Prev: Incontinence Surgery...
Next: No more Fosamax
From: Califchief on 15 Aug 2008 00:48
>> http://www.nytimes.com/2008/08/12/health/12well.html

> Pure nonsense! Lead-time bias can affect 5-year relative survival
> statistics. It has no effect on the mortality rate. The truth is that
> there has been a 37.4% reduction in CaP mortality (1991 - 2005) since
> the inception of PSA testing.


If the number of men diagnosed with PCa doubled in that 15-year time
frame because of PSA testing, and the number of deaths remained the
same, there would be a 50% reduction in the mortality rate.


___ Blue Wave/QWK v2.12
From: len on 15 Aug 2008 11:40
On Aug 14, 11:48 pm, califch...(a)fidotel.com (Califchief) wrote:
> >>http://www.nytimes.com/2008/08/12/health/12well.html
>
> > Pure nonsense! Lead-time bias can affect 5-year relative survival
> > statistics. It has no effect on the mortality rate. The truth is that
> > there has been a 37.4% reduction in CaP mortality (1991 - 2005) since
> > the inception of PSA testing.
>
> If the number of men diagnosed with PCa doubled in that 15-year time
> frame because of PSA testing, and the number of deaths remained the
> same, there would be a 50% reduction in the mortality rate.
>
> ___ Blue Wave/QWK v2.12

That is true. But it is a bit more complicated than that. First
of all, the number of deaths in any given year depends not on the
number of men diagnosed that year, but those diagnosed something like
5 to 15 years earlier. Second, because of PSA testing, men tend to be
diagnosed earlier. That means you have to wait even longer from time
of diagnosis to death.

It is true that initially the number of cases diagnosed increased
significantly, but it didn't do anything like doubling. Also, after
several years, it started decreasing. the upshot of all this is that
one would need a fairly sophisticated model to decide if there apears
to have been an effect.

There is one fact which can't be denied. Not only did the prostate
cancer death rate decline during this period, but the absolute number
of deaths declined. No fiddling with how you count can explain
that. The usual explanation by critics of testing is that treatment
improved, but I don't think anyone has given any proof that such is
the case. It may be the explanation or it may not.

There is one ongoing prospective randomized study which is attempting
to determine whether testing has an effect either on prostate cancer
mortality or overall mortality. Unfortunately, for any disease with
a very long time horizon, such a study may be of limited usefulness,
and this one is not without problems. You can never truly isolate
one factor. A true prospective randomized test would make sure that
in the group undergoing testing the treatment following diagnosis was
the best available. But in the current trial there is no quality
control of treatment. The argument usually made is that it is best to
just allow the treatment common in the medical community because that
is what patients on the average will get. But medical practice also
changes, and it may change significantly over the course of a study
with a long time horizon. In the case of prostate cancer, obvious
examples is the vast increase in the number of surgeons trained in
Walsh's never sparing technique. That is bound to affect the
prevalence of impotence following surgery. Another example is the
improvement in radiation treatment which allow much larger does to
kill the cancer without damaging surrounding tissues.

It seems to me that the focus is wrong. We will probably never
establish beyond doubt that testing overall is beneficial or not.
More important, it is clear that testing is beneficial for some men.
I am an example. I was diagnosed at age 67 with a Gleason 7=3+4
case. That typically happens in 15 - 20 percent of diagnoses. Such
men stand a fair chance of developing clinical metastatic prostate
cancer within 10 years. Such an outcome would be far from certain---
the odds might not even exceed 50-50---but the risk is certainly high
enough that most rational men wouldn't want to face it. Suppose
testing were banned. That means a substantial number of men would be
sacrificed in order to avoid serious side effects for men whose
prostate cancer need not have been treated. Suppose that number were
even as high as 40 percent and the number of men who were sacrificed
was only 10 percent. Would that be a reasonable trade-off?

I think it is foolish to engage in a moral calculus of that nature.
It makes more sense to make testing available and, in the case of a
diagnosis of prostate cancer, make it very clear to men what the
potential risks and benefits of treatment are, along with all the
undertainties about what is known about that. In the end we are all
responsible for the decisions we make, but we should make our own
decisions, not havie them forced on us by public health
epidemiologists looking at overall effects.
 | 
Pages: 1
Prev: Incontinence Surgery...
Next: No more Fosamax