[ RadSafe ] Re: Radon and Lung Cancer

Wed Feb 14 12:30:52 CST 2007

Dear Wesley, and Professors Cohen and Raabe,

This was written in response to Wes Van Pelt's reminder about his
elevation paper (below, after prof Cohen's response).

Firstly, allow me to thank both Prof B L Cohen and Wes Van Pelt sincerely
for making their data available.

I thought Wesley’s oxygen proposition sounded interesting, but my problem
is that if one examines Cohen’s lung cancer mortality data in 1601 US
counties as a function of Wesley’s altitude data in more detail, the major
reduction in lung cancer (roughly ~30%) takes place below the average
county elevation e0 of 1259 feet.  It seems difficult to reconcile such a
large cancer reduction over a range in which the oxygen concentration is
reduced by only about 3.5%.  If I am not mistaken, the altitude
stratification used by Wesley was in ten groups of 160 counties, of which
the first eight happened to lie below the average elevation.

Prof Cohen at some stage posed a challenge to propose an independent
variable showing a similar lung cancer dependence of his radon data. 
Wesley came very close, but has not made the claim in so many words.  It
therefore remains for yours truly to state unequivocably that Cohen’s lung
cancer mortality rate data (for both male and female and for both
observational periods 1970-79 and 1979-94, i.e. all 4 data sets) when
plotted against the rank of county elevation, is virtually identical to
the corresponding plot of LC against the rank of average radon
concentration.  Rank is simply extreme stratification into 1601 counties,
i.e. sorted from 1 to 1601 according to the size of the independent
variable, be it county elevation or average radon concentration.  The
similarity naturally remains irrespective of the stratification groupings,
be it 8 of 200 each or 10 of 160, etc. The use of rank of the independent
variable facilitates a comparison in terms of differently dimensioned
quantities, e.g. elevation and radon concentration.  Lung cancer mortality
shows an almost identical negative gradient with increasing rank of either
independent variable.  Suitably presented, elevation is apparently as
credible an independent variable regarding lung cancer as is average radon
concentration.  I trust this is of some interest to you and to others.
Chris Hofmeyr
Chris.Hofmeyr at webmail.co.za

Prof Cohen wrote:

