Stream: conformance
Topic: VitalSign effective rule (vs-1)
Thomas Papke (Feb 01 2022 at 08:31):
For the vitalsign profile, the rule "vs-1" requires that effective datetimes "shall be precise to the day".
Is this intended to be shall be at least date granularity or exactly date only?
Having a Vitalsign observation with
"effectivePeriod": {
"start": "2021-11-08T10:00:00+01:00",
"end": "2021-11-08T11:00:00+01:00"
}
Is this "valid"?
At least hapi currently complain about this date. The validation seems to allow a date only.
https://github.com/hapifhir/hapi-fhir/blob/master/hapi-fhir-validation-resources-r4/src/main/resources/org/hl7/fhir/r4/model/schema/vitalsigns.sch#L56
Is the validation in hapi too strict here or this expected behavior?
Grahame Grieve (Feb 01 2022 at 09:41):
the formal definition of the invariant is ($this as dateTime).toString().length() >= 8
Grahame Grieve (Feb 01 2022 at 09:44):
but the schematron doesn't specify the end pattern must match. So the schematron looks right to me
Thomas Papke (Feb 02 2022 at 07:47):
@Grahame Grieve : Thank you for verifing. If the specification and the schematron are valid, i am still wondering wheter the mentioned example is valid or not? Currently a vitalsign with the given effectivePeriod above is rejected by the schematron rule:
"severity": "error",
"code": "processing",
"diagnostics": "vs-1: 'if Observation.effective[x] is dateTime and has a value then that value shall be precise to the day' Rule 'if Observation.effective[x] is dateTime and has a value then that value shall be precise to the day' Failed",
"location": [
"Observation.effective.ofType(Period)",
"Line 3, Col 42"
]
Grahame Grieve (Feb 02 2022 at 09:22):
it shouldn't be rejected
Julian Sass (Feb 02 2022 at 10:35):
I think the problem is in the vitalsign profile that the "vs-1" invariant is defined on effective[x] but should be on effectiveDateTime to work. Because $this as dateTime returns an empty result if $this is of type Period.
Grahame Grieve (Feb 02 2022 at 20:07):
oh interesting. Yes, that'll be the problem, and HAPI will be using the FHIRPath not the schematron - I should've paid attention to that.
Thomas Papke (Feb 03 2022 at 13:30):
Hello @Grahame Grieve . If i understand your remark correctly, this is a short bug in hapi and it make sense to create a ticket for this in hapi?
Alexander Zautke (Feb 03 2022 at 14:22):
It's a bug in the specification. See https://jira.hl7.org/browse/FHIR-29654
Alexander Zautke (Feb 03 2022 at 14:23):
It's a bit worrying that we are moving towards R5 and this change has not been applied yet
Lloyd McKenzie (Feb 03 2022 at 15:22):
We have 800 outstanding unapplied change requests. Volunteers to help are welcome...
Last updated: Apr 12 2022 at 19:14 UTC
