Stream: Medication
Topic: prescribe with kinetics and no dose form
Olivier Boux (Aug 27 2021 at 13:15):
I need to prescribe, for example, "tramadol, prolonged release, 100 mg BID, oral route", without specifying a particular dose form (tablet, capsule, …).
In Medication, contained in referenced by MedicationRequest, I can express tramadol, but not tramadol prolonged release.
Kinetics is usually coded as a property of the dose form, e.g., prolonged release tablet or prolonged release capsule.
Thus it is currently not possible to specify kinetics in a prescription without choosing a dose form.
A solution would be to add to the Medication resource the drugCharacteristic element as it is defined in MedicationKnowledge, assuming that one of the possible characteristics is the kinetics.
Thoughts on this?
Jean Duteau (Aug 27 2021 at 13:50):
Yes, that does seem like something to do - specify the "prolonged release" as a drug characteristic extension.
Jose Costa Teixeira (Aug 27 2021 at 18:49):
We have the same challenge. Indeed the medication can be added with the tramadol. Then for more characteristics, we could include a MedicationKnowledge, but since there is no link between Med and MedKnowledge, we cannot contain the medicationKnowledge, and we'll need to add extensions for those attributes
Jose Costa Teixeira (Aug 27 2021 at 18:49):
Same thing for ATC class
Jose Costa Teixeira (Aug 27 2021 at 18:51):
@Olivier Boux happy to see you here
Jose Costa Teixeira (Aug 27 2021 at 18:56):
@Jean Duteau Device has just that, for same/similar purpose: a link to the DeviceDefinition where more characteristics of the product are available. Is there a downside of having that for medication?
The upside is simple: Sometimes you express a product identifier and want to attach more product characteristics in the same data object.
Jean Duteau (Aug 27 2021 at 18:58):
there is a link from MedKnowledge to Medication, so you can contain a MedKnowledge in a Medication if you want.
Jose Costa Teixeira (Aug 27 2021 at 19:16):
Ok thanks.
I was not expecting that link direction. What is the meaning? I understand the other way around (as "this medication resource is fully defined / has more information over there"
Jose Costa Teixeira (Aug 27 2021 at 19:18):
Jean Duteau said:
there is a link from MedKnowledge to Medication, so you can contain a MedKnowledge in a Medication if you want.
Just a note - in mine and Olivier's cases I think we'd need to contain them both in the medrequest, not one inside the other
Jose Costa Teixeira (Aug 27 2021 at 19:19):
But if Olivier is suggesting that Medication should have the DrugCharacteristic element, I would be in favour of that (Extension would work for now, but I think it's a good idea to think about it)
Jean Duteau (Aug 27 2021 at 19:23):
given that Medication is about to go Normative and we haven't had this use case before, I suspect it will not be accepted into the core. that is why I suggested an extension would be the best place for this.
Jean Duteau (Aug 27 2021 at 19:28):
depending on your drug code database, you may have a drug code for this. as an example, in my drug database, I have codes for "NIFEDIPINE EXTENDED RELEASE", so I would just use this code in my Medication.
Jose Costa Teixeira (Aug 27 2021 at 19:48):
I think the use case was suggested before
Jose Costa Teixeira (Aug 27 2021 at 19:49):
in my use case there is no code for the ensemble of attributes I want to convey - that is a (the?) goal of openMedicine/UNICOM AFAIK
Jose Costa Teixeira (Aug 27 2021 at 19:50):
(this is why I recall the requirement/use case being suggested before)
Jean Duteau (Aug 27 2021 at 20:00):
i just checked the JIRA issues for Medication and it was never raised there before.
Jose Costa Teixeira (Aug 27 2021 at 20:32):
not sure there was a ticket, I think not (it should, I know) this was in a meeting some years ago. And the discussion was short. Sorry made you look.
Lloyd McKenzie (Aug 27 2021 at 23:19):
Sort of agree with Jose that the direction of the linkage seems backward. (Though on Medication I'd expect it as an extension rather than core)
François Macary (Aug 31 2021 at 07:16):
@Olivier Boux was requesting the addition of "drugCharacteristic" as a 0..* (code, value[x]) pair to medication. @Jean Duteau you say Medication is jumping from FMM 3 in R4 to Normative in R5, which precludes this addition? In that case, is reversing the link between Medication and MedicationKnowledge doable? It makes sense from Medication to MedicationKnowledge. Not the other way round.
By the way, the Medication References section fails to capture MedicationRequest, MedicationDispense and MedicationAdministration. I assume that this computed list takes in account only Reference datatype, and misses CodeableReference. Shouldn't it capture it as well? (@Lloyd McKenzie )
Lloyd McKenzie (Aug 31 2021 at 13:50):
Moving to normative doesn't prevent changes. Some elements can be marked as remaining STU even if the overall resource progresses to normative.
François Macary (Sep 01 2021 at 09:39):
OK. So Olivier and I will post a jira with the addition to Medication described here.
@Lloyd McKenzie how about my question above regarding including CodeableReference references to the "References" sections of the resource, so as to have the full list of resources referencing the one at hand?
Jose Costa Teixeira (Sep 01 2021 at 11:12):
I think that is a publishing question - whether CodeableReferences appear on this References section.
@Grahame Grieve ?
François Macary (Sep 01 2021 at 16:32):
Lloyd McKenzie said:
Moving to normative doesn't prevent changes. Some elements can be marked as remaining STU even if the overall resource progresses to normative.
Jira created: https://jira.hl7.org/browse/FHIR-33329
Grahame Grieve (Oct 12 2021 at 23:23):
fixed. will become visible sometime this week
Jose Costa Teixeira (Nov 16 2021 at 21:33):
Back to the issue, I see it was resolved because UNICOM is working on it but there needs to be action by ISO TC215?
Jose Costa Teixeira (Nov 16 2021 at 21:34):
I don't know if anyone reached out to UNICOM about this, and I'm not sure whether Pharmacy expects this to be UNICOM or ISO TC215 WG06 ?
Jose Costa Teixeira (Nov 16 2021 at 21:36):
@Melva Peters @Jean Duteau what is the feedback needed?
Jean Duteau (Nov 16 2021 at 21:44):
@Christof Gessner was on the call so he might be able to help, but basically once the Unicom project decides how drug characteristics for a particular dosage form are to be modelled, we didn't want to pre-anticipate how to model this ourselves. We expect that we would create a standard extension with the details needed.
Jose Costa Teixeira (Nov 16 2021 at 23:05):
Ok so we can trigger UNICOM first, and then ISO). I think the ask is how "drug characteristics for a particulat dose form are to be modeled"
Last updated: Apr 12 2022 at 19:14 UTC
