Observational Health Data Sciences and Informatics

Meeting Notes June 19, 2019

Topics

Attendees: Rimma, Dmytry, Christian, Michael

We went over the backlog of milestone items to prepare us for beta testing. See here:

https://github.com/OHDSI/OncologyWG/issues?q=is%3Aopen+is%3Aissue+milestone%3A%22Beta+testing+preparation%22

• We discussed scoping NAACCR items to schemas. This is the requirement that all the necessary NAACCR items to diagnose a site/histology combination are linked and represented within the OMOP vocabulary. Example: Gleason Score for Prostate Cancer; HER2 for Breast Cancer. And so on. Rimma argued that this is a feature that all oncology information models need to suport. Dmytry confirmed that this is not currently done in the staging NAACCR ingestion. Michael argued that it is important but expectational and not necessary for ETLing, so it can be done in the next version. Dmytry agreed to try to accomplish this for version one of ingestion but will defer it to the next version if it is too much work. See here: https://github.com/OHDSI/OncologyWG/issues/54.
• We discussed how the fundamental site/histology data points, NAACCR item 400 'Primary Site' and NAACCR 522 'HISTOLOGIC TYPE ICD-O-3' are represented within the OMOP vocabulary. Dmytry and Michael confirmed that these NAACCR items had been curated into the 'Observation' domain and marked as non-standard. The group agreed that the ETL documentation will need to specify how to handle these “special” NAACCR items.
• We discussed that the best way to provide feedback to Dmytry on the NAACCR ingestion is to query the staging tables.
• We discussed finalizing the details of supporting numeric concepts. Dmytry and Rimma agreed to meet offline to finalize.
• We discussed mapping staging NAACCR items to NAACCR items. This is to address the fact that some NAACCR items have overlapping semantics. See here: https://github.com/OHDSI/OncologyWG/issues/51. Dmytry asked that we map staging NAACCR items without possible values to staging NAACCR items with possible values. Mike argued that he wanted to populate the staging NAACCR items without possible values with version 8 staging values from here: https://www.facs.org/quality-programs/cancer/ncdb/registrymanuals/cocmanuals. So we could map the older staging NAACCR items to the newer staging NAACCR items. Dmytry argued that we stick with version 7 staging values for the first ingestion of NAACCR into the OMOP vocabulary tables. And thus stick with mapping new staging NAACCR items to old staging NAACCR items. Mike agreed to this.
• We discussed mapping radiation therapy NAACCR items to NAACCR items. This is to address the fact that some NAACCR items have overlapping semantics. See here: https://github.com/OHDSI/OncologyWG/issues/51. Dmytry asked that we not have to create generic NAAACR extension concepts across the phases. Mike agreed to this.