On 20-Sep-03 Adrian Midgley wrote:
> Genomic research may let us predict the best drug (or other treatment)
> for someone, compared to others with the same health need, or the
> required frequency and type of monitoring of a standard treatment on
> a better basis than we do at the moment, but I would find it handy to
> be able to show that people vary, and show that it is possible to
> look at someone and choose the best first or second hypertension drug
> for instance.
>
> I am quite sure they do, and I can, but I can't _prove_ it.
>
> Or can I?
A very interesting question. I think the difficulty in proving it by
"RCT" methods is that the effectiveness of "looking at" depends
(a) on general experience and knowledge refined into well-tuned
perceptiveness
(b) on particular knowledge and perception one has of a patient,
built up by acquaintance
How do you randomise something like that relative to the "consult
a database" approach?
> --
> From the provably superior Linux desktop of Dr Adrian Midgley
> http://www.defoam.net/
So you think you can _prove_ that? I (and undoubtedly you) know it for
certain, but I think this is an apt case of (a) and (b) above!
Best wishes,
Ted (sitting at Red Hat 9, typing into SuSE 7.2 a message which will be
"From:" FreeBSD).
--------------------------------------------------------------------
E-Mail: (Ted Harding) <[log in to unmask]>
Fax-to-email: +44 (0)870 167 1972
Date: 20-Sep-03 Time: 13:20:39
------------------------------ XFMail ------------------------------
|