Stream: terminology
Topic: ucum {ScoreOf}
Richard Townley-O'Neill (Feb 10 2022 at 00:11):
What does ucum {ScoreOf} mean ?
See https://www.hl7.org/fhir/valueset-ucum-common.html
Grahame Grieve (Feb 10 2022 at 01:06):
nothing, pretty much
Grahame Grieve (Feb 10 2022 at 01:06):
an arbitary unit with an uninterpretable meaning
Lloyd McKenzie (Feb 10 2022 at 02:03):
Then why is it in our list of common units?
Lloyd McKenzie (Feb 10 2022 at 02:03):
In fact, why do any of our units have comments in them?
Grahame Grieve (Feb 10 2022 at 02:12):
it was assembled from what people are actually using in practice. That unit probably came from canada
Lloyd McKenzie (Feb 10 2022 at 02:24):
Ok, but that doesn't mean they should be in the list of common units.
Grahame Grieve (Feb 10 2022 at 04:10):
you asked me why it was there, not why it should be
Richard Townley-O'Neill (Feb 10 2022 at 04:23):
The Apgar examples in Core (e.g. http://hl7.org/fhir/R4/observation-example-5minute-apgar-score.html) have {score} as the unit for Apgar scores. What would be appropriate here? A quantity with no unit at all?
Lloyd McKenzie (Feb 10 2022 at 04:32):
Would you object to a change request to strip all the comments (and strip all extra units made duplicative once the comments were removed)?
Lloyd McKenzie (Feb 10 2022 at 04:33):
And yes, having no unit is appropriate for an APGAR score.
Lloyd McKenzie (Feb 10 2022 at 04:33):
(Though using a code with an ordinal value would probably be better than using Quantity)
Richard Townley-O'Neill (Feb 10 2022 at 04:39):
@Lloyd McKenzie
I debated what it should be, code, integer or quantity. Why code?
Richard Townley-O'Neill (Feb 10 2022 at 04:39):
VS = {0,1,2,3,4,5,6,7,8,9,10}?
Richard Townley-O'Neill (Feb 10 2022 at 04:52):
In terms of removing the comments, will systems cope if "/{Specimen}" is removed?
Lloyd McKenzie (Feb 10 2022 at 05:24):
The components are definitely coded ordinals as each number means something specific. What to do with the overall score is less clear as there isn't a defined meaning for each numeric value (beyond high=good, low=bad).
Lloyd McKenzie (Feb 10 2022 at 05:25):
Removing "/{Specimen}" won't make that suddenly illegal - any valid UCUM code is still permitted. It just won't be in the value set used for drop-downs. Which will hopefully nudge people to not use units for things that aren't. "per specimen" isn't a unit - it's a qualifier on how the specified amount is supposed to be used.
Richard Townley-O'Neill (Feb 10 2022 at 05:30):
Seems like Apgar components are codes and the total score is a Quantity with limited numeric values and no unit.
Lloyd McKenzie (Feb 10 2022 at 05:31):
Could just be an integer too. Given that units are not relevant.
Eric Haas (Feb 10 2022 at 23:41):
just to clarify the Apgar examples:
- each component has a code a ...eg. "2" AND an ordinalValue extension which is an integer 2
- These examples based on the corresponding LOINC Lforms.- There is probably an NIH LHC form using Q/QR doing the same thing.
- Creating examples in the FHIR spec is an absolutely thankless task and I don't recommend it.
Richard Townley-O'Neill (Feb 10 2022 at 23:42):
Examples are very valuable. Thank you for creating some.
Richard Townley-O'Neill (Feb 11 2022 at 05:33):
Given that the data type count includes the ucum code '1', should that be added to the valueset ucum-units?
Grahame Grieve (Feb 11 2022 at 05:34):
probably
Richard Townley-O'Neill (Feb 11 2022 at 07:12):
https://jira.hl7.org/browse/FHIR-36010
Lloyd McKenzie (Feb 11 2022 at 17:31):
And FHIR#36012
Last updated: Apr 12 2022 at 19:14 UTC
