Stream: implementers
Topic: Not Pregnant: Condition or Observation.value false?
Alexander Henket (May 22 2019 at 13:16):
I've read up a lot in this stream for pregnancy and other conditions, but so far I've not found how to say "Person is known to be not pregnant".
Currently we seem to be forced into an Observation with a valueBoolean false because Condition really does not seem to support "NOT" having a Condition. Do I miss something?
Lloyd McKenzie (May 22 2019 at 14:35):
Normally you'd capture the specific test you did for pregnancy - with a negative result.
Alexander Henket (May 22 2019 at 14:37):
I guess that will happen too. Predictor stick for example. All our functional people gave though is a Y/N indicator. We don't know how they know (they might not either). Hence it is hard to go down that angle
Lloyd McKenzie (May 22 2019 at 14:39):
I'd still capture that as an Observation. It's a declaration as of a point-in-time based on some (unknown) method.
Lloyd McKenzie (May 22 2019 at 14:40):
The fact you were declared to be non-pregnant 2 weeks ago doesn't mean we don't need to test you again today if we're going to prescribe something new.
Alexander Henket (May 22 2019 at 14:49):
Right. That confirms our thinking afaik.
Yunwei Wang (May 22 2019 at 14:57):
I think you could use Condition resource with Condition.verificationStatus=refute (Refuted:This condition has been ruled out by diagnostic and clinical evidence.)
Lloyd McKenzie (May 22 2019 at 14:59):
@Yunwei Wang That's used when the patient has something but you've ruled out a particular cause. It's not appropriate to capture the absence of a condition that might have an onset at any time.
Chris Moesel (May 22 2019 at 14:59):
You could use an Observation with LOINC code 82810-3 (Pregnancy Status) and SNOMED-CT value 60001007 (Not pregnant (finding)). This approach is outlined here: https://www.healthit.gov/isa/representing-patient-pregnancy-status
Yunwei Wang (May 22 2019 at 15:06):
@Lloyd McKenzie Maybe I misunderstood the original question.
I was reading this section: https://www.hl7.org/fhir/condition.html#9.2.3.5
Lloyd McKenzie (May 22 2019 at 15:25):
@Yunwei Wang You are correct that Condition allows you to capture that someone doesn't have something. I'm just saying that the use of that isn't intended to cover Alexander's scenario. It's more for situations when you have a number of candidate diagnoses and you've ruled out some of them or there was a belief at one point that a patient had a given condition and you've subsequently confirmed that they don't (and never did) - presumably because they had something else instead.
Grahame Grieve (May 22 2019 at 20:03):
@Michelle (Moseman) Miller that's a surprising outcome for me - what does PC intend here? I expected Condition would be the go
Michelle (Moseman) Miller (May 22 2019 at 20:12):
The guidance in the spec is what PC last discussed/agreed to recommend: https://www.hl7.org/fhir/condition.html#9.2.3.5
I can raise the topic again to confirm, but I don't have any reason to believe anything has changed
Grahame Grieve (May 22 2019 at 20:18):
hmm ok. thanks
Michael Lawley (May 24 2019 at 01:52):
Or 250425007 | Pregnancy test negative |
Last updated: Apr 12 2022 at 19:14 UTC
