Stream: implementers
Topic: Preliminary Reports
Michael Wang (Nov 17 2019 at 16:22):
Hi all, I think I've read through the spec, but couldn't figure out how to handle preliminary reports, specifically for labs and radiology. Should preliminary versions of a report generate separate observations (linked to the same diagnostic report) or should we replace the values on the existing observation as the values get updated. A specific example for this could be a blood culture has result "no growth to date at 24 hours" that evolves to "no growth to date at 48 hours" and finally "no growth to date a 6 days". If we generate separate observations do they all keep their preliminary statuses or do they all switch to final when the culture itself is finalized. Thanks.
Lloyd McKenzie (Nov 17 2019 at 16:45):
The same Observation instance would be updated when the final results are available and the status would be changed. No need for separate Observation instances (though they would be distinct versions of the same instance).
John Silva (Nov 17 2019 at 19:15):
@Lloyd McKenzie - "no growth to date at 48 hours" -- this sounds like a microbiology result. Isn't there a FHIR profile that suggests/recommends a way to handle microbiology results?
Lloyd McKenzie (Nov 17 2019 at 21:12):
I'm not aware of existing work in the microbiology space - @Eric Haas ?
Grahame Grieve (Nov 18 2019 at 08:27):
yes there is a profile, though I've asked OO to describe it better.
Grahame Grieve (Nov 18 2019 at 08:27):
I agree that you would keep the same observation resource (same id) and just change the status
Craig Newman (Nov 18 2019 at 13:32):
The DiagnosticReport example page gives a couple of ways to structure discrete micro results. Not sure if it's an official profile or not. http://hl7.org/implement/standards/fhir/diagnosticreport-examples.html
Last updated: Apr 12 2022 at 19:14 UTC
