Re: Re[2]: Elephants vs mice

["Dale E. Boyce" <dale@radpro.uchicago.edu>]

Boy I hate to wade in here (no pun intended Wade), but I can't resist the fun.

Deriving a linear response based on a purely probabilistic model is
fallout of the assumption that each step of even a multistep process
is based on a single intiation in one cell.  Any functional form you
pick will be by necessity linear as it goes to zero _BUT_ the slope
at low dose/dose rate does not necessarily need to be the same as at
high dose high dose rate.  This applies even if you throw in a repair
process.  There would be some adverse effect expected at low doses from
such models (although probably experimentally unobservable).

However, such models do not take into consideration any possible
intercellular interactions.  I'll also use Dave's caveat that I am
not a biologist, but everyone in this business should do a little reading
about intercellular communication.  I'll try to dig up a reference
I believe there was a good introduction in Sci. Am. within the last
year.

Cells help regulate each other.  Thats why humans grow to one size
and mice and elephants to other sizes.  We're usually comparable to
our own species in size.

The point is that one damaged cell among many healthy cells might
not be able to become cancerous.  It might even be killed off by
immune response.  However a cluster of damaged cells might be
able to promote to a malignant growth.  Granted I am way out of
field here, but trying to point out just one way that the basis
of LNT may break down.

We know that cells communicate to regulate growth.  Cancer is a growth
process.  We totally ignore growth regulation possibilities in
the models that are the basis of LNT.

I do not know if LNT is true or false.  (I know which ism I lean towards)

I have no problem in using LNT for setting protection standards.
 (as long as we don't keep ratching beyond the domain of the model)

I don't have a problem with using LNT to establish probable cause 
 (if it is used correctly).

But I do have a problem in using LNT to estimate how many people we are
killing with small doses.

The basic point of fact that we need to keep our eye on is that when
everything is added up, the use of radiation extends the average life
expectancy by improving medical care, reducing pollution, and improved
engineering safety.  Over-regulating it out of business and scaring the
public of being exposed to any radiation will cause negative health
effects.

The use of radiation that probably causes the greatest loss of life
is in its use in density gages in cigarette manufacturing. ;)

Enough of my rantings for one afternoon.

Dale Boyce
dale@radpro.uchicago.edu