[Gnumed-devel] another vista at abnormality/relevance flags

[Karsten Hilbert]

Let's look at this from another angle. There's two questions
a user asks herself:

        *Should* I sign this result ?

The answers are IMO:

1) Yes, because no-one else signed it so far and so I'm
   likely the first to see it.

2) Yes, because I am the responsible clinician, in which
   case I don't care who else signed already (but I sure
   may care about their *opinion* in making up my own).

3) No, because neither am I responsible nor am I the first
   to see this result (and they did sign it). In which case
   I still *can* sign if I choose to.

Case 1) and 2) would be flagged as unsigned in the review grid.

Case 3) would not regardless of whether there actually is a
review or not on the result (the tooltip will, of course,
inform me of that).

So this marker in the UI doesn't really mean "this is
unsigned" but rather "I should sign this".

*Then* there's the second question:

        Whose review do I want to see straight away ?

IMO the answer is:

1) Mine if available.
2) Else the responsible clinician's if available.

Note that the order of 1) and 2) can be debatable.

3) Anyone else's if there's *only one* available.
4) The most recent of anyone's ?!?

Let's assume we add a field "primary provider" to each
patient:

1) Mine if I am the primary GP AND if available.
2) Else the responsible clinician's if available.
3) Else mine if available.

else see above 3) and 4).

Does that make sense ?

The whole issue, of course, still begs the question:

        *Should* there be more than one review per result ?

Or let's paraphrase that: Should I be able to document a
review differing from that of the repsonsible clinician if
need be ?  (Note that that's talking about just two reviews
but technically means "any number of".)

I think so. Or maybe not, forcing me to seek clinical debate
with the other clinician which likely results in better
care. In the systems I *regularly* work with (3 different
ones for that matter) this would be solved by putting a
free-text note in the chart stating my opinion.

Karsten
-- 
GPG key ID E4071346 @ wwwkeys.pgp.net
E167 67FD A291 2BEA 73BD  4537 78B9 A9F9 E407 1346