Allow DocumentReference.category to be 0..* instead of 0..1

[gcangioli]

Background

In the IHE MHD profile, DocumentReference.category is currently constrained to 0..1. This limits the capability to classify and filter documents according to multiple classification dimensions. In some real-life implementations, this additional need has been addressed by overloading the event element, which is not ideal and reduces semantic clarity.

Both Composition and DiagnosticReport allow multiple categories, each serving slightly different purposes. It is not clear why some of this information should be lost when document/report metadata are shared through DocumentReference.

This limitation will become even more evident with the European Imaging and Laboratory Report use cases. The current EU cross-border specifications, which adopt DocumentReference for document sharing, do not impose such a restriction and support richer classification.

Proposal

Change the cardinality of DocumentReference.category from 0..1 to 0..*.

[JohnMoehrke]

It seems you are assuming that ITI-67 has constraints on the DocumentReference resource. It has no constraints on the DocumentReference. There is no requirement in ITI-67 that the DocumentReference.category is constrained to 0..1. That requirement is only enforced on ITI-65 to assure that MHD -> XDS will function without conflict. John Moehrke 🔥 Architect: Healthcare Informatics Standards - Interoperability, Privacy, and Security ***@***.*** | M +1 920-564-2067 Consulting: https://MoehrkeResearch.com Blog: https://healthcaresecprivacy.blogspot.com

…

On Wed, Nov 19, 2025 at 9:35 AM Giorgio Cangioli ***@***.***> wrote: *gcangioli* created an issue (IHE/ITI.MHD#261) <#261> *Background* In the IHE MHD profile, *DocumentReference.category* is currently constrained to *0..1*. This limits the capability to classify and filter documents according to multiple classification dimensions. In some real-life implementations, this additional need has been addressed by *overloading the event element*, which is not ideal and reduces semantic clarity. Both *Composition* and *DiagnosticReport* allow multiple categories, each serving slightly different purposes. It is not clear why some of this information should be lost when document/report metadata are shared through DocumentReference. This limitation will become even more evident with the *European Imaging* and *Laboratory Report* use cases. The current *EU cross-border specifications*, which adopt DocumentReference for document sharing, *do not impose such a restriction* and support richer classification. *Proposal* Change the cardinality of DocumentReference.category from *0..1* to *0..* *. — Reply to this email directly, view it on GitHub <#261>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AHBWC56IZ3RRYLNAECUXYUT35SE2XAVCNFSM6AAAAACMS2BF4WVHI2DSMVQWIX3LMV43ASLTON2WKOZTGY2DGMRZHE3TEOI> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

[gcangioli]

Yes this was my assumption (see also my comment on #261).

[JohnMoehrke]

Please see CP-ITI-1327 -- in Ballot 66.