Re: [Gnumed-devel] Measurement workflows - value colour red , display of � , and gnuplot auto-invocation?

[Busser, Jim]

On 2013-07-14, at 9:35 AM, Karsten Hilbert <address@hidden> wrote:

> What about grid cells containing more than one value: In the
> extreme case there can be a low, a normal, and a high value
> in the cell.
> 
> Use one color for "out of bounds" AND show the lab indicator
> (OR add a +/- (or up/down arrow)) ?
> 
> You do realize we are losing the value of
> .clinically_relevant, aren't you ?

I started to think about that in my last post.

There is some value to keeping aware that a patient value remains "abnormal" as 
defined by the lab, even when we do not find the result to be clinically 
important.

For example, an HDL cholesterol value of 1.6 mmol/L might be reported by the 
lab together with a normal range of 1.0 - 1.5 and so would be "high".

By definition, this result is abnormal because the value falls outside the 
range in which most people fall.

It seems to me redundant for the clinician to re-define, in GNUmed, whether 
something is abnormal when the lab has already done it.

It seems to me that the utility, in GNUmed, of

        Sign as: [    ] Abnormal

is limited to those situations where the clinician manually-entered the result, 
along with any situation where the lab lacked the means to supply along with 
the test result a determination of whether or not it was abnormal.

This then brings us to "clinically relevant (important)"

- I can see the value of leaving this "unchecked" for something like incidental 
findings as commonly occur with imaging, for example hepatic hemangiomas and 
hepatic and renal cysts.

- for lab tests however, these generally should not have been requested if they 
were not "relevant"

- when should a clinician "check-mark" a cholesterol level as "irrelevant"

-- JIm