FHIR Chat · Observations - multiple reference or copies? · implementers

Stream: implementers

Topic: Observations - multiple reference or copies?


view this post on Zulip ...Paul (Apr 23 2020 at 05:28):

In trying to suss out the relationship between a DiagnosticReport and Observations, it's clear the DiagnosticReport will have 0 or more Observations associated with it. The Observation data model makes no specific reference back to a specific DiagnosticReport, so what I can't be sure about is whether or not it is expected that an Observation may be included in _multiple_ DiagnosticReports, or if the set of Observations for any given DiagnosticReport is a completely separate set (even if the dates, et al, are the same)?

In the JSON schema, I noticed this bit under "Observation":

        "contained": {
          "description": "These resources do not have an independent existence apart from the resource that contains them - they cannot be identified independently, and nor can they have their own independent transaction scope.",
          "items": {
            "$ref": "#/definitions/ResourceList"
          },

...which implies that Observations included under a DiagnosticReport are, in fact, unique elements of that report and should only be referenced "through" the report -- but in the HTML documentation, there's a reference to the "last n Observations query" (https://www.hl7.org/fhir/observation-operation-lastn.html) which makes apparent that these records _can_ be referenced "independently".

Can an Observation be referenced by multiple different DiagnosticReports or does each belong to only one report?

view this post on Zulip Grahame Grieve (Apr 23 2020 at 05:31):

you're right to be unsure. THe specification does not prohibit a given observation being referenced from more than one report. That's deliberate. In general, though, it would be unusual for a source system to do that - typically, the diagnostic report owns the observations - they are part of it's life cycle. Though you might get more than one version of the same diagnostic report referencing the same observation, if it doesn't change

view this post on Zulip Lloyd McKenzie (Apr 23 2020 at 05:32):

There are no rules prohibiting the same Observation from being referenced in multiple DiagnosticReports, though that would not be common.

The 'contained' element is only used for resources that cannot stand independently and are generally not accessed independently. The intention with a DiagnosticReport is that all of the referenced Observations exist as independent instances, not sent as 'contained' resources.

view this post on Zulip ...Paul (Apr 23 2020 at 05:34):

Cool, cool. Would it be expected, though, that an Observation referenced in a DiagnosticReport would, though, be also referenced by the other types, such as "Appointment", "ServiceRequest", etc?

view this post on Zulip Grahame Grieve (Apr 23 2020 at 05:34):

couild be - specially as evidence for sometihng

view this post on Zulip ...Paul (Apr 23 2020 at 05:35):

Excellent, thanks for the advice, guys!


Last updated: Apr 12 2022 at 19:14 UTC