DQ: CCDA data for Other Counts

[chrisroederucdenver]

This is a ticket for the backend work for the seccond batch of metrics, here on the CCDA side.

• [ ] visit counts by type
• [ ] implausible death and birth dates count ( year < 1900 or year > 2025 )
• [ ] data of visits per person
  • [ ] count of persons with 0
• [x] data of diagnoses per person
  • [ ] count of persons with 0
• [ ] data of medications per person
  • [ ] count of persons with 0
• [x] data of measurements per person
  • [ ] count of persons with 0
• [x] data of observation per person
  • [ ] count of persons with 0 If the widget doesn't automatically create buckets, do something reasonable and compare it to what was done in the N3C dashboards.

[chrisroederucdenver]

@AdamLeeIT please use pull requests to commit so that we can do code reviews. Questions/Comments here.

• use the JHU mapping tables:

  • https://foundry.cladplatform.org/workspace/data-integration/dataset/preview/ri.foundry.main.dataset.28fe6af8-0b22-4b45-86e2-b394c62dcd09/master
  • https://foundry.cladplatform.org/workspace/data-integration/dataset/preview/ri.foundry.main.dataset.5ba1594c-ef36-4eeb-b6c4-d3d005dbd127/master
  • ccda_counts_by_condition
  • comment the section_code = '1145-4'. In fact, it should be != 10157-6 the one for family history. The idea is that if there is a section we overlooked, it will show up. This will work if exclude family history rather than restricting to problem list.
  • ccda_counts_by_observation and measurement
  • Could you comment why the counts are restricted to value_types CD and CE ? T
  • Double rows for "Age at onset of clinical finding" don't look right.
  • Though not used in the presentation, including the concept_id in the table makes it easy to eliminate one line of thought.

  The code 95971004 is what I added trying to create observations from C-CDA_R2-1CCD.xml

  3052,3053c3052,3054
  <                                   <id root="aed821af-3330-4138-97f0-e84dfe5f3c35"/>
  <                                   <code code="3094-0" displayName="Urea nitrogen, Serum" codeSystem="2.16.840.1.113883.6.1" codeSystemName="LOINC"/>
  ---
  >                                   <id root="aed821af-3330-4138-97f0-e84dfe5f3c35-1"/>
  >                                   <!--<code code="3094-0" displayName="Urea nitrogen, Serum" codeSystem="2.16.840.1.113883.6.1" codeSystemName="LOINC"/> -->
  >                                   <code code="95971004" displayName="Urea nitrogen, Serum" codeSystem="2.16.840.1.113883.6.96" codeSystemName="SNOMED CT"/>
  It comes up now as a drug in SNOMED and an observation in Nebraska.  That needs to be corrected with something from SNOMED that is an observattion...and not necessarily urea nitrogen.