I wanted to add my own tuppence to Robbie's very well articulated points.
I've just made a series of short videos on how computers can 'help' the
consultation process (I recall asking acad-ae-mailbase readers for help -
belated thanks to those that offered services, inc Robbie I think!). One of
the films showed a telephone consultation, in which the questions the
(nurse) operator asked were prompted by the computer. I thought this would
be the easiest film to make, since the software was pretty mature, and has
been used in a number of previous studies. What went wrong was that I
assumed the nurse, already being an expert in the field (genetic risk
assessment in cancer) and answering calls like this all day long in her job
with the ICRF's cancer helpline, would have no problem using the software.
But no. Our nurse has developed her own "patter" in handling calls, which
works for her, and is flexible and responsive to the different personalities
who call her. Although a lot of thought has gone into the way the this
particular system is designed, I didn't really (in retrospect) give the
nurse enough time to think about how she was going to alter her "patter" to
fit around the order of questioning and information giving the system
imposed. The result was a very clumsy consulation (and a fair few retakes
and some heavy editing).
One of the criticisms of NHS direct has been that the nurses manning it seem
to be more interested in 'treating the computer' than talking and listening
to the patient. I suspect NHS-direct staff are encountering the same
problems I encountered - i.e. that the computer system imposes too rigid an
order on the asking of questions, and ends up distracting perfectly capable
clinicians from doing their job, whilst they instead "treat the computer".
This is not because the principle of computer systems supporting telephone
helplines is necessarily fundamentally flawed, but because few programmers
have a inkling of how reactive and flexible a clinical line of questioning
has to be - it's a tad more emotionally charged than buying a plane ticket
or choosing car insurance. Unfortunately, most programmers (understandably)
have a rather linear, INPUT - PROCESS - OUTPUT view of life ...
So I agree with Robbie - clinicians should take on a greater role in
developing these tools than just acting as advisors on the content, and
start forcing the designers to make tools that are USEABLE, not just USEFUL.
I guess Manchester Triage works because patients generally understand that
it is not a "consultation", and are thus willing put up with a more
The films will be available on the web over the next few weeks - digital
streaming technology permitting!
Clincal Research Fellow
ICRF Advanced Computation Laboratory
And A+E SHO from time to time too.
Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com