Stream: fmg
Topic: QA - GF#13064 - Partial approval
Grahame Grieve (Mar 15 2017 at 22:12):
GF#13064 - patient modifiers clean up.
Grahame Grieve (Mar 15 2017 at 22:12):
+1
Brian Postlethwaite (Mar 15 2017 at 22:28):
+1
David Hay (Mar 15 2017 at 23:57):
+1
Lloyd McKenzie (Mar 16 2017 at 00:29):
Not sure I agree on deceased. If they're deceased, then for many purposes, they lose their capacity to be a patient anymore. It's the same sort of effect as active = false - and that's a modifier.
Grahame Grieve (Mar 16 2017 at 00:34):
actually, the patient record still exists and is active. billing still occurs. Some activities are no longer sensible
Lloyd McKenzie (Mar 16 2017 at 00:35):
Right. Like I said, from some perspectives, they're no longer a patient. Patient.active=false doesn't necessarily stop billing from happening either. (Takes a lot to stop billing from happening :))
Grahame Grieve (Mar 16 2017 at 00:36):
should because it's saying that the record is not valid. not that the person is not valid
Lloyd McKenzie (Mar 16 2017 at 00:37):
Neither the name nor the definition limit the use to invalid. Can absolutely be used for "not a patient of this clinic anymore" too.
Lloyd McKenzie (Mar 16 2017 at 00:39):
(The fact we can't distinguish between 'not a patient anymore' and 'was never a patient' is a bit of an issue . . .)
Grahame Grieve (Mar 16 2017 at 00:39):
that's not the issue on the table here.
Lloyd McKenzie (Mar 16 2017 at 00:42):
The issue is "does being deceased change the fundamental nature of the resource instance or not?" And I'm arguing that it changes it in exactly the same way that Patient.active does.
Grahame Grieve (Mar 16 2017 at 00:44):
then you're wrong. you can (mis)use Patient.active to indicate something similar to Patient.deceased, but that's not why it's marked as a modifier
Lloyd McKenzie (Mar 16 2017 at 00:45):
It's a modifier because it impacts the answer to the question "is this instance really still a patient or not?". And so does deceased date. (birthDate, address, gender, etc. do not have that impact)
Grahame Grieve (Mar 16 2017 at 00:46):
it really still represents a patient. And that's not what relates to the definition of isModifier
Lloyd McKenzie (Mar 16 2017 at 00:47):
For some purposes it will represent a patient, for others it won't anymore.
Lloyd McKenzie (Mar 16 2017 at 00:47):
I don't see a significant difference between the impact of Patient.active and Patient.deceased on implementer behavior
Grahame Grieve (Mar 16 2017 at 00:51):
well, clearly, we're not going to come to agreement on this anytime soon
Lloyd McKenzie (Mar 16 2017 at 01:00):
+1 to everything except deceased
John Moehrke (Mar 16 2017 at 01:01):
I understand the status field being an indication of the status of the Resource instance; not the status of the human described. I understand the deceased indicators to be an indication about the human. I thus think that status is clearly a modifier, but deceased is simply an attribute about the human described. Thus...
John Moehrke (Mar 16 2017 at 01:01):
+1
Grahame Grieve (Mar 16 2017 at 01:05):
as it stands now, approved for everything but deceased. Need someone else to vote for that @Hans Buitendijk @Josh Mandel
Brian Postlethwaite (Mar 16 2017 at 01:19):
We have lots of use of the patient resource while the patient is deceased and very active. AgedCare institutions and Community Care berevement stuff.
John Moehrke (Mar 16 2017 at 01:22):
I agree, deceased is not a modifier. It is a fact about the human, much like all the other elements.
Brian Postlethwaite (Mar 16 2017 at 04:24):
Since its not marked as approved for deceased, hows this for text?
This element is labeled as a modifier because Lloyd wouldn't agree that it wasn't, and once a patient is marked as deceased, the actions that are appropriate to perform on the patient are significantly different. E.g. An organ harvesting is appropriate on a desceased patient, and less appropriate on a non-deceased patient.
Grahame Grieve (Mar 16 2017 at 04:28):
+1
Brian Postlethwaite (Mar 16 2017 at 04:29):
Yes, the seruous text has been updated to:
This element is labeled as a modifier because once a patient is marked as deceased, the actions that are appropriate to perform on the patient may be significantly different.
Brian Postlethwaite (Mar 16 2017 at 04:59):
@Lloyd McKenzie I've updated the tracker with voting information, can you check that I've done this right when you get to it.
Lloyd McKenzie (Mar 16 2017 at 05:18):
@Brian Postlethwaite This vote was just to allow the change. The actual vote needs to come from the owning WG - but you can treat it as a pre-applied change
Brian Postlethwaite (Mar 16 2017 at 05:21):
Too bad if the owning group doesn't like it (not that I'm expecting it)
Lloyd McKenzie (Mar 16 2017 at 05:22):
Yes, that would be problematic . . .
John Moehrke (Mar 16 2017 at 11:55):
With this logic we would then need to mark as a modifier just about everything. Even if someone simply changes their name, it is no longer appropriate to address them by their old name. This seems like it is a modifier, in the way deceased is being described. Stronger argument can be made for gender. Less strong for telecom, communication, generalPractitioner, managingOrganization, address, maritalStatus...
Grahame Grieve (Mar 16 2017 at 11:56):
well, further discussion deferred to Madrid
Paul Knapp (Mar 22 2017 at 10:01):
+1 and not deceased
agreed that active is talking about the record, not the person (or animal) and when active=false I don't think you can safely assert anything about the record contents other than these being the contents at the point in time when the record was flagged inactive.
By that I mean, maybe the named person left the care of this facility, maybe we discovered they never existed, maybe this was a complete data entry mistake, maybe anything else.
Last updated: Apr 12 2022 at 19:14 UTC
