I appreciate your point about the more abstract nature of a LOINC code Dan, and that these are less intuitive than the metadata hierarchy that you build for your KU research community. When moving into a broader community of shared users however, the trick is to convince everyone to employ the same descriptive characteristics as you have built when identifying a data element to share in response to a query. By agreeing on the reference concept_cd system we will employ, you can build your metadata hierarchies to suit your users and I can do so for mine and yet we can answer each other's question with the correct piece of data from our systems - assuming that we share an understanding of the reference concept model of meaning.
Unfortunately there are a few concepts in the PCORI model that do not have a reference standard, but that is a minority of what we need. Jim ________________________________________ From: Dan Connolly [[email protected]] Sent: Monday, June 16, 2014 1:50 PM To: Phillip Reeder; [email protected]; Campbell, James R; [email protected] Subject: RE: [gpc-informatics] #114: Milestone 2.7 GPC harmonizes with PCORI CDM V1.0 No, we don't have a proposed demographics hierarchy; at least not in the GPC section of babel. All we have so far is Diagnoses, Meds (Drug Products by VA Class) and LOINC Codes (a mix of labs and other stuff). We seem to be iterating between the overall design issue (#114) and the specific parts (e.g. demographics #67). In this case, it really would be nicer to be further along on the specifics. Meanwhile, as noted in #67, we do have a PCORI demographics term hierarchy, and as far as I know, it suffices for GPC purposes. I'm not aware that we need any finer distinctions w.r.t. age, sex, race, nor ethnicity. Oh... and "A leading proposal was to follow i2b2's [demo] ontology". While paths like \GPC\Patient\... are handy for debugging, I'm agnostic about using less mnemonic standardized codes (as we're doing for Diagnoses) as long as I know how to look up the codes. But 29694-4 doesn't give any results when I try to Search by Codes. I lose at search.loinc.org too. 42784-9 looks fine: Ethnic background stated. I was hoping that Jim's proposal would take the form of actual i2b2 queries that he had executed, either as XML in the document or as queries on babel. For some terms, he might have to use something from the UNMC hierarchy or another site and explain how the relevant terms would end up in a shared GPC ontology. -- Dan ________________________________________ From: Phillip Reeder [[email protected]] Sent: Monday, June 16, 2014 12:36 PM To: [email protected]; [email protected]; Dan Connolly; [email protected] Subject: Re: [gpc-informatics] #114: Milestone 2.7 GPC harmonizes with PCORI CDM V1.0 Regarding specifically the PATIENTS section of document, do we have a proposed hierarchy for the patient demographics? Jim lists a path like: '\LP29694-4\LP7850-3\42784-9\%¹ (Ethnicity) But I¹m not sure where that path came from. I think if it were created on Babel and looked something like: \GPC\Patient\Ethnicity (L????) where the ?? Is the code the loinc experts decide on. Then the document would be consistent with the shared hierarchy that everyone is mapping to. For me, working in i2b2, that is more intuitive than the paths/codes that are currently shown in the document. It would look very much like the PCORI terminologies that Dan created a while ago. The same could be done for vitals with a Œ\GPC\Vitals\¹ hierarchy and other data types. Phillip On 6/16/14, 10:08 AM, "GPC Informatics" <[email protected]> wrote: >#114: Milestone 2.7 GPC harmonizes with PCORI CDM V1.0 >----------------------------------------------+--------------------------- >- > Reporter: campbell | Owner: campbell > Type: task | Status: accepted > Priority: major | Milestone: >initial-data- >Component: data-stds | domains > Keywords: PCORI CDM V1, GPC data standards | Resolution: > Blocking: | Blocked By: 17, 23, 67, > | 120 >----------------------------------------------+--------------------------- >- > >Comment (by nathan.wilson): > > Attached part I of my comments here: > >https://informatics.gpcnetwork.org/trac/Project/attachment/ticket/114/Mile >stone_27_TestSQLv4_nsw_comments_part_i.docx > > These comments focus mainly on the current selection of LOINC codes which > are to be used in the queries. Once I know what the purpose and the > expected results are, I will comment on the SQL queries themselves. > >-- >Ticket URL: ><http://informatics.gpcnetwork.org/trac/Project/ticket/114#comment:20> >gpc-informatics <http://informatics.gpcnetwork.org/> >Greater Plains Network - Informatics >_______________________________________________ >Gpc-dev mailing list >[email protected] >http://listserv.kumc.edu/mailman/listinfo/gpc-dev ________________________________ UT Southwestern Medical Center The future of medicine, today. 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
