I don't see any evidence of the bug here. I get nothing like an accession number from select distinct itemname from tumor_reg_facts.
-- Dan ________________________________ From: Lenon Patrick [[email protected]] Sent: Thursday, February 05, 2015 4:38 PM To: Dan Connolly; [email protected] Subject: RE: NAACCR Ontology - Accession and Seq No? I don’t know if there’s an actual bug in your code, or if it’s just a part of the Heron code I couldn’t use as is. Your tumor_item_value is a view while I had to do mine as a table (Netezza insisted), for instance, so you may have something in the view definition that was added later. There could also be something in the definition of the NAACCR fact table (naaccr.extract_eav) that tumor_item_value draws on. At any rate, I’ve already excluded those fields from my ontology. Thx for the response. From: Dan Connolly [mailto:[email protected]] Sent: Thursday, February 05, 2015 4:30 PM To: Lenon Patrick; [email protected] Subject: RE: NAACCR Ontology - Accession and Seq No? No, the accession number should definitely not be in the observation_fact table. That would be a bad bug. I'm trying to reproduce it here... -- Dan ________________________________ From: [email protected]<mailto:[email protected]> [[email protected]] on behalf of Lenon Patrick [[email protected]] Sent: Thursday, February 05, 2015 10:42 AM To: [email protected]<mailto:[email protected]> Subject: NAACCR Ontology - Accession and Seq No? Hello again all, In building NAACCR concepts via my modified versions of KUMC’s naaccr_txform.sql and naaccr_concepts_load.sql, I’m producing concepts and facts for each value of both Accession No. and Hospital Sequence No. This seems unnecessary, since I can’t see accession number being part of a researcher’s query, not to mention possible de-identification issues. So, my questions are: 1) Is this actually the correct outcome? 2) Assuming it isn’t, did I miss a step where these two fields (or their section, 6 Hospital-Specific) are filtered out of the fact and concept builds? I’d expect that to be in the definition of the “big flat view” aka tumor_item_value. Awaiting your input eagerly. Patrick Lenon HIMC Informatics Specialist 608 890 5671
_______________________________________________ Gpc-dev mailing list [email protected] http://listserv.kumc.edu/mailman/listinfo/gpc-dev
