#90: Diagnoses Modifiers for data attribution
--------------------------+----------------------------
 Reporter:  campbell      |       Owner:  lv
     Type:  design-issue  |      Status:  accepted
 Priority:  major         |   Milestone:  data-domains2
Component:  data-stds     |  Resolution:
 Keywords:                |  Blocked By:
 Blocking:  70, 91, 120   |
--------------------------+----------------------------

Comment (by dconnolly):

 Replying to [comment:10 lv]:
 > Are there are requests from researchers at other sites to use these
 alternative types of diagnoses?

 My experience with HERON at KUMC includes at least one researcher who said
 something like "you can't distinguish primary billing diagnosis? Never
 mind then. Call me when you can." So when i2b2 1.6 came a long with
 modifiers, we enhanced HERON to use modifiers to make the distinction in
 our Sep 2012 release:
  - [https://informatics.kumc.edu/work/blog/heron-waconda-update HERON
 Waconda update introduces new searching modifiers for Medications and
 Diagnosis]

 The CDM spec says "Please do not include diagnoses generated from problem
 lists, which are a different workflow and data-generating activity". HERON
 at KUMC includes problem list diagnoses, so using a modifier is a
 straightforward way to filter them out for CDM purposes.

 The PCORNet CDM also includes `DX_SOURCE`:
    - Admitting
    - Discharge
    - Final
    - Interim
    - No information
    - Other
    - Unknown

 So PCORNet DRN queries presumably require, or at least expect, those
 distinctions to be made. I don't know to what extent it would meet PCORNet
 requirements to just map everything from your site to "Unknown" or
 "Final". Actually, we plan to have a shared mapping from GPC terms to CDM
 terms (#145), so the choice between Unknown and Final will go there. I
 suppose it would map GPC's "Billing Diagnosis" top level modifier to
 "Final".

 That's what I can think of for requirements. More on the other topics
 separately...

--
Ticket URL: 
<http://informatics.gpcnetwork.org/trac/Project/ticket/90#comment:11>
gpc-informatics <http://informatics.gpcnetwork.org/>
Greater Plains Network - Informatics
_______________________________________________
Gpc-dev mailing list
[email protected]
http://listserv.kumc.edu/mailman/listinfo/gpc-dev

Reply via email to