Stream: implementers
Topic: text element
Brian Reinhold (Sep 11 2016 at 14:09):
This is a simple question that I have not found an answer for. IT may be buried somewhere in the FHIR site but I have not stumbled on it yet.
'text' elements show up in the resource definitions in various places. But from the examples there is a suggestion that a text element can be added anywhere. It does make sense that text could be present anywhere since they are really only for human consumption.
So the question is, can I insert a text element anywhere with another construct?
Brian Reinhold (Sep 11 2016 at 14:12):
I do not want to retype this whole message. Lost the first one.
Can I insert a text element anywhere within another element?
Lloyd McKenzie (Sep 11 2016 at 14:27):
"text" is defined in DomainResource, which is the parent of most resources - that's where the narrative for the resource goes. Some other resources have explicit "text" elements in other places. These are part of the resource definitions. You can't add "text" wherever you like though. In fact, you can't even add "extension" wherever you like. We define most elements in FHIR as including extension, but there are a few places (root of Bundle, inside Narrative.div and various other locations where extensions aren't allowed).
Brian Reinhold (Sep 11 2016 at 14:30):
Okay thanks. That answers it. So I cannot do something like Observation.valueRange.low.text = human readable string
Lloyd McKenzie (Sep 11 2016 at 14:56):
No. But you could add an extension. I'd be in favor of a general extension on Quantity that let you capture a human-readable representation. Feel free to submit a change request :)
Grahame Grieve (Sep 12 2016 at 18:12):
when would human readable representation of a quantity not be (compoarator) value units?
Lloyd McKenzie (Sep 12 2016 at 20:10):
Well, if we're talking original text or rendered text, it could be "greater than 50 milimoles". If we're talking about exceptions it could be "approximately 35".
Grahame Grieve (Sep 13 2016 at 04:34):
doesn't seem like a value proposition to me
Lloyd McKenzie (Sep 13 2016 at 15:09):
Not for core, no.
Richard Townley-O'Neill (Sep 14 2016 at 01:17):
Is there room or need for a comparator value of 'approximately'?
Of course, this opens the question of how approximate?
Grahame Grieve (Sep 15 2016 at 09:30):
yes. indeed. I've never seen it used. The only place I've ever seen it is the openEHR data type
Eric Haas (Sep 15 2016 at 18:50):
this is already baked into the Quantiy data-type type "Quantity.unit
Definition
A human-readable form of the unit."
Lloyd McKenzie (Sep 15 2016 at 19:12):
That's just of the unit, not the full quantity
Last updated: Apr 12 2022 at 19:14 UTC
