Thomas Beale wrote: > Sam Heard wrote: >> It is a flag that says the value is very uncertain - accuracy is not >> known - how do we say this - or a quality factor makes the reading >> very uncertain. I just want to be able to see how we express when >> accuracy is poor but not quantifiable. >> Sam >> > doesn't it mean that the value is completely useless, and that instead, > a null-flavour flag should be set in the Element, and no data value be > recorded at all?
There is no such thing as a pre-condition in clinical medicine: the clinician has to compute with whatever parameter values are provided, no matter how shabby, incomplete and inconsistent they are... Floyd and Hoare would have hated being doctors of medicine, I suspect. ;-) Tim C
