Epic generates MANY encounters which are not care events and do not need to be billed, hence having no diagnosis. I suggest that this is only a problem if we have no diagnoses for encounter types AV, IP and ED. Otherwise the diagnostic information is to be found in the orders events Jim
From: [email protected] [mailto:[email protected]] On Behalf Of Nathan Graham Sent: Tuesday, August 26, 2014 1:41 PM To: [email protected] Subject: ETL Annotated Dictionary Bug: Diagnosis Table includes many encounters with no diagnosis Importance: High Phillip, As discussed, I'm forwarding to gpc-dev to continue the conversation regarding entries in the CDM diagnosis table with no diagnosis code (see ticket 144<https://informatics.gpcnetwork.org/trac/Project/ticket/144> and Annotated Data Dictionary (BitBucket)<https://bitbucket.org/njgraham/pcori-annotated-data-dictionary>). I've created a bug report: "Diagnosis Table includes many encounters with no diagnosis"<https://bitbucket.org/njgraham/pcori-annotated-data-dictionary/issue/1/diagnosis-table-includes-many-encounters>. From just a quick look, it appears that we have many rows with NULL diagnosis at KU as well - I think I may want a join rather than a left join as you suggest. Anyone can create issues (or at least that was my intent when configuring the BitBucket account). I like feedback and want everyone to feel free to report issues or add comments, etc. I wonder if it just adds more confusion to use the BitBucket bug tracker and also have the ticket on informatics.gpcnetwork.org<https://informatics.gpcnetwork.org/trac/Project/ticket/144>. I guess we'll see how it goes... Thanks. Regards, Nathan From: Phillip Reeder [mailto:[email protected]] Sent: Tuesday, August 26, 2014 10:10 AM To: Nathan Graham Subject: i2b2 to PCORI Importance: High On line 418 of the CDM_transform.sql, you do a left join of the encounter table to the diagnosis. For me, this is creating a ton of encounters that don't have an associated diagnosis. I haven't tracked down why that is the case yet, but I'm guessing it has to do with some of our fake encounters we generate, along with unsynced encounters between systems and such. Should this be a join instead a left join, or possibly revers the order so that it is diagnosis left join encounter? That way we would ensure that all rows had a non-null diagnosis. And if for some reason, an encounter/visit didn't exist for the row, the diagnosis would still show up. Thoughts? Phillip ________________________________ 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
