Re: "Screening" tests
From: Leonard Evens (len_at_math.northwestern.edu)
Date: 07/04/04
- Next message: George Conklin: "Re: "Screening" tests"
- Previous message: harmony: "Re: TOP 20 FRUITS AND VEGGIES TO BEAT DISEASES"
- In reply to: dale.j. : "Re: "Screening" tests"
- Next in thread: Robert A. Fink, M. D.: "Re: "Screening" tests"
- Messages sorted by: [ date ] [ thread ]
Date: Sun, 04 Jul 2004 10:21:16 -0500
dale.j. wrote:
> In article <YrWdnU-NEr6dRnvdRVn-sA@comcast.com>,
> Leonard Evens <len@math.northwestern.edu> wrote:
>
>
>
>>Again it depends. If there is no known effective treatment, knowing
>>about it early rather than late may not be helpful.
>>
>>For prostate cancer I think it is generally helpful. You don't
>>necessarily have to treat prostate cancer just because you know you have
>>it. It depends on the details of the diagnosis, the age of the patient
>>and some other factors.
>
>
>
> What cancer do we not have an effective treatment for if found early
> enough? I'm curious because I thought that if cancer were found early
> the treatments would be effective and I was also under the impression
> there are treatments for all types of cancer.
There are certain cancers which have very low survival rates. Dr. Fink
mentioned some brain tumors, and he would know more about that than I
do. If you want to scare yourself you can go to the American Cancer
Society web site and find other examples.
>
> Perhaps we are still not finding the cancer early enough. Example, if
> we somehow could find that first abnormal cell and eliminate it wouldent
> that be a cure? Most likely not practical, but just a thought of mine.
A problem with early detection is that you may detect cancers which will
never amount to anything. So you may end up treating perfectly healthy
people needlessly and in some cases there will be undesirable side
effects. This is a subject of intense study and some debate in the
medical research community, and I don't believe there are going to be
easy answers. Ideally, if you could actually find those abnormal cells
long before they can do damage, you could be relatively certain they
would not be harmless, and you could excise them with little risk of
side effects, it would be worth doing. But in practice that is hard to do.
Take prostate cancer as an example. It is known from autopsy studies
that a large number of men who die of other causes show microscopic
evidence of prostate cancer. The estimates vary by age and they also
vary quite a bit in different autopsy studies. But it seems to be at
least 50 percent in men over 80, and perhaps a lot higher. It would be
lower in younger men but still fairly high. Remember that only about 16
percent of men in the US are likely to be diagnosed with clinical cases
of prostate cancer during their lifetimes. So there is a difference
between the microscopic kinds of cancers found on autopsy and those
actually diagnosed through current techniques. If you do PSA testing,
you will catch some cases that don't need treatment (and you will miss
some that do, but that is another story). No one really knows how many,
but again it varies by age. Few experts would recommend testing and
aggressive treatment in men over 80, but most urologists, at least,
generally feel there is no question that men under 70 should be tested
and treated if necessary. For men 70-80, it depends on a variety of
factors, including life expectancy. The available diagnostic tools,
like PSA and Gleason scores are helpful, but they are not entirely
reliable. Current research using modern techniques such as DNA analysis
are beginning to find markers which can distinguish the cases needing
treatment from those that don't, but they aren't yet available for
general use. I suspect that in a few years, physicians will be able
through appropriate testing to distinguish those cases which do need
aggressive treatment from those that don't, and at that point the debate
will become moot. But we are not there yet.
Biostatisticians talk about risk-benefit analysis. That is the benefit
of treatment has to be balanced against the risks of treatment. But as
I keep pointing out that analysis really has to be done separately for
each patient, and that will depend on the specifics of the case. But
what we have to help make the decision is statistics, much of it of
questionable validity, about large populations. It is hard to translate
those statistics to odds as they apply in individual cases. I've taught
a lot of premeds in my day and I know quite a few physicians. Few of
them really have the background in advanced statistics to be able to
make decisions of this kind based on formal theory. That is probably a
good thing because it is unlikely that going by the numbers would
actually take into account all the relevant facts. So doing it on the
basis of experience and feeling about a given patient is probably a
better way to go.
>
> Dale J.
>
- Next message: George Conklin: "Re: "Screening" tests"
- Previous message: harmony: "Re: TOP 20 FRUITS AND VEGGIES TO BEAT DISEASES"
- In reply to: dale.j. : "Re: "Screening" tests"
- Next in thread: Robert A. Fink, M. D.: "Re: "Screening" tests"
- Messages sorted by: [ date ] [ thread ]
Relevant Pages
|
