Hi Jim,

Thanks for the good discussion on gpc-dev yesterday on the medication modifiers 
for pcornet and the potential to load procedure orders in the procedure table.


1.       Procedures and procedure orders

I am very interested in orders as the driver of activity and so having them in 
alongside the billing representations could be very nice for informatics 
research proposals that want to generalize across several sites.

I did some investigation this morning after talking with Michael, Matt and Dan.

Now that we are using Epic for charge capture, it's possible to get coverage on 
especially lab orders by linking to the hospital billing transaction log in 
Epic and seeing where the hsp_transactions match to the order_proc and vice 
versa on the specific order transaction (order_id or order_proc_id); note that 
for some things like a Basic Metabolic Panel, the proc_id (concept of the 
order) is Epic's ordering table (678 for us) doesn't match the transaction 
table (has 10233657 as the proc_id)  because I think our preforming lab is 
using Sunquest so there may be an intermediary relationship somewhere.  Both 
are in the CLARITY_EAP table.

/* this looks at the order and finds matching transactions  */
select cht.tx_id, cht.cpt_code, cht.order_id, cht.pat_enc_csn_id, 
cht.procedure_desc, cht.proc_id, cht.service_date, cht.tx_filed_time, 
cop.order_time, cop.description, cop.order_proc_id,  cop.result_time, 
cop.ordering_date, cop.proc_code
from CLARITY.ORDER_PROC cop
left join clarity.hsp_transactions cht on cop.pat_enc_csn_id=cht.PAT_ENC_CSN_ID 
and cop.ORDER_PROC_ID= cht.ORDER_ID
where cop.pat_enc_csn_id =---put in a csn id here---
order by cop.order_time
;

/* this looks at the transactions and finds matching orders  */
select cht.tx_id, cht.cpt_code, cht.order_id, cht.pat_enc_csn_id, 
cht.procedure_desc, cht.proc_id, cht.service_date, cht.tx_filed_time, 
cop.order_time, cop.description, cop.order_proc_id,  cop.result_time, 
cop.ordering_date, cop.proc_code
from clarity.hsp_transactions cht
left join CLARITY.ORDER_PROC cop on cop.pat_enc_csn_id=cht.PAT_ENC_CSN_ID and 
cop.ORDER_PROC_ID= cht.ORDER_ID
where cht.pat_enc_csn_id =---put in a csn id here---
order by cop.order_time
;

In our environment we get pretty good hits to CPTs for the labs and occasional 
hits for ECG

But lots of other good stuff like nursing activities don't to a CPT (which is 
why I suppose Michael Prittie's work is research).  But it's possible someone 
might make a order to flowsheet measure to LOINC mapper and then crosswalk 
those to the orders for things like "Vital Signs".

Question to Jim: was your bias to build an interoperable ontology based on CPT 
hierarchies like we pull out of the UMLS?  Or, did you have one developed based 
on LOINC by linking to the reported results related to the parent order?

-          I think having someone prototype this on real data is going to be 
key to see how usable it will be.

-          My bias would be to support some things as interoperable like a BMP 
by exploiting this billing link but also allow sites to load other procedure 
orders with their local terminology to at least expose the magnitude of the 
problem (as we've done at KUMC on babel in our procedure orders ontology).  We 
built our local procedure orders hierarchy off native trees in Epic and there 
are probably pros and cons here.  Could do a CPT tree with added branches for 
stuff like nursing and consults not in CPT and base those branches off perhaps 
something Dan Vreeman and the LOINC team might be creating.

Question to SCILHS but also Charles with Pedsnet and Data committee members: is 
anyone else creating an interoperable procedure order ontology and bringing 
into the CDM along with OHDSI or i2b2?  Did you go down the road of CPT/billing 
based or LOINC/observables based?



2.        Medications.  I updated ticked 280 
https://informatics.gpcnetwork.org/trac/Project/ticket/280 with my thoughts and 
would like to hold off until we see the thinking emerging from the cancer CRG 
and the recommendations of the team doing the antibiotic study interfacing with 
the DRNOC.

Russ Waitman, PhD
Director of Medical Informatics
Associate Vice Chancellor for Enterprise Analytics
Professor, Department of Internal Medicine
University of Kansas Medical Center, Kansas City, Kansas
913-945-7087 (office)
[email protected]<mailto:[email protected]>
http://www.kumc.edu/ea-mi/
http://informatics.kumc.edu<http://informatics.kumc.edu/>
http://informatics.gpcnetwork.org<http://informatics.gpcnetwork.org/> - a 
PCORnet collaborative

_______________________________________________
Gpc-dev mailing list
[email protected]
http://listserv.kumc.edu/mailman/listinfo/gpc-dev

Reply via email to