I have noticed quite a few archetypes making mention of particular FHIR resources, ‘CIMI’, HL7v2, and other data / interop standards. Unless the standard referenced is absolutely immutable, i.e. it is a past release, this is not a good idea. It may not be a good idea anyway, because these standards come and go in time.
Particular issues:
- CIMI is not something we follow, and as far as I know, is not publishing much within HL7. I would suggest these references be removed.
- FHIR resources such as the AllergyIntolerance once were sort of close to the openEHR counterpart but have since been remodelled, and text in the CKM archetype is likely to be out of date. Much of FHIR that might relate to CKM (i.e. not infrastructure resources) is still changing, and cannot be reliably linked at a detail level to openEHR archetypes.
In the past, we used to publish the openEHR primary (abstract) specifications (Reference Model and so on) with CEN/ISO 13606 and also HL7v3 cross-references. The latter standard is now obsolete, and the former has been greatly altered by the 2018 version such that all previous analysis of its correspondence to openEHR was not just obsoleted, but actually wrong. When we upgraded the publishing system to the current web-based one, I had to take hundreds of these references out from 1000 pages of specification.
Referring to very specific and stable standards, such as an IHE HL7v2.5 lab profile is probably safe.
In general however, I would suggest these kinds of cross-references be maintained in separate documents, e.g. wiki pages or similar. Otherwise archetypes risk containing text that has to be changed / removed later on as the standards change, or their relevance changes.