Open JohnMoehrke opened 3 years ago
when a client receives multiple patient identities (id / identifier) for a given patient, which value do they use for which server. unlike with PMIR where PMIR gives a single golden patient.id to use within that community scope
whitepaper might become handbook
whitepaper would help inform Connectathon 'floor' layout. How multiple QEDm test servers would be presented relative to their patient identities. whitepaper primary audience are those with the problem of multiple servers hosting patients that are cross-referenced using PIXm (not PMIR).
see https://github.com/IHE/IT-Infrastructure/blob/master/Proposals/Patient.id%20in%20FHIR%20Environments.pdf
=Original Problem= I want to write a FHIR client application that will push/pull/… FHIR clinical resources (Observation, …..) If my target is an environment that has a centralized system that owns/services all of the resources (Patient, Observation, ….), I can make a lot of simplifying assumptions and work in that environment. My angel investors will be happy. If my target is an environment where the resources are distributed (say Radiology Diagnostic Reports live in a Radiology only system), then the assumptions are not so easy to make. My angel investors will not understand why this is so hard.
=What Could IHE ITI Do?= ● Per Charles, HL7 FHIR has not addressed how to handle the issue of distributed systems. IHE does not solve it. ● Same conditions, but IHE writes a white paper that says: ○ Here are the assumptions in the consolidated environment. ○ Here are unresolved issues in the distributed environment ● Write a white paper that says: ○ Here are the assumptions in the consolidated environment. ○ QEDm and other profiles need to be amended to support queries that include patient.identifier so we do not need to assume a patient resource id. That might open another can of worms ● There are likely other short term options that would educate Steve ○ Who would then follow guidance at IHE Connectathons rather than inventing