Re: Fwd: Re: Radon and lung cancer

[Al Tschaeche <antatnsu@pacbell.net>]

David Scherer wrote:
The case control studies will be done.  Let's see whether they
> confirm Cohen's work.  What's the hurry?  The universe isn't going anywhere.

Well, that could be argued.  However, billions are being spent for no
measurable health benefit.  Therefore, there is a hurry.
> 
> As far as the PC thing goes, there has always been a difference between
> science and popular-science.  Public forums are not doing science, but
> public policy.  Hopefully scientists take their time and hold their
> conclusions as tentative.  

But the politicians want scientific answers.  Please tell me how the
scientists can tell the politicians to wait.

On the other hand, this list looks for a
> definitive answer before the year is out.  My concern is that we are
> becoming part of the rabble instead of maintaining some degree of
> objectivity.  The question of how radiation affects health is a SCIENTIFIC
> question.  The LNT model is a tentative model used for SOCIETAL purposes
> (i.e. radiation protection poicy).  There may be time pressure to address
> the latter, but not for the prior.

Very, very true.  So, how do we do it?

> 
> >>Hve we not lapsed into a consideration of semantics, however, when we say
> >>that "burden of proof" is significantly different from "weight of the
> >>evidence"?

These are two very different concepts.  The burden of proof should be on
those who say a low dose of radiation is harmful.  Instead, the burden
of proof is being placed on those who say there is a threshold and low
doses are safe.  The latter is absolutely wrong!  The NCRP, ICRP, EPA,
etc. have corrupted the American idea that one is innocent until proven
guilty.  The problem is, there is little chance of science saving the
day.  No one will spend the money to do the experiment to demonstrate
low doses are safe and/or hormetic to humans.

Weight of evidence is quite different.  It is a legal term and has no
place in science.  Science is science.  If an hypothesis is demonstrated
to be wrong by only a single piece of data, then it is wrong and another
must be presented.  One does not need the weight of evidence to call a
hypothesis wrong.  So far the LNTH has been demonstrated wrong by much
data (e.g. Evans for radium and Cohen for radon).  Therefore, the LNTH
should be discarded and another hypothesis presented.  I vote for a
threshold. So far there are no data that demonstrate 5 rem per year is
harmful.  So, my hypothesis would be: 5 rem per year is safe.  When, and
if, data are presented that demonstrate that hypothesis is false, I
would have to devise another one.
> 
> "Burden of proof" suggests that one explanation is "accepted" or at least
> preferred.  

And the explanation may be absolutely wrong.  Take the OJ Simpson
trial.  The burden of proof was on the prosecutors.  They could not
present the burden of proof that the jurors would accept. But, if OJ
didn't do it, who did?  And why has OJ not spent his fortune to find
that person(s) as he said he would?  So, I say the explanation in that
case is absolutely wrong.

Others must overcome that preferrence.  In a weighing process
> one takes all the evidence at the same time, not in a sequence were later
> results must overcome the preferrence of earlier.  One not only looks at
> the conclusions of each study, but the power of the methods and the
> strenght of the data.  For example, I have seen papers that take
> measurements of the speed of light (c) from the Renaissance until today and
> suggest that it may be changing.  When I look at the methods used 500 y
> age, I give little weight to the conclusion that c is changing.  If c is
> changing, we will know this from several centuries of modern, precise
> measurements.  No need to form a conclusion now based on the "proof"
> offered.  I see a similar situation with regard to radon and other
> radiation bioeffects.  

I don't.  There are data that demonstrate non-linearity at low doses
(the hormetic data for some).  There is much data that demonstrates zero
effects (much of it because the statistical power of the experiment is
not very powerful).  I agree that, if we had time (the exact value of
the speed of light is not very important in everyday life (except for
some things nowadays).  There is time to measure and measure and
measure.  But the radiation effects question is costing billions.  There
is not time now to wait.

Al Tschaeche antatnsu@pacbell.net