Re: The Latest on Low Doses....

["H.Wade Patterson" <hwade@talltown.com>]

Chris Davey wrote, in part:
> 
> Here's a quote from the latest Bulletin of the Canadian Radiation
> Protection Association.  (Vol 18 Number 1 January 1997, page 6)
> 
> (Editorial by Chris Pomroy:)
> 
> "A report in Nucleonics Week for 14 November 1996, reviews the
>latest RERF data which covers the period 1985-1990, and shows a
>cancer risk at doses as low as 50 mSv.

Chris then asked for comments. Here are mine.
1. Nucleonics Week is not a peer reviewed publication. Therefore, the
chance of having errors of fact or of logic is greater.

2. I have read many such papers (but not this one), and often their
method of analysing the data is this:

a. In so many words the authors state, and inspection of their figures
shows, that they have fitted the data (from Miners, Nuclear workers,
Japanese survivors etc.) to a dose response curve whose origin is at
zero and whose slope is everywhere positive

b. They have then extrapolated downwards to residential and occupational
radiation exposure levels.

c. Their method, therefore, guarantees that any derived risks will be
positive.

d. For this reason, it's circular reasoning to cite such papers as
supporting a dose response curve whose origin is at zero and whose slope
is everywhere positive (the LNT) because the authors have derived their
results from such.

3. Authors of "review reports" such as those from UNSCEAR, BEIR, ICRP,
and NCRP invariably cite papers that have used circular reasoning to
support their conclusions and recommendations.

4. I have documented this faulty reasoning in my paper:

"Setting Standards for Radiation Protection: The Process Appraised." It
will appear in Health Physics in March
Abstract
"Present radiation protection standards are based to a large extent on
data that has been forced to conform with the linear nonthreshold (LNT)
model. A review of the literature shows that there are examples of both
data and theory that disagree with such a model. Established standard
setting bodies seem not to have recognized this disagreement; indeed, as
will be shown, there are many studies that they have neither cited,
discussed, nor refuted. Additionally, examples of data adaptation and
circular reasoning are to be found in the standard-setting process.
Consequently, a new approach to the process is desirable.  Numerous
citations and quotations are given."


5.I would be grateful if anyone can send me a reference to a paper in
the peer reviewed literature where this error has not been committed. In
other words to a paper where first a "best fit" to the data has been
made, and then conclusions have been reached. Please don't bother
referring me to Cohen and Kondo. I already know about them. 

Cordially,

Wade

<hwade@talltown.com>

H.Wade Patterson
1116 Linda Lane
Lakeview OR 97630
ph 541 947-4974