More explanations and more restrictive fhir specification

[SmalsJulien]

When an integrator looks at the specification, he doesn't know how to construct the payload to be accepted by UHMEP, the specification is too general.
After a discussion with Bart and Jean-Michel, it seems to be alright to be more restrictive for some fields and for others to write guidelines to explain what is recommended for UHMEP.

Here you will find the list of some changes :

• Subject, requester, performer
  • Reference - Accept only a ssin in the "identifier/value" field
  • (in the future) We will accept contained resources
• validity, latest, latestDraft, authoredOn, executionStart/EndDate
  • Format accepted for dateTime - YYYY-MM-DD
• statusReason
  • Bound this field with the existing values set.

[bdc-ehealth]

WG:

1. Bart will propose an invariant that will be discussed in the next meeting
2. This can be implemented

[bdc-ehealth]

WG: waiting for feedback.

[costateixeira]

Please confirm if this is useful - https://jira.hl7.org/browse/FHIR-38726
Create an extension "uncertainDate" with lower and upper bound e.g. "lower= 16:0:00, upper= 17:00:00". The software should populate that extension according to the input from the user and the rules that are defined.

[bdc-ehealth]

WG: we accept the solution for the formatting of the dates. On the second topic about the references, we have not yet reached a mature solution (we also discussing this in the architecture WG). The third topic is solved.

[bdc-ehealth]

WG: the discussion with the architecture group was not conclusive. A new meeting will take place.

[bdc-ehealth]

WG: in general, we do not forbid the use of literal references but this system (UHMEP) can only use the logical references (Patient, Practitioner)