Proposed new Freeset format
Hi everyone
The old Freeset format (PDF with just SCTID + FSN listed) is not proving very versatile. It's currently used for 4 different products:
1. IPS (HL7)
2. DICOM
3. IHE
4. ERA/EDTA
Now that we've tested out the new format with GPS, how does everyone feel about standardising the format and migrating all freesets to the new format:
ConceptID Active FSN USPreferredTerm
We also need to discuss possible use cases for including RF2 versions of each of the Freesets (including GPS), for those who do already have SNOMED to use?
Thanks very much!
Andrew
Comments
Can we please just include the free set in the International Edition so those of us with RF2 based tooling don't have to special case this format? It won't matter to those who don't have a licence because they won't get it, but it will help us.
I do agree with @Matt Cordellthough, the lack of versioning makes it hard to understand how people will use this over time.
Okay great, so we've got more support for an RF2 version of GPS - the good news is that we've already got this planned in alongside the September 2020 GPS release, so unless we get any negative feedback on this idea, we'll publish 2x versions of the GPS from September onwards, one in the existing GPS format, and the other in a standard RF2 refset, with it's own moduleID to allow it to be optionally consumed.
Hope that helps?
Cheers!
I wondered about including an (lastChange)effectiveTime field. I think similar to what LOINC has? Though I'm not sure of the details of the intended audience, maybe they don't care. Given it's already published, probably not worth it.
If there were RF2 version(s), it's ideal to have them in a separate optional module. So as not to clutter the core any further. I notice the set really contains a little bit of everything, so implementers probably need to filter somewhat (Presumably on semantic tags) to do much with it. So if made RF2, a single refset only exactly per the GPS would be best. Any further breakdown could get messy.
FYI - the contents of the set look like this.
Body structure 1811
Clinical finding 9684
Environment or geographical location 105
Event 100
Observable entity 329
Organism 4082
Pharmaceutical / biologic product 360
Physical force 2
Physical object 260
Procedure 2745
Qualifier value 730
Record artefact 1
Situation with explicit context 388
SNOMED CT Model Component 49
Social context 54
Special concept 3
Specimen 171
Staging and scales 10
Substance 893
It seems like a pretty random assortment of concepts. Just looking at medicines, There's no codeine, paracetamol, morphine, ibuprofen, cefalexin etc... I didn't find many (if any) of the "WHO Model List of Essential Medicines" I sampled in there..Some products are available with strengths, others (most?) not.