Re: Correlation: In Memory of My Former Academic Colleagues
- From: G Robin Edwards <robin.edwards@xxxxxxxxxxxxx>
- Date: Sun, 19 Jun 2005 22:13:41 +0100
In article <1119049309.580884.27530@xxxxxxxxxxxxxxxxxxxxxxxxxxxx>,
Reef Fish <Large_Nassau_Grouper@xxxxxxxxx> wrote:
> Jerry Dallal wrote:
> > Reef Fish wrote:
> > >
> > >
> > > Half a century later, there STILL has not been a single valid
> > > controlled experiment on that issue -- on HUMANS. :-) And the
> > > evidence remain slim to highly questionable, notwithstanding
> > > all the warning labels the Surgeon General have put on
> > > cigarette boxes.
> >
> > While there may be things to criticize
> > http://tinyurl.com/cvcxd gives one pause.
> > How does one proceed when controlled experiments are unethical?
> > I agree that it is often worse to act on bad data than not act at all.
> Welcome back, Jerry.
> Excellent question, and excellent comment.
> My answer would be, if you CAN'T do a control experiment on the
> subjects, then DON'T draw strong conclusions from them AS IF
> it had been done.
I am interested in the Controlled Experiment discussion for a rather
practical and unfortunate reason.
Those of you living in the USA will be unaware of a very large clinical
"experiment" that is currently under way in the UK. It is a trial of
five drugs that are believed to have potential in retarding and
alleviating some of the unpleasant (to say the least) symptoms of
multiple sclerosis. My daughter is a sufferer, having been diagnosed
about 10 years ago, though in hindsight some typical symptoms occurred
at least ten years prior to this, when she was about 20.
Under very great pressure the government agreed with the drug
manufacturers to set up a "trial", on a shared risk basis. If the
drugs were shown to be effective (how this was to be done seems not to
have been made clear as far as I know) the government would pay for the
treatment. If not, the drugs suppliers would pay. Now this is no
trivial matter. Somewhere around 10,000 people were expected to be
enrolled, although it may be that only half of these were finally
accepted for assorted clinical and personal reasons. The annual cost
(that is, what a patient might have to pay for treatment) is around GBP
10,000 (ten thousand pounds) per year.
This trial began in April 2004, IIRC. The participants were allowed to
choose freely which drug they would test, and the choices were based on
personal reasons. For example, my daughter chose Interferon B, which
requires self injection on alternate days, because some others required
more rigorous treatment schedules.
As far as I know there are no "controls" - people who are getting a
placebo treatment. The assessment of patients' conditions seems to be
done on a haphazard basis, with no quantitative aspects except seeing
how well or badly they can cope with a 50 metre walk along hospital
corridors, done about every three months. They do not have to keep
diaries or notes of their daily condition as seen by themselves. In
fact now that the trial is running smoothly from the logistic
standpoint there seems to be little interest in any quantitative aspect
of my daughter's symptoms. Her neurologist, whom she sees once a year
only, tells her that as far as he is concerned over the past three
years her condition is stable. From the practical aspect of being able
to manage her life she has clearly deteriorated sadly.
It is a sorry tale, in my opinion, and looks to me as if it will be a
total waste of public money, since I fail to see how the good effects
(if any) of the drug can ever be ascertained with the trial "organised"
in this manner.
Would anyone care to comment, or to suggest just how our government and
the drug companies will try to reach an understanding on the relative
effectiveness of the five drugs?
If I were one of the suppliers I would insist on proper and agreed
evaluation of the patients' condition at regular intervals. Note that
this trial is expected to last several years, and is very expensive
indeed.
I anticipate enormous legal bills for one or both parties in due
course, so someone is sure to benefit!
Robin
.
- Follow-Ups:
- Re: Correlation: In Memory of My Former Academic Colleagues
- From: Jerry Dallal
- Re: Correlation: In Memory of My Former Academic Colleagues
- Prev by Date: Binary Clustering Help
- Next by Date: Re: My Editorial on Science and Statistics
- Previous by thread: Binary Clustering Help
- Next by thread: Re: Correlation: In Memory of My Former Academic Colleagues
- Index(es):
Relevant Pages
|