VetXML
Dennis Ballance
Recently, at the North American Veterinary Conference in Orlando,
Florida, a group of people representing Practice Information
Management System (PIMS) vendors, pet insurance companies, microchip
companies, corporate practices, and AAHA met with representatives of
VetXML, a UK consortium that has defined a set of XML schemas for
sharing certain types of data between PIMS and service providers,
including pet insurance companies and microchip registries.
VetXML is designed to be simple to use, and it includes a transport
layer definition (in fact, the detailed specification was donated to
VetXML by its designer, VetEnvoy, the sole implementation of VetXML
middleware at this time). It also includes a growing set of defined
terms, including a diagnostic code list that maps to SNOMED, and a
breed/species list that is not mapped to SNOMED. It appears that the
current schema structure is amenable to being adapted to HL7
structures and concepts, with some work. The schemas and transport
layer are open, meaning they are freely available without restriction
to anyone wishing to use them. To have input on the design of these
requires consortium membership.
The offer from VetXML is to extend the consortium into the US,
allowing US companies to join the consortium and contribute to its
further development. There will be a followup meeting on Monday, Feb
15 at Western Veterinary Conference (time TBA) to discuss the offer.
In the meantime, this forum is designed to facilitate discussion among
all stakeholders in veterinary standards, including any companies
selling products in the IT space, academic staff and faculty,
governmental bodies and agencies, professional organizations, and
practitioners or other consumers of IT products and services.
The most significant questions that have been posed to me (from
various sources) are:
1 - Does the fact that VetXML is proven in the market and continues to
grow make it valuable to the US market?
2 - Do HL7 and other human-centric standards have the capability to do
all that VetXML does?
3 - Are HL7/IHE/SNOMED too expensive for veterinary companies to
implement (meaning, require more expertise than companies in our
market have to spare)?
4 - Does VetXML have the right organizational structure to become an
international standard body?
5 - What is the relationship between VetEnvoy and VetXML, and between
SPVS and the two?
I'll post my thoughts on these questions separately.
Dennis
Robert Featherston
Reference the question: "Does VetXML have the right organizational
structure to become an
international standard body?
Is there a reference to a description of VetXML terminology structure,
definitions and maintenance?
Mike Fletcher
While the Vet-XML Consortium has developed a series of proposed data
schemas that can be viewed openly online (at www.vetxml.org), it has
yet to officially formalise and document its operating procedures and
governance that it's been working under informally since its inception
(i.e. late 2005). This process of documentation is currenly underway
and the resulting document will be made available through the website
once completed.
Mike
On Jan 25, 4:15 pm, Robert Featherston <rhfeathers...@gmail.com>
wrote:
> > Dennis- Hide quoted text -
>
> - Show quoted text -
Matt Wright DVM DACVR www.animalinsides.com
the direction of vetXML, outside contributions, who can sit on
committees, who gets to vote etc?
Matt Wright DVM DACVR www.animalinsides.com
is adopted for the veterinary profession, all we ask is that ONE AND
ONLY ONE standard is used and accepted. Although I was not at the NAVC
meeting, I am told that there was alot of grumbling as the VETXML
folks did not make the case that that this is an OPEN standard in the
sense that it is open to discussion and modification from all
interested parties. It is clear that this is OPEN for anyone to use
but unlike other standards organizations it is not clear that the
future development is open to input from others.
Failure to demonstrate that the organization is configured to run like
an objective and independent standards organization will result in
another failure for veterinary standards because of the inevitable
political fighting between the companies with a vested interest in
this movement.
Matt Wright DVM DACVR
www.animalinsides.com
Mike Fletcher
standards only those who are members can contribute to its
developement. However, everything is OPEN to debate, so if there is a
groundswell of support for making it truely open for all regardless of
membership then I'm sure this is how the Consortium can be structured
and governed going forward. One to discuss at the next meeting at WVC.
Hopefully see you there.
On Jan 25, 8:10 pm, "Matt Wright DVM DACVR www.animalinsides.com"
Dennis Ballance
working group. Observers can be on the mailing list, can contribute ideas
and engage in discussion. However, only "voting members" (I'm forgetting
the right term) who have been nominated and approved by the Standards
Committee and turned in a patent waiver can vote on the action items created
by that working group. To have a vote on the standards committee itself, I
believe academics can do so for free, but industry members and professional
associations must pay an annual membership fee (or else supply a secretary
for one of the working groups).
I do not know how HL7, SNOMED, and LOINC are structured as far as who can
participate and what fees or costs are affiliated with that participation.
Can anyone else fill in this detail?
Dennis
On 1/25/10 12:10 PM, "Matt Wright DVM DACVR www.animalinsides.com"
Michael K. Martin
organizational. Membership gives both voting rights and access to all
the standards. Organizational memberships are based on size and/or
revenue and vary the number of voting representatives. Cost can be an
issue but is unavoidable (I'm told) because the HL7 process is very
resource intensive. This can be managed by careful use of
organizational memberships, etc.
