[ RadSafe ] Cohen's radon results and LNT

Fri Oct 7 13:18:14 CDT 2011

Hi, Mike Brennan,
Excuse my belated response, but after a lull there has been a kind of solar
flare on Radsafe that reached up to existential problems. I hope we can resume
at ground level.

I do not feel obliged to comment further on Field et al's radon paper, except
perhaps to point out that individual dosimetry with respect to radon is
basically as good as impossible, as unambiguously pointed out by dr Field in
his 1999 booklet "Radon occurrence and health risk". Working levels are
probably more for regulatory purposes; the fairly crude assumptions are
basically relevant to mining conditions. The same radon concentration can
result in completely different actual lung doses, depending on a range of
external factors and even level of activity and physiological factors of the
subject. So in the end one is back at having to sample sufficient numbers so
that the factors sort of average out. It is humanly impossible to control for
all of them. How far can and does one need to go if you insist on being
'properly scientific'?

This is of course the attraction of Cohen's study. It was a genial insight that
LNT could be tested as a null hypothesis: one only needed to determine a
collective dose or a quantity proportional thereto per county, e.g. average
domestic radon concentration.  The other required statistics were readily
available. 

In the planning phase Cohen could not have foreseen the extent of the problem
of an unfortunate (anti-)correlation between smoking prevalence and average
county radon levels, nor the strength of the correlation. Ideally there should
have been no correlation or only a very weak one. In that case the smoking
correction would have been almost immaterial and not critical. My conjecture is
that Cohen cannot illustrate the veracity of his smoking treatment, compromising
his claim against LNT. If the smoking correlation with radon had been positive,
the data would have appreared to support LNT!

I believe Cohen had to vary the smoking parameters quite drastically for his
data to resemble LNT, whereas a comparatively moderate variation would have
produced the zero gradient, which I believe is correct.  I back up my
supposition by investigating Cohen's LC data in terms of elevation and county
population size.  
I believe that my presentation of Cohen's results is sufficient to contradict
LNT for domestic radon. A plausible smoking correction would be nice to have,
but is not essential, as it is in Cohen's approach. 

I am still waiting for somebody to fault my argument that Cohen's data in my
presentation support lung cancer independence from average domestic radon,
thereby contradicting LNT.

Regards
chris.hofmeyr at webmail.co.za

On Mon, 3 Oct 2011 10:24:33 -0700 "Brennan, Mike  (DOH)"
<Mike.Brennan at DOH.WA.GOV> wrote

