> >Therefore results would need to be written into unmatched results *not > >only* if the patient cannot be uniquely identified, but also if the > >test_org cannot be identified. > I disagree here. If the patient matches we should import so it becomes > available straightaway. That sounds reasonable, medically. My importer doesn't yet, though.
> I think it is reasonable to create new test_org (and org) entries based on > what info in in the results file. > (labs may mis-spell a patient's name, but if they can't get the their *own* > name right, > we have bigger problems than duplicates in test_org ;-) True. > I'm lost here, Either 1 of two situations exist: > - the results format has enough metadata to allow us to infer (HL7), or > specifies a priori (?? LDT), what coding system is being used, No it does not. LDT does not specify a coding system. In fact, no coding system is even used. Notice that a test_result code does not have to have a coding system for it to be imported. Such things can be added at a convenient time (eg never in a busy practice) ! Karsten -- GPG key ID E4071346 @ wwwkeys.pgp.net E167 67FD A291 2BEA 73BD 4537 78B9 A9F9 E407 1346 _______________________________________________ Gnumed-devel mailing list [email protected] http://lists.gnu.org/mailman/listinfo/gnumed-devel