LOINC uses a different model. It is managed by Regenstrief Institute on
a very open basis. Anyone can participate, request codes, etc.
Attendance at committee meetings is open and free; just pay your way
there. Voting membership on the committee is open to anyone who
represents a constituency not already represented. Most reasonably well
researched requests for codes are handled at the staff level and never
require an official ballot process like is followed in HL7 (and most
other SDOs.)
SNOMED has recently been transferred from the College of American
Pathologists to the unpronounceable acronym IHTSDO (International health
terminology standards development organization). Membership is on a
national basis for the founding member countries with affiliate
memberships for other countries and organizations. Details of how input
on content gets coordinated are still being worked out. In veterinary
medicine we have an extension namespace for the National Animal Health
Laboratory Network managed by Virginia Tech, that gives us a fast-track
for terminology we need. The National Library of Medicine is the US
representative.
ps. Each of these organizations is based on consensus rather than
majority-rule, so "vote" is a little misleading. It is not the case
that an organization with more voting participants can get its way and a
smaller one cannot. It all comes down to gaining consensus from the
group as a whole.
Mike
> .
>
>
--
Michael K. Martin, DVM, MPH, DACVPM
Clemson Livestock Poultry Health
PO Box 102406
Columbia, SC 29224-2406
email: mma...@clemson.edu
personal email: michael.mar...@gmail.com
phone: (803) 788-2260 ext 230
work cell: (803)312-1439 (no personal calls)
personal cell: (803)348-1879 (no work calls)
fax: (803) 736-0885
There are no passengers on spaceship earth.
We are all crew.
MARSHALL MCLUHAN
Matt Wright
Can you describe the uses and differences between these organizations. My understanding was that HL7 was interested in information transfer while LOINC and SNOMED were about codes. They go hand in hand but, at least for now, isnt HL7/IHE the main organization that we should be interested in?
Thanks.
Michael K. Martin
The slightly over simplified version is this: HL7 defines the syntax of messages, documents, and services. LOINC provides codes for the names of observations (the "questions"), and SNOMED provides a general medical terminology (the "answers"). HL7 attempts to provide unambiguous context for the information while LOINC and SNOMED provide unambiguous semantic meaning to the individual concepts.
I say over simplified because each standard takes on a little of the other concepts. For example, HL7, especially version 2, defines a lot of code tables that more correctly fit under SNOMED and are moving that way in newer versions.
Our experience in the NAHLN is that the central focus is on HL7 but that much of the detail work has to take place in LOINC and SNOMED. You get the basic messages defined and then the things that change are LOINC codes for new tests, SNOMED codes for new specimen types, animal species, diseases, etc.
Mike
Matt Wright DVM DACVR www.animalinsides.com
would it be incorrect to say...
If we were building a house the IHE acts as the contractor saying this
is what the house is going to look like. HL7 is like a hammer and the
skills necessary to use the hammer. LOINC and SNOMED would act as the
nails in this analogy.
The bottom line with all of this is that these organizations all have
a function, are all necessary, and all work together to get us where
we need to go.
Matt
Michael K. Martin
as the carpentry subcontractor and LOINC and SNOMED as materials
suppliers. As prospective homeowners we are interested in both the
shape of the rooms and things like the carpet and plumbing fixtures. A
good contractor will help us make sure they all work together well.
That is the role IHE plays and the one that Dennis has said we need some
organization to play for veterinary medicine.
DICOM is a little harder to fit into the analogy because it deals with
both syntax and semantics. Maybe they are the landscaping contractor.
Or as a radiologist, you might want to call them the foundation contractor.
Mike
Matt Wright DVM DACVR www.animalinsides.com
THEY want their role to be. I am not clear if they even want or strive
to be the general contractor. Is their goal to be a subcontractor
(building a house there are many subcontractors working side by side)
or is their goal to be the general contractor?
If they do not want to be the general contractor our challenge then is
find or start a group to work as a general contractor. Previously,
the human side IHE has offered their assistance in this regard. Should
we start a veterinary "branch" of the IHE in the same manner as we
started a DICOM working group. That method was very successful with
regard to DICOM. If this is the road map that we take, it seems like
it would be reasonable and effective to look at what vetXML has done -
keep what is useful - and add on other schemas that have already been
sorted out on the human side.
I guess the question that needs to be asked is whether or not vetXML
even wants the role of general contractor.
VetXML - comments?
Robert Featherston
Thank you for your response.
I apologize for not being more specific in the previous query.
Actually, my question was intended to address the nomenclature or
common terminology used by the various users in data input. For
example do the insurance companies provide a list of terms and
definitions for the submission of data?
Thank you!
Robert
On Jan 25, 1:45 pm, Mike Fletcher <mikef...@gmail.com> wrote:
> Good question!
>
> While the Vet-XML Consortium has developed a series of proposed data
> schemas that can be viewed openly online (atwww.vetxml.org), it has
> > - Show quoted text -- Hide quoted text -