Re: Exercise and PCa

[Leonard Evens]

ron wrote:
> Most recent studies on the effect of exercise upon PCa have concluded,
> as does this study, that there is no observable connection between
> exercise and overall PCa risk. A few previous studies suggested that
> there might be a link between exercise and advanced PCa. This
> prospective, 17 year study finds that "recreational physical exercise
> is associated with reduced risk of advanced prostate cancer and
> prostate cancer death."...ron
>
>
> Int J Cancer. 2006 Oct 3
>
> Recreational physical activity and risk of prostate cancer: A
> prospective population-based study in Norway (the HUNT study).
>
> Nilsen TI, Romundstad PR, Vatten LJ.
>
> Department of Public Health and General Practice, Faculty of Medicine,
> Norwegian University of Science and Technology, NO-7489 Trondheim,
> Norway.
>
> Physical activity has been studied in relation to prostate cancer risk,
> but the findings have been inconclusive. We prospectively examined the
> association between self-reported recreational physical activity and
> overall risk of prostate cancer, risk of advanced disease and risk of
> prostate cancer death in a cohort of 29,110 Norwegian men.

I'm not sure why this is called a prospective study. They didn't
randomly divide men into exercisers and non exercisers and then follow
them to see what happened. So this seems to me to be basically a
retrospective study, even if the men were identified first and followed.
It wouldn't have made any difference had they instead gone back and
picked men out by the same criterion and then seen what happened to them.

For example, the usual objection arises. How do we know that some
other factor both makes men more resistant to prostate cancer and makes
them more interewted in exercising? It is like the old theological
dispute about predestination. Is Godly behavior a choice that makes it
more likely you will go to Heaven or is it just a sign that you were
predestined to be Godly and go to Heaven in the first place?

On the other hand, some carefully done retrospective studies do add some
evidence to the mix, and I don't think they should be rejected out of
hand, as some statisticians would have it. This may be one such. I am
also willing to believe that men who exercise regularly will get through
treatment better and suffer fewer side effects.

> Incident
> prostate cancers were obtained from the Norwegian Cancer Registry, and
> prostate cancer deaths were obtained from the national Cause of Death
> Registry. During 17 years of follow-up, 957 incident cases were
> identified, 266 of them were advanced (i.e. metastases at diagnosis)
> and 354 of the incident cases died from prostate cancer. In
> multivariable analysis, frequency and duration of exercise were
> inversely associated with the risk of advanced prostate cancer (p for
> trend = 0.04 and 0.02). We computed a summary score that combined
> frequency, duration and intensity of exercise, and this score showed
> inverse associations with advanced prostate cancer risk and mortality
> (p for trend = 0.02 and 0.07). Compared to men who reported no
> activity, the relative risks (95% confidence intervals) among men in
> the highest category of physical exercise was 0.64 (0.43-0.95) for
> advanced prostate cancer and 0.67 (0.48-0.94) for prostate cancer
> death. We found no association between physical activity and overall
> risk of prostate cancer. We conclude that recreational physical
> exercise is associated with reduced risk of advanced prostate cancer
> and prostate cancer death. (c) 2006 Wiley-Liss, Inc.

Notice the careful wording. Being associated doesn't necessarily imply
a causal link.

>
> PMID: 17019717
>

[ron]

Leonard Evens wrote...snip...
> I'm not sure why this is called a prospective study. They didn't
> randomly divide men into exercisers and non exercisers and then follow
> them to see what happened.

Leonard...Wouldn't what you describe above be a prospective, randomized
study? If these 29+ thousand men were identified 17 years ago (say
they were all the men in a county or some similar grouping) and their
exercise habits monitored over time through inquiry, and their PCa
incidence / outcomes monitored through a tumor registry, wouldn't that
be a non-randomized prospective trial?..ron

> So this seems to me to be basically a
> retrospective study, even if the men were identified first and followed.
> It wouldn't have made any difference had they instead gone back and
> picked men out by the same criterion and then seen what happened to them.

[Leonard Evens]

ron wrote:
> Leonard Evens wrote...snip...
>
>>I'm not sure why this is called a prospective study. They didn't
>>randomly divide men into exercisers and non exercisers and then follow
>>them to see what happened.
>
>
> Leonard...Wouldn't what you describe above be a prospective, randomized
> study? If these 29+ thousand men were identified 17 years ago (say
> they were all the men in a county or some similar grouping) and their
> exercise habits monitored over time through inquiry, and their PCa
> incidence / outcomes monitored through a tumor registry, wouldn't that
> be a non-randomized prospective trial?..ron

This is coming down to a matter of terminology. It is the randomization
that is important from a statistical point of view. The point is to be
as sure as you can be that there aren't any signficant confounders that
can bias the results.

>
>
>>So this seems to me to be basically a
>>retrospective study, even if the men were identified first and followed.
>> It wouldn't have made any difference had they instead gone back and
>>picked men out by the same criterion and then seen what happened to them.
>
>