Here is Simon Carley's response to my somewhat luddite posting!
Forwarded at his request.

  ------- Forwarded message follows -------
It's a little tricky suggesting a reliance on clinical acumen in the same reply
as stating that you need something for your inexperienced SHO's. The two do
not go together ;-)

I do take your point about clinical acumen - in effect the experienced clinician
is able to estimate pretest probabilities in the way that checklists may also
do (I have no problem with how someone gets there although scoring systems do
work well for some things & some people (esp. juniors on the learning curve)).


I would say that making the way in which we estimate risk of disease explicit
helps not just ourselves get our minds around the problem, but it is also a
useful way of teaching that very same "clinical acumen" to others.

As for the design of these things - well I guess that could vary depending on
what you want to achieve. However, making it presentation, rather than diagnosis
led would seem the way to go.

Blood tests do have a place, as do all other investigations, but we need to
focus usage amongst those patients in whom it will really make a difference.
(very) Generally speaking those with very high risk of disease don't benefit
because we intuitively ignore a negative result (e.g. normal ECG & CK BUT loads
of cardiac risk factors and past CABG). Similarly, those with low pre test risk
don;t benefit because we intuitively ignore a positive result (e.g. 2 hours
after direct blow to chest in fight, pain on inspiration, slightly raised d-dimers).
We often get the greatest population benefit from testing in the moderate risk
patients.

Simon

>For departments with a high throughput, excessive cerebration about test
>results introduces delay for patients and often results in indecision
>from our inexperienced SHO's.
>
>What today's busy clinician needs is a simple yes/no answer with minimum
>fuss and maximum speed and an absent margin for error. Clearly, this is
>not always achievable in every case. Certainly not with D-Dimers.
>
>This over-reliance on blood tests at vast expense to the NHS and A/E
>trolley time diminishes our standing as clinicians.
>
>What you do not want is a checklist based system where there are twenty
>items to check for each of up to twenty differential diagnoses, followed
>by waiting for the results of multiple tests.
>
>I think you will be hard pushed to beat old fashioned clinical
>judgement.
>-- 
>Stephen Hughes
>

I agree that the basis of clinical acumen is subconscious probability
weighing to some extent, as well as a good degree of pattern
recognition.
-- 
Stephen Hughes


%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%