COVID-19 Testing Issues

[Gregory Briddick, Jr.]

Depending on how institutions are configured, their LOINC codes may or may not align with the current list of SARS LOINC codes available online. 

e.i.  In the beginning of the pandemic (many hospitals this is into August/September of 2020) all/most of the SARS-CoV testing was performed by the state's labs.  Gathering send-out labs from Epic can be problematic as they are not always in the same format, don't follow the same data fields, etc. 

If we want to construct a system that can determine long covid patients, we will need to be able to capture this send-out lab data so that we can determine when they first had a positive test and that test's date/time. 

For our system, we have all of the send-out PCR data stored as culture data (as cultures contain large alpha-numeric data fields).  The issue is that the LOINC id for cultures will capture a large swath of cultures, every send out culture.  This would include a large number of STD send-outs, special cultures, etc.  This will require either attempting to gather this data in alternative ways (looking for COVID-19 ICD-10-CM codes or other structured data elements which could indicate a positive COVID-19 test was obtained, etc.) or using a wildcard search within the culture fields to capture SARS PCRs (unknown how intensive this would be for search duration). 

Newer testing capabilities (September 2020 - present), can include various testing strategies (POC, Rapid, qPCR, qtPCR, and complete sequencing), which do not seem to be completely enumerated in the LOINC ID listing.

Gregory Briddick MSHI, BSN, RN