Help
Login
Search
CDA-CH - Issues

 
Select object
 
false
Results (1 / 1)
Id Issue Status Priority Type Date Assigned To Label
16 Format of the document ID Open Normal Change Request 2020-09-04 15:24:19
 
Change Request Format of the document ID
Id hl7chcda-issue-16
Status Open
Priority Normal
Last Tracking 2020-09-04 15:24:19  by  Quentin Ligier
    
Concerns
Template
false
Document Id (chdachpharm-template-header-23)
/
/
Template 2.16.756.5.30.1.1.10.2.23 (2018-04-18) Document Id
Status Active
/
-
/
-
Events
Tracking Open 2020-09-04 15:24:19 : Tracking by Quentin Ligier
Edited by Quentin Ligier ( 2020-09-04 15:26:24 )
Description
Alright, thank you!
The Schematron regex accepts both cases:  ^[A-Fa-f\d]{8}-[A-Fa-f\d]{4}-[A-Fa-f\d]{4}-[A-Fa-f\d]{4}-[A-Fa-f\d]{12}$ There also is a info assertion about that:  Lower case UUID "[...] " found in @root. UUIDs SHALL, under official HL7 V3 Datatypes Release 1 (and 2) rules, have upper case hexadecimal digits A-F. RFC 4122 and HL7 FHIR state lower case UUID display.
Tracking Open 2020-09-03 21:06:08 : Tracking by Oliver Egger
Description
yes, cda-ch limited it to a guid which meant uuid as far as i can tell, at least it is what we have in all the examples (e.g. https://github.com/hl7ch/hl7ch-cda/blob/master/projects/HL7.ch/HipJointReplacement-de/v2.0/2-Radiologischer%20Befund.xml)..
in the examples all id's are uppercase but as grahame notes in his log there are different versions (uppercase, lowercase),
so I don't know if we could fix it to a lower or upper case variant in switzerland and I think you're assumption of both cases are valid is correct (however I don't remember how art-decor schematron handles it).

in fhir it is specified as lowercase.

Tracking Open 2020-09-03 17:44:29 : Tracking by Quentin Ligier
Description
Thank you for the references Oliver. So if I understood correctly, the CDA spec requires it to be either an OID, UUID or RUID and CDA-CH limits it to UUID only?
For the letter case, it's sad HL7 did choose uppercase as lowercasing UUIDs is recommended by the original RFC and required by FHIR. As it's not specified here, should we assume both cases are valid?
Tracking Open 2020-09-03 17:14:18 : Tracking by Oliver Egger
Description
the base CDA spec defines ClinicalDocument.id as Instance Identifier (II).

InstanceIdentifier root is defined as UID 
cda_r2_normativewebedition2010/infrastructure/itsxml/datatypes-its-xml.htm#dtimpl-II

and the spec defines it as an either UUID or OID, the format of the UUID is defined also in CDA
( A DCE Universal Unique Identifier is a globally unique string consisting of 5 groups of upper- or lower-case hexadecimal digits having 8, 4, 4, 4, and 12 places respectively.)
See also http://www.healthintersections.com.au/?p=1762 where its pointed out that UUID should be uppercase.

The description in the cda-spec with GUID is misleading, I propose the wording should be corrected in a future version to UUID.


Tracking Open 2020-09-03 15:05:55 : Tracking by Quentin Ligier
Description
Finding:

The document's ID is described as a Globally Unique Identifier (GUID) but no formal definition is available.
In RFC4122 , GUID is given as a synonym of UUID but it's not quite the case in real world usage (Microsoft's own GUIDs are not 100% compatible, RSS's GUIDs are alpha-numeric strings, etc.).
The Schematron currently seems to accept an UUID, an OID or an alpha-numeric string (datatype 'uid'?).
What exactly is the expected format for this ID? Is it an RFC4122 UUID or a 128-bit alpha-numeric UUID-like string, an OID, a free alpha-numeric string?

-

Suggestion:

-

Further explanation:

-
    
Labels
Preview Code HTML color Display Name Description