There's no particular benefit to worrying about sharing value set definitions unless they're complex or hard to maintain or there needs to be an authoritative version. Having 7 or 10 different value sets that all say "all DIN codes" won't negatively impact interoperability. Valueset sharing is more important when you're going to have a complex set of filters and enumerations. For example a value set of "all comorbidities for hypertension" might be one worth sharing to ensure consistency (and to allow the benefit of the thought and work that goes into maintaining it to be shared)

Thanks Lloyd, that is consistent with my understanding.

Coming back to whether or not there is a need to define a ValueSet for DIN (full or a subset of DIN), it depends on the use case. One potential use case is if one is working on a FHIR implementation guide/conformance package for a project, and certain resource data element needs to be using DIN code, the right way of defining the FHIR conformance resources are defining a Manufacture Drug valueset, define the coded data element to ValueSet binding in a FHIR profile (StructuredDefinition), and having the FHIR IG resource reference both the profile and the valueset.

Bindings in v3 and FHIR are always to a ValueSet. However, in the instance you don't generally identify the value set, only the code.

I'm not sure in HL7 V3 or FHIR, you could bind a coded data element/attribute directly to a CodeSystem? Looking at the updated HL7binding syntax http://wiki.hl7.org/index.php?title=Binding_Syntax Regardless if it's "direct" or "indirect binding" (or in V3 the "model binding" vs "domain binding"), the code element/attribute is supposed to be bound to a ValueSet?

I understand in Coding or CodeableConcept in FHIR, you don't necessarily need the valueset, but from Implementation Guide and it's referenced conformance resource package perspective, the way to go still seem to be defining (or referencing existing) valuesets. Someone who is more familiar with HL7 Vocab or CGIT group could probably share more insight on this.

No, we didn't. I'm not sure there is a need to define a value set for the full DIN.

Thanks Igor. I am aware of this code system. Just wondering if you also created an associated value set URI which is meant to contain the full DIN.