> My paper responding to this is attached. To summarize briefly, The
> regression of lung cancer vs radon has a negative slope, B = -7.3,
> whereas LNT predicts B = +7.3, a discrepancy of about 25 standard
> deviations. Wes Van Pelt found that there is a very strong correlation
> between radon and elevation (above sea level), and between lung cancer
> and elevation. If this is taken to be a confounding factor in the radon
> vs lung cancer relationship, it reduces the negative value of B by half,
> to B = -3.6. This is still a long way from the LNT prediction, but Wes
> feels that it is close enough to B = 0 (it is still about 6 standard
> deviations away) -- he calls this a "flat line relationship" -- to
> negate any support for a hormesis interpretation.
>     On the other hand, if the correlation between elevation and lung
> cancer is interpreted as an explanation for the correlation between
> elevation and radon,  there is no effect on the radon vs lung cancer
> relationship and B = -7.3 is not affected. Thus, saying that the
> elevation vs lung cancer relationship cuts the negative value of B in
> half is an upper limit on its effect.
>     The issue here is what really affects the lung cancer rate, radon or
> elevation? Wes thinks it is elevation, due to differences in oxygen
> concentrations in air. It should be noted that this is not a matter of
> the difference between mountain tops and sea level. The median elevation
> of all counties is 803 ft, and half of them are between 497 and 1313 ft,
> which is not a great difference in oxygen concentrations. I interpret
> the correlations as due to the fact that high radon levels tend to be
> far from sea shores (Colorado and Iowa are the highest states)  and the
> entire center of the country is at the above elevations. But everyone is
> entitled to judge between our two viewpoints. I don't consider this to
> be vitally important because my goal was to test LNT and the conclusion
> from that test is unequivocal.
>
> Wesley wrote:
>
>>Dear Otto, Bernie and Radsafe,
>>
>>I would like to remind you of my paper in Health Physics (October 2003):
>>
>>EPIDEMIOLOGICAL ASSOCIATIONS AMONG LUNG CANCER, RADON EXPOSURE AND
>> ELEVATION
>>ABOVE SEA LEVEL - A REASSESSMENT OF COHEN'S COUNTY LEVEL RADON STUDY
>>
>>In this paper I plot the county average lung cancer rate vs. the
>> elevation
>>above sea level (altitude) and show an inverse association between county
>>average lung cancer rate and elevation. I postulate that the decrease in
>>lung cancer rates with higher elevations is caused by the carcinogenic
>>effect of higher absolute oxygen concentration in the inspired air at
>> lower
>>elevations. Stratifying Cohen's lung cancer vs. radon data into ten
>> groups
>>of counties with similar elevations removes some, but not all, of his
>>inverse association between radon and lung cancer.
>>
>>This correction for the confounding effects of altitude produces more of
>> a
>>"flat line" relationship between radon and lung cancer.... i.e., no
>>association between ambient indoor radon levels and lung cancer.
>>
>>Those who are convinced that ambient indoor radon causes lung cancer have
>>not, in my opinion, satisfactorily explained away Prof. Cohen's study.
>>
>>Best regards,
>>Wes
>>Wesley R. Van Pelt, PhD, CIH, CHP
>>Wesley R. Van Pelt Associates, Inc.
>>
>>-----Original Message-----
>>From: radsafe-bounces at radlab.nl [mailto:radsafe-bounces at radlab.nl] On
>> Behalf
>>Of Otto Raabe
>>Sent: Monday, February 12, 2007 12:02 PM
>>To: Bernard L. Cohen
>>Cc: 'RADSAFE'
>>Subject: Re: [ RadSafe ] Re: Differences in Background radiation and
>> disease
>>incidence
>>
>>At 08:28 AM 2/12/2007, Bernard L. Cohen wrote:
>>
>>
>>>         ---The problem is that the "powers that be" refuse to consider
>>>such evidence. My radon study does what you say with tremendous
>>>statistical accuracy, and it treats over 500 potential confounding
>>>factors, and does a lot of other things with no statistical uncertainty,
>>>but it is ignored. They refuse to even consider an ecological study and
>>>they ignore all the evidence I have presented.
>>>
>>>
>>>>**********************************
>>>>
>>>>
>>
>>February 12, 2007
>>
>>I believe that Prof. Cohen's studies provide the most reliable available
>>information about radon risk to the general public because they are based
>>on appropriate dosimetry. In his studies he used overall average
>>radon-in-home measurements in each U.S. County as a surrogate for
>> relative
>>lifetime average alpha radiation dose rate to the respiratory tract for
>>people living in each County. It is the LIFETIME average alpha radiation
>>dose rate to the lung that needs to be considered in estimating the risk
>>from radon in homes.
>>
>>In contrast, the case-control studies of indoor radon such as are
>> presented
>>in BEIR VI (e.g. Figure G-1, page 377) use selected isolated measurements
>>of radon concentration in each individual's home for each case or control
>>and there is no way that these measurements can properly represent the
>>LIFETIME average alpha radiation dose rate to the lung of any individual
>> in
>>contrast to other selected individuals. In addition, the resulting widely
>>variable results are used to assign a risk relative to ZERO lifetime
>> alpha
>>radiation dose to the lung even though there is not one single person in
>>the study whose lifetime dose is zero. Then, with these totally imprecise
>>surrogate estimates of lung dose for specific individuals, a regression
>>analysis is performed fitting these imprecise data to a log-linear (in
>> the
>>case of Figure G-1) or linear function beginning at zero dose. I think
>> the
>>results of these types of studies are inherently unreliable.
>>
>>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
>>***********************************************
>>
>>
>>
>
> --
> ÐÏà¡±á
>
> _______________________________________________
> 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://radlab.nl/radsafe/radsaferules.html
>
> For information on how to subscribe or unsubscribe and other settings
> visit: http://radlab.nl/radsafe/
>

-------------------------------------------
For super low premiums, click here http://www.webmail.co.za/dd.pwm