Re: [COAS-List] qualifiers from HL7/CEN

["David W. Forslund" <dwf@lanl.gov>]

At 11:17 AM 4/6/99 , Larry Hamel wrote:
>Dave,
>
>At 12:20 PM 4/5/99 -0600, David Forslund wrote:
>> > section 4.2.6 provides a list of possible qualifiers, but we specifically
>> > decided not to provide IDL definitions for these (reversing a previous
>> > decision in previous COAS drafts).  many of the items in 4.2.6 are
>provided
>> > by HL7 in the examples in section 12.5, and others are probably covered by
>> > reusing HL7 definitions.
>> >
>>
>>The problem in my mind is whether there we will achieve sufficient
>>interoperability between servers without getting a larger list from the
>>HL7 or CEN definitions into the document. If none are required to be
>>supported, it isn't clear there will be any interoperability between
>>COAS servers.  Also, is the data defined by these qualifiers sufficient
>>unambiguous from the HL7 or CEN spec?  Do I assume that the values from
>>some of these qualifiers are then in HL7 segments which must be parsed
>>with an HL7 parser, or are they further broken down into atomic HL7
>>pieces? 
>>
>>Does this question make sense to you?
>>
>
>
>yes, this question makes sense.  
>
>note that with the naming convention in place, all definitions of HL7 and
>CEN are included in the document, in one manner of thinking.  are those
>specifications sufficiently unambiguous?  probably not. they are all we
>have right now, so i suppose that putting energy into tightening up those
>specs (like the HL7 Templates effort) is the best path right now for
>interoperability.
>
>note that a big HL7 chunk can be broken down with the existing naming
>convention, e.g.
>
>const QualifiedCodeStr SpecimenSource = 
>	"DNS:omg.org/DSObservationAccess/HL72.3/OBR/SpecimenSource";
>
>refers to a big chunk of the OBR, but a finer grain code like 
>
>const QualifiedCodeStr SpecimenSourceBodySite = 
>	"DNS:omg.org/DSObservationAccess/HL72.3/OBR/SpecimenSource/BodySite";
>
>would specify a smaller piece.  

A brief statement along the linese of your example sited here would help
clarify how the 
HL7 segments are to be used, without having to explicitly put them into the
document.

Thanks,

Dave

>
>i think we need to get some implementation experience before we add to the
>specification.  right now, we've said that observation data definition
>(qualifiers are observations) is out of our scope.
>
>larry
>