The implementation success has been somewhat patchy, with at least some proponents scaling back their enthusiasm for full adoption of the V3 Standard as some see it is lacking the necessary robustness and internal consistency for ongoing use.
The following blog from Professor Barry Smith had the effect of reminding me there are a variety of views on the topic - some of which may not be all that heartily embraced by HL7 and NEHTA. If the dissenters are right the implications are pretty serious.
The views expressed here are quite provocative but I believe need to be aired.
Monday, February 28, 2011
Cries for Help
One of the disturbing aspects of the HL7 phenomenon is the degree to which so many of those who have strong critical views are reluctant to express these criticisms in public. Here a sample criticism of the HL7 RIM, by a leading healthcare IT specialist in Asia who has given me permission to quote from his email communications, as follows:
It is abundantly clear that V3 is an unnecessarily complex, incoherent and confusing messaging standard while V2 is simple, workable, elegant and deployed at more than 95% of healthcare institutions wherever HL7 is used.
Why not scrap V3 and simply work on improving the V2.x standard (say V2.7)?
.....
I am concerned about the growth and proliferation of a meaningless standard because it is fraught with danger and is leading to:
(a) Increase in the cost of Hospital IT implementation
(b) Making coding complex which could lead to errors and thereby lower care of (and even endanger) patients
(c) Increasing the cost of software
(d) Increasing the cost of training and implementation
(e) Forcing the use of two standards in parallel when one could have sufficed
(f) Diverts funds from useful IT apps to feed V3
The only people/groups that could possibly gain from this are:
(a) HL7 org and their collective egos
(b) Trainers who will make money from teaching this "complex" 'new' standard
(c) IT companies that make money from pushing newer versions and widgets to "upgrade" to V3
.....
Here is another example of a desperate push through a back door: Pushing and converting a perfectly good CCR to a CCD by unnecessarily insisting on a RIM based approach (where none was required) shows desperation to push the RIM, come what may.
It is my personal opinion that there should be no shame in accepting that a certain approach has failed. We are doctors and scientists and accept, humbly, that we do not always succeed. The real shame and loss would come from stubbornly trying to flog a dead horse putting time, money and people at stake, just so as not to declare failure. I say, be brave enough to move on and use the excellent collection of thinkers at HL7 to develop a fresh, focused and simple solution. To try and capture the complete medical domain (present and future) into one core model is not only foolish but also unachievable. We as doctors know that and therefore only focus on our core specialty.
The full blog and many others are found here:
http://hl7-watch.blogspot.com/2011/02/cries-for-help.html
It is definitely worth reading this blog where some Australian contributors express their concerns.
http://hl7-watch.blogspot.com/2010/11/are-iso-21090-data-types-too-complex.html
While I am not sufficiently across the details of some of this to be able to form a trustworthy opinion a number of very smart people I have chatted with have expressed similar concerns.
I highlight the material for people to review and consider on that basis. Form your own conclusions, from a position of being aware that there are a range of views out there about the whole HL7 V3 project and just where it is leading!
David.
15 comments:
Absolutely 100 percent correct "To try and capture the complete medical domain (present and future) into one core model is not only foolish but also unachievable. We as doctors know that."
Anyone who has never studied and practiced medicine across many of its domains will never ever comprehend the depth and perceptiveness of the above statement. Some allied health people just might - like nurses and physiotherapists but non-medically trained individuals who live eat and breath IT and artificial intelligence and IBM supercomputer knowledge systems will never ever grasp this very simple fact.
The multiplicity of disciplines embraced by medicine combined with the diversity of the basic knowledge base upon which medicine is founded when mixed with the intuitive capacities of a well trained experienced medical practitioner is something that is extremely hard for lay people to comprehend.
That is why "To try and capture the complete medical domain (present and future) into one core model is not only foolish but also unachievable. We as doctors know that."
hi David
As co-chair of MnM and consultant to NEHTA on CDA, I guess I old the bag for v3 in Australia.
There's no reason for anyone to be afraid to comment. HL7 is not a police state, and there's an endless list of people who criticise and carp. They can expect a vigorous in kind response, but nothing more.
v3/RIM is not perfect. But it's not intended to be a ontology of everything. If only Barry would understand. It's just a model that has some use for interoperability. It's got some go-down-with-the ship type supporters, of course.
As does v2 (and all the other stadnards - they all have the folks who are going to go down with the ship). It's not enough to say it's simpler. There's more to it than that. The discussion is being had elsewhere, and here is not the place for it. HL7 continues to produce v2 (v2.7 is coming) but the community that is HL7 is switching to v3 because of it's power.
v3 itself does have a patchy record. I'm on record as saying that there won't be another full v3 implementation. People will cherry pich the parts that work - like CDA - and use them as they want
I've got more to say but it's starting to rain and I'm out in the bush camping. Have a good weekned.
