CCR compatable by 1/1/2006

Hey... I thought one of the great things about the CCR was supposed to be that it was "vendor independent!?" It sounds like from this thread that it is not. I suppose you could say that the actual CCR file itself will be "vendor independent" (which is an XML file) but the production of the CCR file is very much vendor dependent. And even a standard XML file will have to be parsed in different ways to be incorporated into a specific vendors EMR. So the use of the CCR IS vendor dependent. It seems to me there is already a standard in place that can/could produce a CCR file (which is a good idea) - HL7.

Jeff

CCR is a great idea and will hopefully move us all in the direction of full interoperability, make selection of EMR's more straighforward and ultimately make physicians lives simpler, communication among physicians more streamlined, reduce test ordering redundancy, and help improve quality of care. I've not reviewed the elements in detail, does CCR include a list of procedures- e.g. echo, cath, colonoscopy- done in the past? This info is invaluable, especially when seeing a patient for the first time, hearing a heart murmur and not knowing for sure whether/when an echocardiogram (very expensive test) was performed in the past. Are there plans to make a centralized CCR available over internet for patients traveling out of town seen by unfamiliar physicians?

David

David P. Hurwitz, M.D.
President, Dynamic Digital Healthcare Systems, Inc.
Indialantic, FL
david.hurwitz@ddhsinc.com

jeffxxx,

The easy answer is "garbage in, garbage out". That is, if an EMR collects a poor quality of data, then the CCR will reflect that, but so too would HL7.

Writing a robust interface for HL7 is far more complicated. Exceptions to the rule are the problem. While hospital A may use the NTE segment to carry any notes regarding a part of the labwork, hospital B does not. Since NTE segment can be placed anywhere in HL7, the complications of writing an HL7 interface build quickly. So writing a single interface that will work with all lab vendors becomes difficult. And once you get close, you must also account for all reference labs and hospitals having their own PROPRIETARY code for each lab test ("own and proprietary" are redundant, for emphasis) . How many map to the LOINC standard?

This is but one small example. If HL7 was so easily used, so robust in design that any coder could write an elegant interface on a long weekend, it would have happened. The truth is that the huge hospital vendors can unload the schoolbus of MBAs at the front door of the hospital, once they receive their multimillion $ check, and unleash their talents on creating all those special HL7 interfaces to multiple systems. It is part of their business model.

The average physician practice cannot afford the "unload the bus" mentality, nor should they expect it. An easier mechanism is badly needed. The CCR is a step in the right direction. Initially the amount of data will be contrained by the limits of time in developing a standard, but as its popularity goes, so will its usefulness.

Are you one of those MBAs by chance? You could enumerate the problems.

Matt Chase
www.medtuity.com

In the NTE example cited, critiquing HL7's "floating standard," could not specific new segments be created that were more rigid/solid, with much more specifically defined fields?

I am a fan of CCR, but, lacking some of the technical perspective in this arena, wonder why not use an existing agency and standard that has waded in these waters longer?

JCP,

The problem is that the standard is not going to change to accommodate us-- it'll break too many programs.

For your problem (in another thread) of creating an interface between your local lab provider (Orchard) and Centricity, it's not such a problem. You just handle the local quirks. For example, maybe Orchard uses NTE.03 for placement of microbiolgy susceptibilities, or maybe they use OBX.04. Maybe they don't even use the NTE segment at all. Still, this is not so bad because you are dealing with a single provider.

But the NTE is also used for many other things by different vendors. It begins to look like a shelf in my garage, cluttered with anything and anything that had no other place for storage. It does not look like my sock drawer. For me, all I have to do is look in my sock drawer for socks. If there aren't any, then they are all in the laundry. I never have to go to the garage. As you add more vendors, you'll have to make lots of trips to the garage, just in case.

My analogy can be easily broken, but I think you are discovering now with your foray into HL7 there are plenty of gotchas. Also, you are dealing with just one part (lab); now add every part of the HL7 standard and start adding multiple (say, many) vendors.

If every physician could spend what hospitals spend for HL7 interfaces, no problem. But that is why you are building your own-- we all cannot afford it. IMHO,

Matt Chase
www.medtuity.com

Matt,

What I was asking was, why not create a new segment for CCR in HL7? Instead of NTE, OBX, ORU, PID, call it "XXX" or something. In other words, upgrade HL7 to a include "new material." Or am I really missing something big here?

>>> What I was asking was, why not create a new segment for CCR in HL7? Instead of NTE, OBX, ORU, PID, call it "XXX" or something. In other words, upgrade HL7 to a include "new material." Or am I really missing something big here?

JCP: The take I have on CCR, InfoPath, and other initiatives is that we have what seems to be a neat language---> XML. Microsoft has hijacked it and put it into most of its offerings. Now, folks are trying to use it in one way or another, hence we have the "CCR" initiative. HL7 is a decent path to follow, but, well, I'll let Microsoft explain why it wants you to purchase its products and go with CCR:

URL- http://www.microsoft.com/resources/casestudies/CaseStudy.asp?CaseStudyID=14931

Overall, though, the concept of the "CCR" is good... Although I think that Microsoft has more to gain than we do!

Regards,
AL

Al Borges, M.D.

• Internist/Oncologist in a Small Group Practice in Virginia
• Columnist, MDNG magazine (“HIT Realist”)
• My website URL: http://msofficeemrproject.com/

• | Post Points: 5