Stream: implementers
Topic: Use id of resource or identifier
Martin Benedict (May 04 2018 at 14:27):
Hi,
Currently I am specifying a REST interface for consuming patient information based on the Patient-resource. I am implementing the IHE PDQm supplement (which is based on the Patient-resource). There, at line 415, you find specifications for _id as well as for identifier.
If an external system provides patient data, should this system write it's own id additionally to the identifier-attribute and add it's Patient.identifier.system property or is this redundancy not recommendable?
What are your experiences?
Lloyd McKenzie (May 04 2018 at 15:45):
Think of the "id" element as a primary key for that server. The "identifier" element, on the other hand, is used for idneitifers that remain consistent across servers and that often have business meaning. You might choose to capture some other system's "id" as an identifier if you want to support mapping, but it would be unusual to expose your own internal key as an identifier.
Martin Benedict (May 08 2018 at 06:55):
Thank you! I will handle this as described.
Last updated: Apr 12 2022 at 19:14 UTC
