Community feedback: Use cases, data sources, goals, struggles, and everything in between

[rtmill]

Please use this discussion to post the details leveraging oncology data in OMOP at your organization. 
- What particular use cases you're trying to address
- What data source(s) you hope to ingest
- Any struggles or hurdles with implementing the model 
- Any general questions or requests for feedback

For concise and transparent fixes where discussions are not needed, such as bugs and missing or incorrect concepts and relationships, please instead create an issue ticket, ideally with an attached spreadsheet of the content to which you are referring. 

[rtmill]

<
NOTES from EU OHDSI Workshop 2023/07/02

TODO: Make into tickets:

// Condition
ICD03 completeness
ICD03 archaeology (standards change over time) see: "Transmart" for possible solution
Management of triple data of diagnosis in cancer OMOP --> need additional type concepts for path note, etc.
Missing dates in source data --> need specific rules for various types of "dateless" events
Uncurated vs. curated data (generally too high level of an issue?)

// Cancer Modifiers
-Handling of negative concepts
-Migration of IHC from binary data to more detailed as we introduce HER2+ into routine care '
-Quality metadata
-Laterality (Christian: is it critical to have this? Everyone else: Yes )
-Genomics ( give canned answer )
-Unconsented somatic mutation data from NGS panels (?)
-Source normalization (?)
-Information about detection method/test (?)
- Signatures (??)
- More up-to-date germline variation annotation on BRCA1 etc.
- Integration of radiomic and digital pathology data

Treatment Regimen

• list should be universal (not US centric)
• Handling chemo-radiation
• Treatment intent and treatment plans

Point of care

• Distinguishing general vs. specialist surgeon
• Distinguishing cancer department visits from other visits in a general hospital
• How to manage multi disciplinary teams
• Consolidation of libraries of care quality compliance signatures

Research
-Prevalence studies with US claims "lives covered" mindset excluding national cancer registries
-Outcome validation
-Where is the data how can we identify collaborators in the community
-Phenotyping (???)

Others

• Problems for the vocabularies committees and open source ETL automation tooliong
• Implicit coding
• Implementation of episode table