Monday, October 24, 2011

Bad Reporting on CDA Consolidation Raises my Blood Pressure

If you haven't guessed from the title, this is a rant.  There are two pieces I want to rant about:

John Halamka's post from today is the first, and specifically, it's this quote from Wes:
Many feel that the consolidated CDA alone will prevent as much as 50% of the programming errors found in C32 testing and that disagreements on the interpretations of the specifications will be far more easily resolved.
OK, who are these "many" that were surveyed, and how come I didn't hear about it?  If there were many who were asked, I'm certain I would have at least heard about it.  And where did this 50% figure come from?  As best I can tell, this is just an opinion with an important sounding number attached to it.  There are also several assumptions in this statement, one of which is that people would be starting the implementation from scratch.  I just don't see that happening.  And the ink on the specification isn't even dry yet, so we have precious little implementation experience as far as I can tell to even begin making these assertions with meaningless numbers.   And we still don't know if the data and models will be imported into a system that automatically generates not just the specification, but also the implementation and validation code.    That's why I signed up for the CDA Consolidation project, and I'm disappointed that I still haven't seen the promise of it fulfilled.

OK, end of that rant, on to the next one

The next one is this report from Modern Healthcare.  It's rather hard to tell what is being reported on here, because the author of the report doesn't even bother to name the specification that he is talking about. It starts out this way:
A federal advisory group is expected to finalize a short-term implementation standard for transitions of care by year-end, according to officials from the Office of the National Coordinator for Health Information Technology.
This is what I know:
  1. The HITSC Federal Advisory Committee is considering using the HL7 CDA Consolidation guide, and will finalize their recommendations soon.
  2. HL7 is still finalizing the guide, I've heard a mid-November publication date.  It will be published as a Draft Standard for Trial Use over a two year time period if HL7 follows its usual procedures.
Then it starts to get a little bit worse:
The coming standard is meant for interim use, while a longer term and more complex standard is finalized over the next two years.
A DSTU in HL7 is usually meant to be implemented over a two-year time period.  After that, implementation experience is supposed to go into the finalization process (usually after the two-year period).  There is NO intention of creating anything "more complex".  Simplifying, yes, and that will probably longer term (i.e., green CDA).

Now this is supposed to be a quote on the aforementioned interim standard:
It's based on a standard that is not actually that well put together from the perspective of interoperability, but since everyone has some version of it, we decided 'Let's let them at least converge to one version of a not-so-great standard, so they can meet objectives today; knowing that the much better, much more robust standard is going to be the standard they will all move to in the next couple years' 
This doesn't even begin to compute.  Either someone made a really stupid statement, or someone didn't understand what was being said.  The comments in context don't make a lot of sense given ONC's support for the CDA Consolidation project.  They appear to be a rather backhanded almost-a-compliment-but-really-a-slap upside the head.  I've been given to understand that this statement was made in relationship to OTHER work simplifying C32 at HIE's in New York, and that the "more robust standard" is the CDA Consolidation guide.  It appears the reporter got it wrong.  I wasn't there for the discussion, but what I'm hearing certainly doesn't agree with what is being reported.  Either way, it was enough to raise my BP at least 20 points.

In any case, if you want the real story, I'm told that this article from Government HIT is a much better report. In that, it discusses the CCD and the CCR and how CDA Consolidation puts them both together and that there will be more work towards green CDA.  The statement about how CDA Consolidation is a harmonization of CCD and CCR is just a little bit of revisionist history.  The CCD was supposed to be the grand harmonization of CDA and CCR, but I'll leave that for another day.

1 comment:

  1. Hi Keith, I read the same Modern Healthcare article and was also puzzled by the wording. I interpreted it as you did, to PROBABLY refer to the HIE C32 efforts in New York (and several other states), since Jitin did mention ONC collaborating with that group. But that group never characterized it as a "standard" since the group is not an SDO, and I'm not sure what an "interim standard" would be coming from ONC, because there's Stage 1 MU and there will be Stage 2, nothing regulatory in between. So it would be at best a voluntary implementation guide like all these standards were prior to HITECH anyway.

    The timing is interesting if Consolidated CDA starts DSTU for two years in November. The end of 2 years would be the likely start of Stage 2 MU. And I read some HL7 documentation that a DSTU period "can be determined by the host Technical Committee, but must be no longer than two years. The more common expectation is that these documents will begin to advance to normative status after 12-15 months of trial use." I didn't see any specific statements in the CDA Consolidation Guide as to the DSTU length. Has that been formally decided and documented?

    ReplyDelete