Stream: implementers
Topic: Invalid ucums in fhir valueset?
Devin Craft (Aug 23 2018 at 14:14):
Hi all,
The unit mmol/kg[H20] in https://www.hl7.org/fhir/valueset-ucum-units.html fails to validate on NLM's UCUM parser https://ucum.nlm.nih.gov/ucum-lhc/demo.html. Is the parser wrong, or is the unit actually invalid?
Devin Craft (Aug 23 2018 at 14:18):
I only know enough about UCUM to be dangerous, but the only place [H2O] appears in the ucum spec is next to a length unit (i.e. m[H2O]).
Bryn Rhodes (Aug 23 2018 at 14:42):
I think I agree, that's invalid, because it's actually H20, not H2O
Bryn Rhodes (Aug 23 2018 at 14:43):
The unit _should_ be mmol/kg[H2O]
Devin Craft (Aug 23 2018 at 14:46):
The unit _should_ be
mmol/kg[H2O]
That unit does not validate either. Note, the typo exists in the FHIR valueset as well, so that probably needs to be fixed.
The ucum spec specifies: "For example 1 m H2O is written as “m[H2O]” in The Unified Code for Units of Measure because the suffix H2O changes the meaning of the unit atom for meter (length) to a unit of pressure."
Devin Craft (Aug 23 2018 at 14:54):
My thought is that mmol/kg[H2O] is invalid because the [H2O] is not modifying the meaning of kg. Instead, it should be expressed as mmol/kg{H2O}. Actually, it probably should just be expressed as mmol/kg because the thing that is being measured should be captured elsewhere.
Stefan Lang (Aug 23 2018 at 15:08):
Looks right to me, H2O is an annotation in case of "per kilogram water", so should be in curly braces.
I suggest to file a tracker item.
Devin Craft (Aug 23 2018 at 15:09):
Thanks @Stefan Lang . Will do.
Tracker item: https://gforge.hl7.org/gf/project/fhir/tracker/?action=TrackerItemEdit&tracker_item_id=17719
Devin Craft (Sep 04 2018 at 13:10):
To anyone interested, I ran the entire valueset through the NLM's batch validator on the site mentioned above and found 9 total invalid ucum expressions. I've logged another tracker item https://gforge.hl7.org/gf/project/fhir/tracker/?action=TrackerItemEdit&tracker_item_id=17791.
Grahame Grieve (Sep 04 2018 at 21:53):
thanks. I'll look at this in a few weeks time - my run into baltimore is full
Grahame Grieve (Dec 06 2018 at 21:45):
mmol/kg{H2O}
Glory Kim (Jan 22 2020 at 19:15):
The valuset ucum common (http://hl7.org/fhir/valueset-ucum-common.html) have multiple repeating values. For example, there are two occurrences of /[LPF] one which stands for the 'per low power field' and the second which represents 'per LPF'. Additionally, there are two occurrences of [beth'U] both of which stand for Bethesda unit
Are there any future plans to clean the ucum common valuset?
Grahame Grieve (Jan 24 2020 at 05:29):
sure . do you want to create task to fix it?
Last updated: Apr 12 2022 at 19:14 UTC