Grahame Grieve.
(anonymous because nothing else works)
Thanks for the perspective Grahame.
A comment or two on how you see it finally playing out from a known expert would be good!
Cheers
David
"To try and capture the complete medical domain (present and future) into one core model is not only foolish but also unachievable."
It's certainly unachievable for many years to come because there are so many issues which need to be solved along the way. continuing to spend many hundreds of millions of dollars at the outset without investing a small percentage in one key core application is terribly shortsighted. Deloitte recommended eprescriptions as the first and highest priority application to bed down. Why is this not being done?
My understanding is that there are two private consortia doing e-prescribing and that NEHTA is working - at a glacial pace - on developing a standards package for the domain - and a few others - referrals etc.
The slow pace at which all this is moving - given the money thrown at it - is really alarming!
David.
This is a very interesting commentary and to belive we can "at this point in time" code the complete medical domain is indeed unrealistic but what tools do we have now that work? I refer readers to how the use of "existing" HL7 models can actually deliver effective clinical decision support tools. As a clinical informatician (on former HL7 committees) the technology of this is not my domain but I do know what helps me improved patient care as shown in the referred to model at http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1839638/
"NEHTA is working - on developing a standards package for the domain". That begs the question do the two private consortia you mention comply with NEHTA's standards?
"My understanding is that there are two private consortia doing e-prescribing and that NEHTA is working - at a glacial pace - on developing a standards package for the domain - and a few others - referrals etc."
Well NEHTA have produced an HL7 V3 CDA specification for electronic transfer of prescriptions (eTP) and have submitted it to Standards Australia to fast track it into an "Australian Technical Specification". I understand that this is to be used as the basis for eTP in Australia, and presumably will be adopted by the two existing private consortia and any newcomers.
However, although this CDA specification introduces a lot of new complex codes, it is actually less prescriptive than the existing HL7 v2 based AS4700.3 standard for exchanging medication information, particularly for supporting safe clinical practice. Let me give one example. I could give many others.
Quality Use of Medicine programs continually push the mantra "Right person, right drug, route dose, right route, right time". I'm not a clinician, but as a patient, I would like to know that the route and dose have been supplied on the prescription. The existing Australian Standard has a whole HL7 segment devoted to recording route of administration. NEHTA's proposed standard has NO specific field to record Route of Administration, and NO terminology to support it. Instead, it has a single free text field that conflates route with dose. Thus, no compliance checking to ensure that route has been supplied would be possible. No compliance checking to ensure that any dose information has been supplied would be possible.
I'm perturbed that from a clinical safety perspective that we might be going backwards in some areas of clinical data standards. This is not a criticism of HL7 v3 per se in this case, but yet again, illustrates the lack of proper governance in e-health in Australia.
Eric, thanks for the detailed comment.
It is my understanding that one of the present providers is committed to implement what Standards Australia comes up with and that the other has said they will look at it when the Standard is finalised.
I should say I am aware of others besides Eric with very considerable concerns about just where the HL7 vs CDA situation currently sits - many of who are on the relevant IT-14 Sub Group.
David.
Eric Browne (Saturday, March 12, 2011 3:45:00 PM) said “NEHTA's proposed standard has NO specific field to record Route of Administration, and NO terminology to support it. Instead, it has a single free text field that conflates route with dose.”
That is amazing, indeed quite incredible. It simply is not good enough. I had to check conflate – which means melds or brings together.
From quality control, accreditation, and safety perspectives NeHTA would fail dismally if, what Eric says, is true.
From a medico-legal perspective any doctor, nurse, dentist or any one else authorised to write prescriptions would be totally liable if a medication error (adverse event) occurred as a consequence of a medication being taken which did not specify precisely the ‘route’ and the ‘dose’.
Much of the rationale behind introducing electronic prescribing and transmission of scripts is to cut down on medication errors. As a doctor I can categorically state that I would never ever prescribe a medication without stating exactly the DOSE (‘n’ mgm, microgram, ml, or whatever) and the ROUTE (Intramuscular, orally, per vagina, in the eye, on the skin, on the affected area, up the nose, in the ear up the rectum, whatever).
It beggars belief to think that NeHTA could even contemplate creating standards which do not include ROUTE and DOSE. So much so that I have to think, despite Eric’s acknowledged expertise in this area, he has to be wrong in his claim that NEHTA has “NO specific field to record Route of Administration, and NO terminology to support it” and simply has a “free text field that conflates route with dose”.
