In Canada we use LOINC to ask the question, so would map our Urine Culture to 630-4 as you have identified. For the result value of e coli, we would map to SNOMED CT and would put the concept ID from SNOMED CT in the Organism dictionary if possible.
LOINC coding is for the purpose of the lab question and is not used for the lab answer or result value.
If the dictionary won’t accept SNOMED CT codes, then at best you could create an internal coding system by identifying each possible result value with a code identifier of your own, and set that up in your table. I am not sure if that would make resulting any easier?
While previewing my response, I see my dear friend and colleague Lorie provided you a response. I concur with her. Here’s my response.
I would suspect that the field might be more correctly labelled as a terminology standard code field. A catalog or listing of genus/species is a perfect location for SNOMED CT codes. The general rationalization is that LOINC would encode what is being tested, while SNOMED encodes non-numeric answers.
You can set this listing aside from the LOINC mapping project, but please keep it as a smaller project on the “to do” list. It shouldn’t take very long when you do tackle it, typical organism lists and qualifier text are usually much less than 1,000 rows. Make certain the field length will accomodate the longer SCT codes, and provide feedback to the Meditech vendor.