Sternglass paper on breast cancers

["Jeffrey A. Leavey" <certhp@vnet.IBM.COM>]

I thought I would post my own comments on this paper to get
some feedback:

Date: 11 November 93, 11:45:12 EST
From: Jeffrey A. Leavey         914-892-3950  certhp@vnet.ibm.com
      IBM X-ray Lithography Program

I've picked out a few sections for comment. I generally take the
position that if one part is suspect - the whole is suspect,
and this paper is suspect. If you don't know, Sternglass is one
of a few activists that is outside the regular body of
scientists in radiation protection.
Overall, this paper conveniently ignores many
confounding factors such as personnel history of the individual,
chemical exposures, population shifts, etc. that play an even
more important role than nuclear power plant releases. You are
correct in saying it is cleverly written and I agree it does
a disservice to the public. I have some specific comments that
may help in response to who ever you are communicating with.

First, Sternglass fails to perform an environmental pathway
analysis to assess population dose (the only TRUE way to gauge
risk). He assumes a priori what is released is what people
intake; bad assumption. He also fails to take into account
ALL exposures: medical, dental, Gov plant releases (which we
now know to be large as part of the Cold War) in determining
population doses. Even though medical doses are not recorded or
regulated, it all adds to a persons total exposure; some procedures
like cardiac imaging give 100+ rem doses (occupational yearly
limit is 5 rem). The bottom line effect finds a dose response
relation at radiation levels much lower than they truly are.
In other words, cancer appears to be caused by lower doses
because only part of the total dose is considered.
At best this is a grossly incomplete study in my opinion.

Next, a good example of data manipulation is Figure 5. This
plot shows the rate of breast cancer incidence in CT and
tries to link Haddam Neck and Milestone plants as the cause.
A more probable explanation is that 1945 was the end of WWII
and the start of rapid industrial growth and population rise.
I attribute the shape of the Sternglass' trend to the
exponential-like growth in all industry. Nuclear power is just
part of the overall industrial base growth. He also plots the
1945-70 trend as linear while I can see a curve in the data.

In Figure 8 and on page 796 Sternglass tries to link
well water contamination later in time with surface water
contamination to NYC. While I don't have the raw data to
examine, Sternglass does not look at other surface water
bodies around NY and other states to be sure the trend
continues away from NYC. Also the range in Figure 8 is
from 17 deaths/100,000 to about 22/100,0000 or a difference
of only 5. This additional 5 could also have been caused by
the increasing ground water chemical contamination that
started in the late 50's - 60's as the chemical industry
expanded - remember Love Canal - how many toxic sites are
unknown? To summarize: other possible causes are not examined
and ruled out if they truly are not the cause.

Lastly, the REAL test is whether Sternglass's results can be
corroborated by independent data. I did some quick dose
assessments to look at whether I-131 (or I in general) and
Sr-90 really are as bad as Sternglass asserts on page 798.
For I-131, the dose to the thyroid is 15,000 times greater than
the dose to breast tissue, because I concentrates in the thyroid.
The breast, however, is 5 times more sensitive to radiation than
the thyroid. Breast sensitivity is not an issue here because when
dose and sensitivity are combined, the thyroid is still about 3000
times more at risk. I would expect to see many more thyroid problems
well before any breast problems.

In the last paragraph on page 798 Sternglass states that long
retention emitters in bone gives long term low level doses.
He fails to tell how this relates to breast cancers since
free-radicals produced in a cell don't last long enough to
leave the cell. In other words, radioactive material in bone
causes free radicals in bone cells, not breast cells. Radiation
emitted by Sr-90 could leave the bone to expose nearby (within
a few millimeters) cells. Doing a dose assessment for the breast
from Sr-90 finds the dose to bone marrow is about 700,000,000 times
higher than to breast. The sensitivity of marrow is very much
the same as breast tissue so I would expect many more bone
disorders than breast cancers.

Sternglass tries to discount any contribution from background
sources like Ra, K-40, Rn, cosmic etc. in an attempt to
make bone seekers more important (page 798 para. 3).
He states that because radium is an alpha emitter it doesn't
expose bone. Wrong. The radium dial painters in the 30's
developed bone cancers. While Ra emits an alpha, the decay
products are beta/gamma emitters and their in-growth is very
quick compared to the biological lifetime of Ra. As for cosmic
rays, they irradiate the whole body, bone, breast, thyroid and
all. And cosmic rays are exposing us constantly where as Sr-90
decays away. I have a hard time following Sternglass' point
in the section other than he is trying to confuse those
less knowledgeable.

Finally, you may have heard of PC - Probability of Causation.
I have a reference that gives the PC based on total dose
obtained by a specific age. Since any environmental levels
of I or Sr would deliver a total long term dose of less than
10 rem, here are the PC numbers for a typical population:
---------------------------
 10 rem by       % Prob of
   Age           Causation
---------------------------
   1              19%
  20              5.7%
  50              0.36%
  70              0.072%