In message <[log in to unmask]>, N.P. van Duijn
<[log in to unmask]> writes
>Bayesian thinking / decision analysis is a necessary mental training
>for clinicians and students, not a sensible clinical method at the
>bedside.
It can be very helpful at the bedside (or deskside usually in my case).
It is particularly useful when patients want a diagnostic test "just to
make sure" and you can demonstrate *before* you are persuaded to do the
test what the possible results are. So if false positive much more
likely than true positive, you don't do the test. The only problem is
finding the sensitivities and specificities of the tests, but you can
still show the pronciples even if you don't have these to hand.
I find that patients are often amazed by the concept of false positives
and false negatives (it never occured to them that tests were other than
100% accurate)
>It is comparable to training in patient-physician comunnication,
>algortimic diagnostic thinking, knowledge of practice guidelines,
>and training in physical examination; neither should be applied
>in full, nor always, nor in detail.
>
Partly agree - but sometimes it's useful to go through the whole
process. I do it with a 2x2 table and calculator in about 5-10 minutes,
but quite rarely and have found patients very appreciative that they are
helped to understand what is involved
Toby
--
Toby Lipman
General practitioner, Newcastle upon Tyne
Northern and Yorkshire research training fellow
Tel 0191-2811060 (home), 0191-2437000 (surgery)
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|