Stream: implementers
Topic: Profiling a contentReference
Chris Moesel (Feb 07 2017 at 15:47):
The Observation resource's Observation.component.referenceRange is defined via a contentReference to Observation.referenceRange. Is it possible to profile the Observation.component.referenceRange (without profiling Observation.referenceRange)? If so, how is this done? Should the profile redefine the snapshot element for Observation.component.referenceRange to *not* be a contentReferencesince following the reference would lead to the *unprofiled* referenceRange? Or is it somehow done via slicing?
Eric Haas (Feb 07 2017 at 17:12):
how about using an invariant? ( profile if component)
Grahame Grieve (Feb 07 2017 at 18:26):
yes, you would just unroll that to be explicit, and profile it accordingly
Chris Moesel (Feb 07 2017 at 20:09):
Thanks, Eric and Grahame. In these cases (which are hopefully few and far between), I'll "unroll" the contentReference to be explicit, as Grahame suggests. As always, I appreciate your help.
Chris Moesel (Feb 16 2017 at 23:09):
Followup question for @Grahame Grieve : If I "unroll" a content reference to profile it (or a part of it), do I put all of the unrolled elements into the differential, or only those that are constrained from the original definition? For example, if I unroll Observation.component.referenceRange, and constrain only the "meaning" sub-element, I assume the snapshot needs to contain the complete unrolling (low, high, meaning, etc) -- but does the differential only include the "meaning" element (and the Observation.component.referenceRange intermediate path)? Or does the differential contain it all since I unrolled the whole thing?
Grahame Grieve (Feb 16 2017 at 23:10):
great question
Chris Moesel (Feb 16 2017 at 23:21):
Ha. I had a feeling I might be in uncharted territory!
Grahame Grieve (Feb 16 2017 at 23:23):
can we get back to that in a couple of weeks time (after HIMSS!)
Grahame Grieve (Feb 16 2017 at 23:24):
(and FHIR R3 deadlines)
Chris Moesel (Feb 16 2017 at 23:52):
I think so. For now I'll do what I think is best and we can adjust as necessary at a later time.
Last updated: Apr 12 2022 at 19:14 UTC
