We don't have loinc in our component table, but they are associated with the individual results. I used those to create our map from component to loinc. That left a large number of lab components still un-mapped so I used the common name field to map labs with the same common name to the same loinc code if one was already mapped.
That got me pretty high, then I manually mapped the top 100 or so labs by volume. Currently, over 90% of our results are mapped. I've validated a few of the labs on a study and it seems to work fairly well to get what the researcher was interested in. Phillip Sent from my iPhone On Apr 6, 2016, at 4:04 PM, Dan Connolly <[email protected]<mailto:[email protected]>> wrote: At KUMC, our experience was similar to yours: Ooh! the CLARITY_COMPONENT table has a LOINC mapping column! Darn; it's mostly empty. We somehow got a component_id -> loinc code spreadsheet or something from the hospital--a by-product of their work on MU2, I think. Jim Campbell and the UNMC dev team<https://informatics.gpcnetwork.org/trac/Project/wiki/DevTeams#unmc> managed to do better than that, I gather. They shared Prototype ETL for CDM V2 Lab LOINC<https://datahandbook.epic.com//Reports/Details/9000297> (in Epic Userweb, linked from #241<https://informatics.gpcnetwork.org/trac/Project/ticket/241>). We have a couple CERNER sites who have figured this out somehow. I don't know how they did it, but perhaps they would like to chime in. p.s. copy to gpc-dev, with permission. -- Dan ________________________________ From: Visweswaran, Shyam [[email protected]<mailto:[email protected]>] Sent: Wednesday, April 06, 2016 12:43 PM To: Dan Connolly Subject: question about obtaining LOINC codes for EMR data Hi Dan, I am a faculty member at Pitt Biomedical Informatics working mainly on the NCATS ACT network but also a little on our PCORnet. Michele Morris here suggested that you may have some answers to the following questions. We have Epic and Cerner here and so far have been manually building a mapping table for labs from local codes to LOINC codes. Do you know if Cerner or Epic or both provide LOINC map tables for labs? I have been looking into Clarity here and came across a table called CLARITY_COMPONENT which looks like it provides a mapping between clarity components codes and LOINC but the LOINC column is mostly empty (only 384 out of 16,683 lines have LOINC codes). We are building a new version of the NCATS ACT ontology and we are hoping to expand the current ~150 labs in the ontology to ~2000 commonly ordered labs. If local mapping tables have to be built at each site, the ETL for the expanded lab ontology would be a lot of work. So I am investigating if there are an y general solutions at least to common EMRs like Epic and Cerner. Thanks. Shyam Visweswaran, MD, PhD Associate Professor, Biomedical Informatics University of Pittsburgh School of Medicine _______________________________________________ Gpc-dev mailing list [email protected]<mailto:[email protected]> http://listserv.kumc.edu/mailman/listinfo/gpc-dev ________________________________ UT Southwestern Medical Center The future of medicine, today.
_______________________________________________ Gpc-dev mailing list [email protected] http://listserv.kumc.edu/mailman/listinfo/gpc-dev
