agree completely!
many researchers and 'researchers' respectively in medicine, esp. in
epidemiology lack the ability for logical reasoning...
media? they live from that kind of garbage :-(
brdp
the researcher

It's not as simple as that.  They usually compare their findings with the
average population, or better yet, a 2nd group of people, which they also
monitor to compare to.  Plus, the bigger the number of people in the group,
the more likely the results are true.  Of course, some studies are designed
better than others, so repeatable of a study is necessary to confirm the
original study, and this is often done.  Sometimes it turns out that it's not
repeatable.  

Also, you have to read the actual article to determine just what was involved
in making the conclusion.  The summaries you see in the media don't reflect
the true design of the study, and the statistics involved in making the
conclusion.  The media often simplifies (and sometimes distort) the
conclusions.

In a previous article, faraz_hussain@yahoo.com (Duke of Hazard) wrote:
->As an engineer I do not understand why journals and the media publish
->health studies that do not demonstrate cause and effect. The typical
->study goes something like this:
->
->Researchers found that those who:
->
->------ exercised regularly
->--or--
->------ ate fruits and vegetables
->
->throughout life were less likely to have contracted xyz disease than
->those who did not.
->
->One can not make such type of conclusions without first proving there
->is a relationship between the variables. The conclusions are as absurd
->as saying "people who eat chocolate are less likely to contract xyz
->disease". While the statement may be true for the sample population ,
->it does not establish there is any relationship between chocolate and
->xyz disease.
->
->So my question is why do respectable medical journals and the media
->continue to publish these types of pseudo-science studies?

As an engineer, you probably encounter people who know nothing
about engineering and yet think that they have simple solutions
to problems you know are hard. Fields that have existed for a long
time typically have a lot of expertise that may or may not come
through in media reports. Yet making sneering comments about fields
you do not really understand seems to be a common pastime of
many groups.

Medical researchers know perfectly well that (as people in the
biz quote constantly) "association does not prove causation".
They do all sorts of clever things to try to adjust for this, some
of which are more believable than others. However, when dealing
with humans, you often cannot run experiments to answer questions.
How, exactly, would you suggest doing an experiment to figure
out whether exercising for twenty years reduces the risk of heart
disease?

Furthermore, you likely completely believe (and rightly so) some
things that have only been proven because of associations. No
one has ever done an experiment in humans that shows that smoking
causes lung cancer. Yet the evidence from observational studies
of the sort that you are suggesting are not worthy of publication
overwhelmingly demonstrates causation.

You picked an interesting example in particular though with exercise.
There are many reasons to worry that in particular with exercise
causation may run in the other direction. Healthier people are
able to exercise more, and so high levels of exercise may be a marker
for health rather than a cause of health. This is extremely hard
to tease out of observational studies.

David Rind addressed your question much better than I could, but I have to
point out that what the media publishes is almost always hyped up,
misinterpreted, grandiose summaries of otherwise reasonable research.  Next
time you see medical research reported in the media, go read the actual
journal articles and I think you'll almost always find valid science with
appropriately restrained conclusions.

I haven't read any of these medical studies myself, and as others have
pointed out, media coverage is necessarily simplified and condensed.
That said, it does seem clear to me that it's worthwhile to pursue
studies which suggest associations.  

First, let's keep our logic straight:  Just because I do not have
proof of a cause and effect relationship between two events -- that
is, just because I don't have precise, detailed knowledge of the
molecular chain of events and energy transfers that may link cause A
to result B -- does not mean that causation does not exist.  

Second, association gives us ground to suspect that a causal relation
does exist, and so serves as a guideline to suggest further research.

Third, the reality is that biochemical systems are very complex.  Our
present state of science, for all its vast sophistication, is still
dwarfed by the complexity of living systems.  It may be that, at our
present point in history, the *possibility* of causality -- inferred
based on association -- may be the best level of knowledge we are able
to achieve.  I'd rather have partial, tentative, preliminary
indications of plausible truths, rather than having no guidelines at
all.

If anything, my gripe with medical science is the opposite.  Too,
often, medicine refuses to consider a possible causal chain unless
it's been validated by formal studies.  The result is that noteworthy
anecdotal evidence is often dismissed.  I've been having high blood
pressure problems.  My doctor put me on a medicine for it, and I
noticed my pulse shot way up.  I checked the Web, and found anecdotal
evidence (accounts from people in newsgroups) that this medicine
occasionally, if rarely, does cause rapid pulse.  But since the side
effect is apparently rare, it's not noted in the literature for the
drug.  Given that we each have a unique body chemistry, it should
certainly be considered that a drug might not show a side effect in a
controlled study of 1000 people, but might show a side effect when
used in a general population of 100,000 people.  (And of course,
occasionally approved drugs are withdrawn when dangerous side effects
are discovered in general use.)  

Anyway, the point is, sometimes knowledge is less than perfect, but
still constitutes useful knowledge all the same.

Steve O.

Standard Antiflame Disclaimer:  Please don't flame me.  I may actually *be* an idiot, but even idiots have feelings.