pH in urine by test strip about loinc2hpoannotation

[kingmanzhang]

@jpgourdine There are potentially many similar LOINC tests for similar things. What you observed above is pretty frequent. "pH in urine by test strip" (5803-2) and "pH in the urine by automated test strip" (50560-2) are DIFFERENT lab tests (have different LOINC codes); at the same time, they are testing on the same thing (pH in urine). Generally, those LOINC codes will be annotated to same HPO terms.

An extreme of the above example can be demonstrated with glucose chanllenge tests, in which a patient is fed with 100g glucose and then urine glucose is measured at 0.5hr, 1hr, 2hr, 3hr, 4hr, 5hr, 6hr. Each time point measurement is actually coded in a separate LOINC code. This is decided by the LOINC people and the purpose is to modularize information transmission. How do we annotate those LOINC codes? I think we should annotate them to same HPO terms, which is to say if the patient has high urine glucose at 0.5hr or 1hr or 2hr etc, he has hyperglycemia. This might not make sense in our application because we may decide that three values has to be higher than normal to call it "hyperglycemia". But we will delay this judgement to the application of the annotations. @pnrobinson @nicolevasilevsky @LCCarmody We may need to discuss this.

[kingmanzhang]

I reopen this ticket for other people's input.

[kingmanzhang]

@jpgourdine Hi JP, what do you mean that you were "not able to flag it or to suggest a term"? Why not?

[jpgourdine]

I think I mistook pH in urine by test strip" (5803-2) and "pH in the urine by automated test strip" (50560-2). One is already (5803-2) and the other is not. I will edit a comment on that.

[pnrobinson]

fixed the error.