Stream: v2 to FHIR
Topic: Connectathon 24
Hans Buitendijk (May 13 2020 at 20:52):
We will kick off v2-to-FHIR Connectathon 24 in 8 minutes. Coordinates: https://zoom.us/j/8295536160.
Ivan Quan (May 14 2020 at 13:09):
will the group be communicating in this stream?
Camila Altman (May 14 2020 at 19:05):
Hi all,
I can't remember who asked yesterday for some ADT messages that I can strip out the PHI from and we could use for testing. What fields exactly were you interested in again?
Camila Altman (May 15 2020 at 16:05):
HI all,
These are the fields I see in ADTs from our EHR that have no mappings:
AL1-9
NK1-34 , NK1-35
PV1-20,PV1-42, PV1-44, PV1-47
PV2-2, PV2-7,PV2-10,PV2-21,PV2-22, PV2-40
Also we are missing GT1 as a segment overall or I can't seem to find it. Then eventually when A18 MRG is missing. I can see that would be a patient update and a delete?
Ivan Quan (May 15 2020 at 16:09):
Anyone playing around with the hbs templates? I have a question in regards to how patient identifiers are referenced in observation resources. I'm use to seeing obs resources subject reference valued with relative path "Patient/" followed by business identifier. For example, LAB-ORU-1.hl7 PID.3.1 is 10006579 for Donald Duck so I'm expecting subject reference to be "/Patient/10006579", but actual transformation from ORU_R01.hbs is "Patient/64bac34e-e611-3549-848b-89416176aa0b". Is this right? Are there fhir best practices around using unique id generators for this field?
Camila Altman (May 15 2020 at 16:34):
HI Ivan,
I don't know the answer but I have their emails from yesterday. modillon@microsoft.com; ranku@microsoft.com. Can you let me know what they say
Ivan Quan (May 15 2020 at 16:34):
yep
Ivan Quan (May 15 2020 at 16:41):
Camila, I don't see any mappings for A18 either. I'm not sure what the process is to start a new message type mapping. However, did find some helpful info on Patient.link https://www.hl7.org/fhir/patient.html
Hans Buitendijk (May 15 2020 at 17:01):
@Camila Altman Thank you for the missing fields that you use in you messages. We'll put them on the list to establish an agreed to mapping. If you have suggestions, please include them in the comments in the spreadsheet.
Regarding GT1, there is no map yet. Do you have many messages using that? Also, if you have a draft map, would be great.
Hans Buitendijk (May 15 2020 at 17:02):
@Ivan Quan If you provide a list of typical ADT trigger events, we can start to put them out re-using segments we already mapped. Question will then be whether based on context those segment mappings need to be adjusted. But it would be a start.
Camila Altman (May 15 2020 at 17:29):
@Ivan Quan Thanks! I will check that out.
@Hans Buitendijk I have many many messages with GT1. I have not drafted any mapping yet
Vassil Peytchev (May 15 2020 at 17:53):
Ivan Quan said:
I have a question in regards to how patient identifiers are referenced in observation resources. I'm use to seeing obs resources subject reference valued with relative path "Patient/" followed by business identifier. For example, LAB-ORU-1.hl7 PID.3.1 is 10006579 for Donald Duck so I'm expecting subject reference to be "/Patient/10006579", but actual transformation from ORU_R01.hbs is "Patient/64bac34e-e611-3549-848b-89416176aa0b".
If all you have is a business identifier, you need to use a logical reference, where Observation.subject.type="Patient" and Observation.subject.identifier.value="10006579"
Ivan Quan (May 15 2020 at 19:22):
@Camila Altman Ranvijay Kumar from microsoft replied: *All the resources are referenced using the resource ids. That is why you see subject reference Patient/64bac34e-e611-3549-848b-89416176aa0b.
As part of conversion, we generate new resource ids. These resource ids are GUIDs, but not random. These are generated using generateUUID helper function that takes an input parameter and outputs GUID. For a given input, the output is always the same. By default, we use the PID.3.1 as input to generate the GUID.
If you want to retain the PID.3.1 as the resource id, you can modify the template from ‘ID=(generateUUID PID-3-1)’ to ‘ID= PID-3-1’. Note that you will also need to modify all the references from ‘REF=(concat 'Patient/' (generateUUID PID-3-1))' to ‘REF=(concat 'Patient/' PID-3-1)’*
Last updated: Apr 12 2022 at 19:14 UTC
