FHIR Chat · Data Absent Reason - why? · argonaut

Stream: argonaut

Topic: Data Absent Reason - why?


view this post on Zulip Michele Mottini (Aug 06 2021 at 21:27):

https://www.hl7.org/fhir/us/core/general-guidance.html#missing-data instructs server to output a data absent reason extension for mandatory elements for which there is no data available, but I sort of fail to see the point?

view this post on Zulip Michele Mottini (Aug 06 2021 at 21:28):

If the client need that data the fact that there is an extension saying why the data is not there does not really help in any way

view this post on Zulip Michele Mottini (Aug 06 2021 at 21:28):

If the client does not need that data ditto

view this post on Zulip Michele Mottini (Aug 06 2021 at 21:29):

and the server needs extra code to generate extensions that client cannot really use for anything

view this post on Zulip Michele Mottini (Aug 06 2021 at 21:30):

and also server author have a sort of escape hatch to get out of the required elements requirement - that in some case is just a fact of life, but maybe if there was no escape hatch they would try harder..

view this post on Zulip Lloyd McKenzie (Aug 07 2021 at 03:01):

This was a workaround to make data elements mandatory that actually aren't. Agree that it's not an appropriate use of DAR - that extension should only be used if there's a legitimate need to capture and use the reason why the data is missing. If the expectation is that it SHOULD always be there, make it 0..x and must support. And if you want to jump up and down about it being really important, put a warning or best practice invariant on the parent that yells if the element is missing. At least then you're not requiring systems to capture and display information they don't need. (I.e. the code indicating why the element is misisng.)

view this post on Zulip Eric Haas (Aug 07 2021 at 04:41):

Why? - Because real life is messy

view this post on Zulip Lloyd McKenzie (Aug 07 2021 at 14:42):

That's not a very satisfying answer Eric. The base requirement of "this is important and SHOULD always be present, but we accept that occasionally in the real-world it can't be" is fine. However, there are some elements where, if missing, you're better off not having the record at all (so DAR is inappropriate). As well, there are mechanisms of declaring the 'SHOULD' in ways that don't involve requiring support for capturing a reason that no-one actually caries about.

view this post on Zulip Eric Haas (Aug 08 2021 at 00:05):

I appreciate your opinion @Lloyd , however after several rounds of successful ballots will ample US implementer feedback who understand the real world limitation vis a vis standards, we feel this is the best approach.

view this post on Zulip Michele Mottini (Aug 09 2021 at 14:50):

Thanks Lloyd


Last updated: Apr 12 2022 at 19:14 UTC