Re: [Gnumed-devel] demographics editor

[J Busser]

At 2:04 PM +1000 9/16/04, Richard Terry wrote:
> I find this whole debate interesting or should I say ridiculour,  for the
> following reasons.
> Imagine you use a paper system. Before sitting down to your desk, you look up
> your manual appointment system and find out who the last patient was. Then
> you go to the file and get out their records. Then you take that file and sit
> at your desk. Next you look up who you really want to see next.

Nice lampoon but a bit off base. Let me give an alternative.

You are with your patient in the exam room. You receive a phone call 
which you must take in your private office, so you must choose 
between leaving the chart open (with the next patient's charts 
accessible underneath), or you close the chart, and either leave it 
on the desk within the exam room, or you put it temporarily in the 
slot just outside the exam room door. You go attend to your phone 
call, and return to your exam room. Where is the chart? Your 
secretary has put it away! Now you have to do initiate an extra step 
to get it back, and to wait for it.

Or your patient had left, and you moved the chart to the comfort of 
your private office, to complete your note and a test request, but 
you realize you will be late for a meeting or you must go urgently to 
the hospital.  So in your absence, and on your return, the chart will 
serve as a useful placeholder for how far you had gotten in your work 
flow. If you still need the chart, you have no delay. If you do NOT 
still require it, there was no harm in it having stayed there during 
your absence.

It is true that in the electronic scenario, the time taken to 
fetch/refresh the patient's data will have been wasted if in fact you 
do not need to access their record. If just prior to your last 
logout, you had somehow communicated to GnuMed "I have finished with 
this patient, put the chart away [loose the focus]"), no patient 
value would be saved at logout, and that way upon next logging in you 
do not need to incur the wasted fetch. But on the other hand, if you 
had NOT finished with that patient, there will have been some value 
to what I had suggested. The fact that GnuMed retains, in the search 
box, a list of recently activated patients reduces the work of 
retrieving the last patient so may not justify the extra work 
required to "dismiss" the last patient before logging out.

I see nothing wrong to consider a work flow (though it disagree with 
your own) that recognizes having been in the midst of a task when you 
had to log out, and therefore to help you rejoin GnuMed in a way that 
lets you resume that work, with a minimum of interruption.

> This whole business of pulling in either 'anyone' or the last patient seems
> ludicrous to me. Why on earth would you want to to it. I cannot think of a
> situation where I've ever wanted either a random patient, or the last patient
> I've just seen when I first boot up my program. How confusing for users.

Nowhere did I suggest a random patient, in fact I specified we should 
*NOT* re-activate the last patient if the user had changed.