Stream: implementers
Topic: Encounter.period
Tom de Jong (Sep 21 2020 at 15:26):
If all I know about an (outpatient) encounter is its date, do I put the date in Encounter.period.start or do I populate both Encounter.period.start and Encounter.period.end with the date? I think the latter option better clarifies that the entire encounter took place on that same date. It would be helpful if the spec's would include a best practice for some common cases like this.
Brian Postlethwaite (Sep 21 2020 at 15:37):
That sounds reasonable.
Tom de Jong (Sep 21 2020 at 15:51):
Thanks for the quick reply! I'm glad we agree :-) The current description of Encounter.period seems to be open to different interpretations though. How would I go about making it a work item for PA to add clarifying text (other than wait for a formal ballot)?
Brian Postlethwaite (Sep 21 2020 at 16:00):
Log a Jira issue I expect.
And very happy to add additional descriptive text in the definitions section.
If you had any proposed text to put in, would be happy to consider that also :wink:
Ardon Toonstra (May 03 2021 at 11:40):
Ha! We are now running into the same problem here... Also thinking of populating both .start and .end date with the same value if you only know one datetime.
Moreover, I think the Encounter resource specification can be improved. In R4 and the current build, the Encounter.period has the following comment:
If not (yet) known, the end of the Period may be omitted.
I think this contradicts the semantics of the Period datatype: 'not known' is not the same as 'on going'.
Ardon Toonstra (May 03 2021 at 11:41):
Would a Encounter.effective[x] that include a valueDateTime and valuePeriod be more explicit and an improvement?
Ardon Toonstra (May 03 2021 at 11:45):
Ah, I see there is already an open ticket: https://jira.hl7.org/browse/FHIR-28621
Last updated: Apr 12 2022 at 19:14 UTC