I suggest someone very quickly asks a Medical Defense Organization to provide an opinion on this pronto.
And I suggest too that NeHTA be asked urgently to very clearly clarify its position on this matter.
I think HL7 V2 is a very under-appreciated standard and I would in general agree with the original comments.
V3 messaging appears dead, even from Graham says and CDA is mentioned as "working", but CDA is a Document format and not a messaging specification.
We do require a messaging specification and V2 is rich and flexible in this arena and there are a lot of existing implementations out there, although most are quite flawed however. CDA on its own just cannot do ePrescribing in anything but a token document transfer. HL7V2 medication messaging has all the behaviours needed to actually enable ePrescribing. By definition CDA has no behaviours as its a document.
The advantages of CDA are that it is xml and people can use a off the shelf parser and it allows richer structure in the body of the document.
Good HL7V2 parsers are available and V2 can be encoded in xml and the parsing side of the problem is a tiny percentage of the task.
The structuring of the clinical data in HL7V2 is an area that needs improvement, but there are several projects active working on that and in any case you could embed CDA documents in HL7V2 to get the messaging and the structured payload. Either or both approaches are appropriate, but standalone CDA will not solve our problems.
Enhancing V2 structure is a path that could be backward compatible and could be incremental but it seems that very few people have a deep understanding of both V2 and V3 and just choose the most recent technology thinking that V2 is on the way out when in fact, its expanding. Taking the lessons learn in over 12 years of V3 development and applying them to enhancing V2 is a path that is doable, but I fear that it appears to much like a failure for people to support it.
In Australia we really only have V2, so enhancing what we have with compliance testing and actually using the V2 Medication standards we have would seem a very sound choice, but Nehta appear to have a complete block when it comes to looking at working standards and enhancing them. Everything has to be done from the ground up and we end up with over complex V0.9 "standards", that would need years of trials to perfect, time they do not have.
In fact Nehta appear to show no interest in looking at what is working, its an arrogance that the country cannot afford. Nothing Nehta have produced so far is out there and working now so the track record is pretty poor and there is high level pressure on Standards Committees to approve the Nehta work, bypassing the normal processes which are there in an attempt to allow proposals to be critically examined. The proposals are all flawed and the new policy is to ignore Nehta work until its proven, which at this stage means ignoring everything they have done.
This Dinosaur has escaped Jurassic Park and its time to hit the kill switch before it eats all the people trying to do real work.
Saturday, March 12, 2011 9:03:00 PM thank you for a very lucid account of the HL7 issues and your comments on eprescribing.
Your comments serve to reconfirm the rationale and absolute imperative of funding an industry led eprescribing initiative in parallel with NeHTAs work. It is clear, from what you say, that is the only way we will see any progress in the foreseeable future in this vitally important application.
Saturday, March 12, 2011 9:03:00 PM said "there is high level pressure on Standards Committees to approve the Nehta work, bypassing the normal processes which are there in an attempt to allow proposals to be critically examined."
The Standards Committees should resist such pressures, remain true to their raison d'etre and under no circumstances bypass their normal checking processes. If they did bypass and take short cuts for political expediency or because they are being paid by NeHTA to deliver to an unrealistic timeline they would be doing the nation as a whole a gross disservice and undermining the credibility of Standards Australia.
We have standards for a very good reason. No one should be permitted to corrupt the process under any circumstances.
As the CDA Schema contains specific elements for both Route of Administration Codes and Dose Quantity, it would be very surprising if the NHETA specification (presumably an application layer Data Model) does not map directly to these elements. Whether all of the systems that will use the NEHTA specification currently store these elements in an atomised and (in the case of Routes) coded form is another question.
Hi David
You asked for my opinion on how this will play out. It's too early to say - the outcomes will be driven by external events and not by the quality of the underlying standards. (A familiar story in IT)
Some of the comments on this thread are profoundly ignorant. NEHTA has an extensive consultation process. If ETP does
conflate route and dose (which sounds like a bad idea to me), then the decision sure didn't come out of NEHTA, but from the consultation process. (The discharge summary does conflate the two, but I don't think ETP does). Nor
is the decision driven by the standards. It's based on requirements analysis.
And I hope that all the anonymous commenters on this blog are also contributing to the NEHTA CMK (http://dcm.nehta.org.au/ckm/). Of course, you have to stand behind your comments with your
name there. But the DCM models matter, particularly with regard to things like route/dose conflation.
It's true that there is tension between the consensus based standards process, and project based deadlines. That's true in all contexts, and it's a problem that IT-14 (which met this
week) is well aware of. While the standards process shouldn't be compromised, committees do need to figure out how to assist meeting politically set deadlines.
Grahame Grieve
Post a Comment