> Hi, Chris.
> 
> I cannot speak for Dr. Field (I know he sometimes monitors RadSafe; perhaps
> he will chime in).  One of the aspects I believe I heard explained was that
> women were chosen because this cohort consisted mostly (entirely?) of women
> who were "stay-at-home-moms", which removed possible workplace exposure as a
> confounder.
> 
> I do recognize that lung cancer rates among women were going up, as smoking
> rates among women had gone up in the decades before.
> 
> I agree there are a lot of assumptions, but that is the way of such studies. 
> One could, I suppose, devise an experiment involving large numbers of
> individuals kept in known radon concentrations, but I suspect the costs would
> be prohibitive, even if the crimes-against-humanity provisions didn't kick
> in.
> 
> -----Original Message-----
> From: Chris Hofmeyr [mailto:chris.hofmeyr at webmail.co.za]
> Sent: Saturday, October 01, 2011 12:47 AM
> To: Brennan, Mike (DOH)
> Cc: The International Radiation Protection (Health Physics) MailingList
> Subject: Re: [ RadSafe ] Cohen's radon results and LNT
> 
> Mike,
> I am not enough of a statistician to give a considered opinion. And I have
> only
> looked at this paper superficially.
> 
> One can see that Field et al have tried to control for numerous possible
> confounders, so the intention was good, but I cannot understand that they end
> up with controls that have a vastly smaller smoking history and also much
> better previous lung health. I am very sceptical that such factors can be
> corrected reliably, despite what statisticians may claim. Surely a closer
> match
> should have been possible?
> 
> Female lung cancer rates were possibly unsettled at the time of the survey.
> There was a very significant increase between the 1970s and the 1980/90s,
> US-wide. Iowa??
> Then the relatively small number of subjects......Why females? One probably
> hoped to get a better dosimetry, but it is still full of assumptions. The
> female lung cancer rate is much lower than the male rate, which happened to
> vary little in the said periods. If radon is suspected as a multiplicative
> modulator of LC, the effect on males would be larger.
> Call me a sceptic.
> 
> chris.hofmeyr at webmail.co.za
> 
> 
> 
> On Fri, 30 Sep 2011 10:43:40 -0700 "Brennan, Mike  (DOH)"
> <Mike.Brennan at DOH.WA.GOV> wrote
> 
> > Hi, Chris.
> >
> > I would be interested on your take on the Iowa Radon Lung Cancer Study,
> > which to me seems to have better power because it controlled for more
> > confounding factors.  http://radsci1.home.mchsi.com/irlcs.pdf
> >
> >
> > -----Original Message-----
> > From: radsafe-bounces at health.phys.iit.edu
> > [mailto:radsafe-bounces at health.phys.iit.edu] On Behalf Of Chris Hofmeyr
> > Sent: Thursday, September 29, 2011 9:03 AM
> > To: Otto G. Raabe; Bernard L. Cohen
> > Cc: radsafe at agni.phys.iit.edu; WesVanPelt at verizon.net
> > Subject: Re: [ RadSafe ] Cohen's radon results and LNT
> >
> > Prof Raabe,
> > Your most recent Radsafe posting is appreciated as implying that my
> > understanding of Cohen's radon data does not conflict with your
> > proposals
> > concerning cancer induction by protracted exposures, as explained in
> > your
> > well-considered 2011 paper.
> >
> > I am thinking aloud about the alpha equal 2 beta effect. It is fairly
> > trivial
> > to calculate that a whole-body short-term lethal dose of a couple of
> > Gray
> > actually only affects one atom in a million. Of course, a DNA strand
> > contains
> > millions of atoms, but a double strand break cannot happen very far
> > separated,
> > or it would be classed as two single strand breaks.  The alpha would
> > tend to
> > break both strands close together, but how is that envisaged for two
> > (uncorrelated?) betas, as the beta-active nuclei are presumably
> > spatially well
> > separated. Consecutive beta-decays are usually well separated in time,
> > but a
> > beta is often almost instantaneously followed by a gamma.
> > Some clarification would be appreciated.
> >
> > Sincerely
> > chris.hofmeyr at webmail.co.za
> >
> >
> > On Wed, 28 Sep 2011 12:00:14 -0700 "Otto G. Raabe" <ograabe at ucdavis.edu>
> > wrote
> >
> > > At 09:40 AM 9/28/2011, Chris Hofmeyr wrote:
> > > >Prof Cohen, Radsafers,
> > > >In the condensed notes on Cohen's radon studies (attached to posting
> > > >on Radsafe
> > > >21 Sept), I attempt to show unequivocally, although slightly
> > > unconventionally,
> > > >that Cohen's lung cancer (LC) mortality rates are basically
> > independent of
> > > the
> > > >average domestic radon concentrations as determined by Cohen on a per
> > county
> > > >basis. This independence clearly means that LNT is REJECTED, since
> > lung
> > > cancer
> > > >is not perceptibly correlated with domestic radon...
> > > *****************
> > > My Forum paper,  "Toward Improved Ionizing Radiation Safety
> > > Standards", Health Phys. 101:84-93;2011, explains why LNT does not
> > > apply to any protracted or repeated radiation exposures.
> > >
> > > Otto
> > >
> > >
> > > **********************************************
> > > Prof. Otto G. Raabe, Ph.D., CHP
> > > Center for Health & the Environment
> > > University of California
> > > One Shields Avenue
> > > Davis, CA 95616
> > > E-Mail: ograabe at ucdavis.edu
> > > Phone: (530) 752-7754   FAX: (530) 758-6140
> > > ***********************************************
> > > _______________________________________________
> > > You are currently subscribed to the RadSafe mailing list
> > >
> > > Before posting a message to RadSafe be sure to have read and
> > understood the
> > > RadSafe rules. These can be found at:
> > > http://health.phys.iit.edu/radsaferules.html
> > >
> > > For information on how to subscribe or unsubscribe and other settings
> > visit:
> > > http://health.phys.iit.edu
> >
> >
> >
> > ____________________________________________________________
> > South Africas premier free email service - www.webmail.co.za
> >
> > Save on insurance with OUTsurance
> > https://www.outsurance.co.za/insurance-quote/?source=webmailmailer&cr=wu
> > y11_468x60&cid=241
> >
> >
> > _______________________________________________
> > You are currently subscribed to the RadSafe mailing list
> >
> > Before posting a message to RadSafe be sure to have read and understood
> > the RadSafe rules. These can be found at:
> > http://health.phys.iit.edu/radsaferules.html
> >
> > For information on how to subscribe or unsubscribe and other settings
> > visit: http://health.phys.iit.edu
> > _______________________________________________
> > You are currently subscribed to the RadSafe mailing list
> >
> > Before posting a message to RadSafe be sure to have read and understood the
> > RadSafe rules. These can be found at:
> > http://health.phys.iit.edu/radsaferules.html
> >
> > For information on how to subscribe or unsubscribe and other settings
> > visit:
> > http://health.phys.iit.edu
> 
> 
> 
> ____________________________________________________________
> South Africas premier free email service - www.webmail.co.za
> 
> For super low premiums, click here. http://www.dialdirect.co.za/?vdn=15828
> 
> 
> 

____________________________________________________________
South Africas premier free email service - www.webmail.co.za 

For super low premiums, click here. http://www.dialdirect.co.za/?vdn=15828