Specifying what gets instantiated as observation_facts goes beyond agreement on paths.
All of the diagrams in the document are based on a diagram of the i2b2 star schema; if they're not intended to communicate constraints on table schemas, I don't understand what they're intended to communicate. In the first diagram, GPC V1 Patient Dimension, I'd rather just see that list of paths: \Demographics\ \Vital_status \Sex \Age \Language \Residence \Race \Marital_status \Ethnicity along with paths for the categorical values, e.g. \Vital_status\y\ and \Vital_status\n\ Details matter at this level; for example, trailing \ and whether the path really starts with \Demographics\ or \\GPC_Demographics\i2b2\Demographics\ to match what came up with in #63<https://informatics.gpcnetwork.org/trac/Project/ticket/63> for diagnoses: \\GPC_Diagnoses\i2b2\Diagnoses\ The other diagrams (provider, visit, ...) don't give paths at all. I made some progress on a paths-are-all-we-need presentation<https://informatics.gpcnetwork.org/2014-05-mexp/babel-paths-iff.slides.html?transition=fade&theme=sky#/> last week. Working out the details took longer than I expected (silly me; when will I learn?) and actually turned up some unanticipated details. Here's hoping I can finish these slides, to some level, before tomorrow's call. -- Dan ________________________________ From: Campbell, James R [[email protected]] Sent: Monday, May 05, 2014 4:32 PM To: Dan Connolly; [email protected] Subject: RE: agreement on paths (c_fullname) is necessary and sufficient Not my intention. I changed the slide on the patient_dimension base3d upon last week’s discussion assuming that all additional attributes other than core i2b2 features of patient_dim would be instantiated as Observation_facts and that metadata build for Demographics\ontology would follow what KU has distributed. As I recall we were not able to discuss provider and visit dimensions last week and so I kept the PCORI attributes we have to support in the model for discussion which we can have tomorrow. When we agree on the GPC standard reference then I will update the scripts for retrieval of PCORI CDM attributes that we will need to run to demonstrate harmonization with PCORI for July 1. Jim From: [email protected] [mailto:[email protected]] On Behalf Of Dan Connolly Sent: Sunday, May 04, 2014 10:35 AM To: [email protected] Subject: agreement on paths (c_fullname) is necessary and sufficient This proposal, like the previous one, directly constrains the patient_dimension, provider_dimension, and visit_dimension. Milestone_27_TestSQLv2.docx<https://informatics.gpcnetwork.org/trac/Project/attachment/ticket/114/Milestone_27_TestSQLv2.docx><https://informatics.gpcnetwork.org/trac/Project/raw-attachment/ticket/114/Milestone_27_TestSQLv2.docx> added Those constraints conflict with practices and preferences at our site (and, as I recall, several others). Plus, those constraints don't specify the concept paths (c_fullname in metadata tables) that would actually appear in the relevant queries. Agreement on paths is necessary and sufficient for our work, as far as I can tell. Given what I've seen so far, I disagree with any proposal that directly constrains the i2b2 table schema. I tried to explain how paths are sufficient in last week's call, and I agreed to elaborate in detail. But as I mentioned, I'm at a conference this week, so I'm not sure I'll have it done in time for the May 6 call. -- Dan The information in this e-mail may be privileged and confidential, intended only for the use of the addressee(s) above. Any unauthorized use or disclosure of this information is prohibited. If you have received this e-mail by mistake, please delete it and immediately contact the sender.
_______________________________________________ Gpc-dev mailing list [email protected] http://listserv.kumc.edu/mailman/listinfo/gpc-dev
