Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Medication - identifier vs code #29

Open
rockphotog opened this issue Dec 10, 2019 · 5 comments
Open

Medication - identifier vs code #29

rockphotog opened this issue Dec 10, 2019 · 5 comments
Labels
bug Something isn't working kodeverk Spørsmål knyttet til kodeverk

Comments

@rockphotog
Copy link
Member

Hei

Jeg tror det riktige stedet for f.eks. FEST-id er under identifier, ikke code. FEST-id er en "business identifier", mens code er "A coded concept that defines the type of a medication." og "Codes that identify this medication" (ikke en helt god definisjon).

Det er bare lov med én code, så det er opplagt at når vi skal leve med parallelle koder (som IDMP, FEST-id, ++) er det identifier som er riktig.

Ping @olemartinwinnem @thomiz @oaassv

Se også: #16

@rockphotog
Copy link
Member Author

Må også ta en kikk på no-basis-Substance.

@rockphotog rockphotog added bug Something isn't working kodeverk Spørsmål knyttet til kodeverk labels Dec 10, 2019
@thomiz
Copy link
Member

thomiz commented Dec 11, 2019

Jeg mener vi har diskutert akkurat denne saken før, men det er bra å få dokumentert det her. Hovedpoenget fra den gang var vel at man kan anse FEST-id som en kode fra et kuratert avgrenset mengde, FEST blir ihvertfall forvaltet på en måte som gjør en slik tolkning mulig. Så kan man alltids innvende at FEST-id kanskje er en misvisende term. Kodeverket FEST identifiserer/koder både virkestoff, virkestoff med styrke, pakninger og doser i tillegg til handelsvarer.

Siden man kan ha mange coding innenfor en CodeableConcept, så man kan representerer flere koder fra flere kodesystem for å definere samme konseptet innenfor en CodeableConcept (og derfor innenfor en Medication). Derfor kan man kode parallelt med både FEST og ATC kode (og eventuelt IDMP-kode) for å identifisere Medisinen.

Spørsmålet blir kanskje hvordan man skal betrakte Identifier i denne sammenhengen. Det er ikke åpenbart at vi trenger en Identifier ei heller hva den skal bety. Det er vel også grunnen til at denne ikke er beskrevet spesielt i no-basis profilen. SLV kan i prinsippet tilby en tjeneste som gir oss Medications ressurser med en fastsatt koding og en statisk forretningsidentifikator som de lagrer.

Dette er også i tråd med hvordan dette benyttes internasjonalt. Identifier for Medication benyttes ikke (profileres ihvertfall ikke), men det velges hvordan man skal kode medication (ATC, tysk-medisinkodeverk pzn, us-core medication codes rxnorm, nederlandske produktkoder).

Verdt å merke seg at us-core er alene om å håndheve en streng koding av medication:
Prescribable medications The codes SHALL be taken from US Core Medication Codes (RxNorm); other codes may be used where these codes are not suitable

@stinejohans1
Copy link

stinejohans1 commented Dec 13, 2019

Jeg spurte Rik Smithies (som jobber for EMA og FDA om "IDMP inn i FHIR") om noe liknende ": how does the link between Medication and MedicinalProductDefinition (in both R5 and R4), work? I can’t quite see it? "

Svaret var:
There is no direct reference but it works via Medication.code = MedicinalProductDefinition.identifier

The links are codes. So, a medication may have the SNOMED code for a drug, and the MedicinalProductDefinition.identifier would be the same "code".

This is the same way that a SNOMED code would link to the same code in a CodeSystem or a ValueSet. It is not a direct FHIR reference. Often the definition of the code is not in any FHIR accessible place, is just "out there" somewhere (in some database).

Code vs identifier seems is a little inconsistent it is true. But I think it is perhaps correct, because the code is a property of one ("has"), but the definition property of the other ("is") (but its arguable).

There has been talk of changing Medication so it has a choice of a code or a reference (to MedicinalProductDefinition). But Medication is not being so actively worked on as the IDMP resources, and we are still slowly working through such issues. The pharmacy and the regulatory/IDMP worlds are a little disconnected and we didn’t want to upset and confuse the pharmacy people by loading all our big regulatory stiff in there. But now the regulatory is getting more established we will want to work out how to link things together better. Another alternative would be to replace references to Medication to a choice between Medication and MedicinalProductDefinition. That seems logical to me, but again it would probably confuse. So the code=identifier or code|reference choice method is probably what we will go for.

@vadi2
Copy link

vadi2 commented Dec 13, 2019

There has been talk of changing Medication so it has a choice of a code or a reference

Just an update on this - CodeableReference has now been introduced in R5 - and the change will apply across the spec: https://chat.fhir.org/#narrow/stream/179294-committers.2Fannounce/topic/CodeableReference

@thomiz
Copy link
Member

thomiz commented Oct 22, 2021

Se også #16

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
bug Something isn't working kodeverk Spørsmål knyttet til kodeverk
Projects
Status: Utvikle
Development

No branches or pull requests

4 participants