thanks for bringing the PCAST discussion to these blogs, and for soliciting comments. Charles Jarvis also noted that he wears multiple hats, including serving as Vice Chair of the EHRA.

The EHRA's comments are here.

A broader set of comments, including the EHRA link is here:

I contributed to and have read the EHRA comments and encourage interested readers to go there for an excellent breakdown of the issues and relevant historical objective facts.

My perspective, in part is captured by my public question to Janet Corrigan, President and CEO, National Quality Forum at this same eHI meeting. The interoperability standards focus needs to follow quality, cost and satisfaction goals. To this end, standards for semantic interoperability should follow objective, validated and shared quality measures.

The example I asked about were quality measures for the in-patient medication reconciliation process, which closely follow the Stage 1 requirements of up-to-date problem, medication, and allergy lists, as well as demonstrating some integration with ordering (CPOE IP, eRx OP). Dr Corrigan validated that these are exactly kinds of measures (with requisite standardization) that NQF is working on today.  It's unclear to me how a universal exchange language and new meta-data would accelerate that goal.

Do we need a Universal Exchange Language in place by Stage 2 or shortly thereafter to improve care in the near term? No. Could that evolution dramatically improve our capabilities in the future? Yes. Of course.  (For more on my take on what we need to focus on to improve care, in addition to a powerful quality framework, see my recent post here.)

But, Could a loss of focus on the current MU beach head with clear and proven standards put attainment of widely adopted improved, valid quality improvement at risk by defocusing our current, complex and consumptive efforts? Based on the comments in the links provided, a lot of people have that concern. You captured that very well in your post.  It's all about focus.  Thanks again.