Identify any loss of data when processing COVID samples through FHIR Engine

[carlosfelix2]

User Story:

As a developer, I want to know if COVID data samples used in the COVID pipeline result in data loss from using our current base schema in the FHIR engine, so I can document the missing data and create new issues to resolve them.

Description/Use Case

We need to use COVID data samples to find out if our configuration is missing anything. Chances are we will see some data we are not converting and will need to add those fields to the respective configs.

Risks/Impacts/Considerations

• There may be other missing fields in the future as we get more data, but we are going to just tackle those fields we find right now instead of tackling any possible field or permutation.
• COVID sample data can be found in the repo in various places.

Dev Notes:

• Refer to the HL7 mapping spreadsheets for how fields are being handled.

Acceptance Criteria

• Use COVID samples available in our repo (this is non-PII data) to determine if there is any missing data we are not converting.
• Document any missing data and create new issues as necessary to have the data in the final HL7 conversion.

[JessicaWNava]

We discussed this as a team and to validate this we want to use prime-router/src/testIntegration/resources/datatests/HL7_to_INTERNAL/CA-20211001-sully.hl7

Start out with sending this through the UP and comparing the data before and after, make sure it is all there. The suggestion was using transforms like the default sending transform paired with the Colorado receiver transform.

Then, run it through the COVID pipeline for Colorado and compare output. Likely this will expose the gap created by the covid-19 schema existing in the covid pipeline and the lack of an equivalent disease transform in the UP. This then gets into the question of whether or not disease transforms should be a thing which may need to be answered as part of this ticket.