Stream: implementers
Topic: Quantity Comparators for "Approximately"
Richard Kavanagh (Jun 25 2018 at 09:54):
We have a need to represent approximate measurements in FHIR for a project where we are mapping a legacy data flow. The flow has the following value comparators. When using Observations with Quantity then the "required" valuseset for comparator at http://hl7.org/fhir/STU3/valueset-quantity-comparator.html does not include "approximately" - any suggestions on options here...
Lloyd McKenzie (Jun 25 2018 at 10:18):
All of the others allow for strict mathematical comparison. My leaning would be to treat "approximately" as something else - though, given it modifies the meaning, having it as core would be good. Can you submit a change request?
Grahame Grieve (Jun 25 2018 at 10:29):
we discussed this at length early in the project, and looked at both PPD and NPPD in the v3 spec, and the openEHR approximate approach
Grahame Grieve (Jun 25 2018 at 10:30):
in the end, we just found that there was little use in practice, hence, leaving it to extension.
Grahame Grieve (Jun 25 2018 at 10:30):
I don't think it would be a modifier extension, though
Lloyd McKenzie (Jun 25 2018 at 10:46):
Ignoring the extension when I see x = 5, I infer that x is definitely not 4. Paying attention to the extension, I see x~=5, where I infer that x may be 4. If paying attention to the extension negates your prior understanding, that makes it a modifier...
Grahame Grieve (Jun 25 2018 at 10:49):
I think you've gone out on a limb with that intepretation
Lloyd McKenzie (Jun 25 2018 at 10:52):
I don't understand how it can be out on a limb. If an extension can falsify an understanding reached by ignoring the extension, then the extension can't be safe to ignore.
Richard Kavanagh (Jun 25 2018 at 11:29):
Thanks you've cofirmed my suspicion that we need to use an extension. I've had a look and can't see any standards ones so we will create our own. We'll discuss within our implementation team whether we go for modifierExtension or not.
With regards to "raising a change request" then I'm not sensing a concensus there, so I'll leave it for others to progress that.
Lloyd McKenzie (Jun 25 2018 at 14:32):
You actually can't have modifier extensions inside Quantity, which is part of the challenge here.
Noemi Deppenwiese (Nov 20 2020 at 16:01):
I recently came across the same problem. Have there been any developments on this in the meantime?
Lloyd McKenzie (Nov 21 2020 at 17:10):
No changes to the spec in this space.
Alexander Zautke (Dec 01 2020 at 17:28):
Would there be sufficient interest in defining a standard extension? @Noemi Deppenwiese and I would like to avoid to define an extension for this purpose in the realm of HL7 Germany as a national extension.
Lloyd McKenzie (Dec 01 2020 at 22:52):
You can certainly submit a change request for us to add a 'standard' extension. We'll have to settle the question of whether it's a modifier or not.
Last updated: Apr 12 2022 at 19:14 UTC
