The LabMCoP is working on the best way to report blood cultures (last notes on this topic are here: https://aphlinformatics.atlassian.net/wiki/spaces/LMCOPL/pages/5354488134/2026-09-17+LabMCoP+Meeting+Notes) - both the results, but also contamination and these 2 questions came up:
#1 Reporting growth of yeast in blood culture bottles: When setting up blood cultures you usually draw 2 sets - each set should have an aerobe and anaerobe bottle; each bottle should have 10 mL. When working up these cultures, Nancy said they can grow bacteria and yeast - which isn’t always contamination. Labcorp example we looked at: 008300: Blood Culture, Routine Information | Labcorp - this uses LOINC https://loinc.org/600-7 = Bacteria identified in Blood by Culture. If yeast is grown how would you report this result?
a. Using https://loinc.org/600-7 even though yeast is not bacteria
b. Reporting https://loinc.org/600-7 as no growth and reporting yeast name under https://loinc.org/601-5, either as a reflexed test, or just as additional result
c. another way
#2 Reporting possible contamination: Currently we think most labs will create a comment added to the organism result, when contamination is suspected - something like this: “A Single-set positive out of two or more sets may indicate the presence of possible skin contaminant, clinical correlation required. Please call the laboratory if further work–up is needed.”
This works well for humans reading it, but is not easily computable (as other comments could also be attached to the organism naming result. In order to more computably identify blood cultures with suspected contamination, e.g. when a commensal organism is identified, should we create a new LOINC with Yes/no answer for Suspected Contamination? The above note could still be sent either with the commensal organism result or with this result, but it would make it easier to analyze data.