Re: I'm get tired of it too, but...

[Jim Muckerheide <jmuckerheide@delphi.com>]

Bill,

William Prestwich wrote:
> 
> Jim,
>         I am certainly one of the unwashed when it comes to epidemiology,
> but my basic instincts as a scientist of some 40 years lead me to conclude
> that Dr.Cohen's data is at variance with current risk assessments, and
> casts grave doubt upon them. I think the response that it is his
> responsibility to show that the rest of the world is wrong is
> incorrect. At one time the world believed in a luminiferous ether. When
> experiments showed that the velocity of light in vacuuo was a universal
> constant, no one demanded the experimenters show why this were the
> case. The reaction was the correct one-to develop an alternative
> hypothesis.
>         Well, having said all this, I can't support your accusations that
> those on the other side of the debate are all members of a conspiracy and
> are behaving in a scientifically unethical manner. 

Certainly not all, not 'most,' but then they don't know much about some of the
scientific data or the history. Do you disagree as a matter of 'faith' or
'facts?'  Do you know of the dozens of msgs over the last few years on this
list that show otherwise? Or do you need the data?

>How, for example,
> could agencies of the United States Government suppress a Canadian study,
> as you seem to imply?

Original study, Miller89, NEJM, with Geoff Howe and others (UToronto), has,
e.g., a breast cancer rate of 0.67 at 15 cGy to the breast from TB lung
therapy fluoroscopies, 2.7 SD below controls. Misrepresents its own data to
claim the LNT applies (plotted and presented to NCRP by Ted Webster). (Claims
there would be 900 excess cancer/million women at 15 cGy, vs the data point at
15 cGy that shows 10,000 FEWER cancers in 1 million women.)

BEIR V reported it as one of their 6 primary sources demonstrating the LNT.
DOE recruited one author, Howe, to Columbia to take over RERF from NAS (an
outcry killed it). (Similar misrepresentation is endemic - draw a slope to a
high dose despite lower doses, e.g., in Cardis et al., the 1995  worker study
- and dozens of studies that are not published or ignored by the
ICRP/NCRP/UNSCEAR/BRER group.)

But Howe revised the breast cancer study in '96 and collapsed the low-dose
groups (most of the cases) to one group from 1-49 cGy. [Howe, G.R. and
McLaughlin, J. (1996) Breast cancer mortality-survivors study, Radiation Res.
145; pp694-707]

The Oct 98 NCRP SC1-6 Draft Report not only simply stated that this is fact,
but also simply stated that an in-press 1998 paper by Howe further confirms
the result. After voluminous chapters of esoteric data and rationalizations
about irrelevant data, they dismissed this "debate" in a paragraph. But Howe's
later paper was never pub'd, was withdrawn, and has not bee made available by
NCRP Sub-Committee 1-6!

He responded to challenges in various scientific meetings about suppressing
the data by diffidence and/or a disingenuous comment, e.g., at a 1997 NAS
meeting: "We found the low dose groups not informative." 

As noted ICRP, NCRP, BRER, UNSCEAR know the results and the challenge about
the data.

See, e.g., the BEIR V summary:
http://cnts.wpi.edu/rsh/Data_Docs/1-2/3/2/12325Beir90.html
and
http://cnts.wpi.edu/rsh/Data_Docs/1-2/3/2/12326po96.html

Plus a summary of the original study (Pollycove 1994), with the original NEJM
table and the Ted Webster plot, adding a curve connecting the data points:
http://cnts.wpi.edu/rsh/Data_Docs/1-2/3/2/12327po94.html

and a 1996 summary of how the 1994 UNSCEAR Report was changed by ICRP/NCRP
(App B and App A) AFTER it was approved by the UN in March:
http://cnts.wpi.edu/rsh/Data_Docs/1-2/3/2/12329po96.html

Note that Howe was appointed to BEIR VII. We objected based on this case. He
was removed. But unlike the anti's, we received no correspondence from the
NCRP acknowledging our comments. :-)

>         Finally, I think the comment by Michael Ford regarding the doctor
> in Colorado is very important. Would a quantitative assessment of his

Dr. Saccomanno passed away. He spent decades of frustration with the
gov't-funded group committed to suppress data on behalf of the LNT. Do you
know other persons who collect, analyze, and publish data when they aren't
supported by the establishment.

> experience be consistent or inconsistent with the current risk estimates
> and could an expanded study of the experiences of doctors in similar
> situations be useful. 

Who do you know that will do what Cohen or Saccomanno have done to further
science at their own initiative and expense in the face of committed
opposition and inability to obtain gov't funds? Or who will provide funds
independently?

>It doesn't seem to me these would fall under the
> category of ecological studies.

But even if you could find people to do the work, the data is very poor. The
statistics are much worse than applying ecological data. (Case-control studies
are "good" compared to "eco" studies ONLY because/when they have, by
'definition,' very good individual dose data to limit the variation in that
parameter. 

But that's not true for radon case-control studies! Measuring a mine or
residence does not equate, in MOST cases, to knowing the dose to an
individual. 

Local radon concentrations are highly affected by air flows and other
conditions. It "pools" and makes it highly unlikely that 2 persons in the same
area at the same time will have an equivalent dose, and even less likely that
2 other persons in another area at the same measurement will have an
equivalent dose. People move in and out of the high concentrations in
different ways. When the French did some work on initial monitoring of
individual miners, they found that great variation in individual doses vs. the
area measurements. (Later data got better with better bases for how/where to
monitor.)

So you can not 'know' with certainty the dose of individual miners. Area doses
are only indicative of a range, and poor at that. So a dose-group of miners
equate to a statistical variation of doses around the measured value - it
suddenly looks like an ecological study!?  

But: The miner studies are small. They do not have the large statistical bases
and validity of a large ecological study. And they have many confounding
factors unaccounted for (allowing the ICRP/NCRP and their closed group to
promulgate studies that "do not refute" the LNT). Some of these factors have
been addressed in the later/larger miner cohorts as ventilation aided in more
consistent radon distributions, there was more and better measurement and data
'at the miner' level (but with much lower doses as the initial rad protection
efforts had great effect on lowering doses (like the dial painters and
radiologists earlier in the century - never a dial painter with bone cancer
after 1925 when brushes were no longer 'tipped' in the mouth), and with the
few studies, like Saccomanno's, that did a better job of
understanding/following individuals (as people not as data points) in terms of
attributing exposures/doses. (And of course, after the high-dose era, only
smoking miners got an excess of lung cancer! :-)

> Sincerely,
> Bill Prestwich,
> McMaster University
> Hamilton, Ontario.
> E-mail prestwic@mcmaster.ca

Regards, Jim
muckerheide@mediaone.net
========================
************************************************************************
The RADSAFE Frequently Asked Questions list, archives and subscription
information can be accessed at http://www.ehs.uiuc.edu/~rad/radsafe.html