----- Original Message ----- 
From: "Deborah Harrell" <[EMAIL PROTECTED]>
To: "Killer Bs Discussion" <[EMAIL PROTECTED]>
Sent: Friday, February 20, 2004 7:13 PM
Subject: Re: Scouted: Nuclear waste court woes



> But we don't _live_ there for years.  Here is an
> abstract which urges caution in interpreting "adverse
> pregnancy outcomes" and exposure to radiation, yet
> *did* find a "statistically significant association
> between uranium operations and unfavorable birth
> outcome was identified with the mother living near
> tailings or mine dumps."
>
http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&list_uids=1399640&dopt=Abstract

They didn't say what statistically significant meant.  I'd guess 1 sigma
was statistically significant, but one sigma above the line would result in
a false positive about 15% of the time or so.

In particular, let us consider the last line of the abstract:

"Also, birth defects increased significantly when either parent worked in
the Shiprock electronics assembly plant. Overall, the associations between
adverse pregnancy outcome and exposure to radiation were weak and must be
interpreted with caution with respect to implying a biogenetic basis."

I interpreted this as data that, if combined with other similar data, would
have a correlation that was high enough to actually point to a correlation.
But, where could we find similar data?

Aberdeen and Denver are obvious choices.  If the tailing pile is that
dangerous, then Aberdeen should be more dangerous, since the background
level of radiation there is very high.  IIRC, it is high enough so that it
would be illegal if it came from an industiral exposure to the general
public.




> I think we've already agreed that uranium miners have
> an increased risk of lung cancer, but here is one
> abstract from a population-based case-control study
> about Navajo men and lung cancer:
>
http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&list_uids=10738707&dopt=Abstract
> "Navajo men who were underground miners have excess
> risk of lung cancer...Smoking did not account for the
> strong relationship between lung cancer and uranium
> mining. The Navajo experience with uranium mining is a
> unique example of exposure in a single occupation
> accounting for the majority of lung cancers in an
> entire population."

I'd be curious to see what fraction of the total population were miners.
I'm surprised that number is not in the abstract.  Its amazing that these
abstracts do not contain the simple information that would allow for a
quick calculation of the statistical significance of the findings.


>
> There is the rub: you seem to be assuming that not
> only can "we" keep track of and contain toxic &
> hazardous substances for centuries, but that whatever
> mistakes we are currently making can be corrected in
> the future -- 

There is suppose to be _one_ nuclear storage site for the US.  The waste
will be stored behind big doors that can be locked.  All that the
civilization 200 years from now will have to do is keep the doors locked.
How hard is that?

>I take the "conservative" approach,
> based on how we have lost track of things like where
> we buried/stored nerve gas bomblets a mere 50 years
> ago, and that certain birth defects are associated
> with the mother's living near hazardous waste sites.*

I also think we differ in how much faith we place in certain statistical
signals.  Lets take, for example, the reports of breast cancer being high
on Long Island than in the general public.  If one takes all known risk
factors into account, they go away.  A second is how many studies can show
a 3 sigma correlation with X and some form of cancer.  This seems to be
very significant until one realizes that there are many different types of
cancer and a different one has the 3 sigma effect in each case.  One could
have the same type of correlation if one were to do a number of different
studies comparing results for people born between 1:15 PM and 1:47 PM and
the general public.

The classic example is the data that was claimed, by some medical
researchers, to show a correlation between power lines and a number of
different diseases.  The APS used the data to show that power lines also
prevented diseases. The methodology simply overdid the risks.

In high energy physics, we had lotsa interesting looking bumps come and go.
For every real 3 sigma signal, there were scores of false alarms.  When
charm was found, I recall an 8 sigma signal that screamed for attention.

> We have different biases, yours an optimistic "can do"
> engineers outlook, mine a more cynical "law of
> unintended consequences" one [yet not quite Murphy's
> Law bad].

My actual criterion is we shouldn't strain at gnats and swallow camels. :-)
I would be happy to compare the dangers to the public from radiation with
the dangers from a number of other sources. I'd suggest that we worry about
the big causes before we get upset about something that might possibly kill
someone maybe in 200 years.  If we did really want that standard, wouldn't
we need to get rid of bridges, for example?

I realize that sound silly, and I'm not trying to mock your position.  But,
since even pessimistic analsyis  has fairly low radiation exposure
associated with it, except for people who are stupid enough to enter the
facility without taking precautions, then we are talking about the
possibility of dangers that correspond to the dangers inherent in old
bridges.


> *This abstract doesn't specify the type of hazardous
> waste site, but is a very large study (N > 13K):
>
http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&list_uids=12018013&dopt=Abstract
>
> Even living near a landfill increases the risk of
> "congenital anomalies," albeit a small increase:
>
http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&list_uids=11509424&dopt=Abstract
>
> And does this one that found "The risk of birth
> defects among index children was significantly
> associated with mother's military service in Vietnam"
> reflect chemical exposure, or something else?
>
http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&list_uids=10982986&dopt=Abstract

Yea, but look at the methodology:

"We compared self-reported pregnancy outcomes for 4,140 women Vietnam
veterans with those of 4,140 contemporary women veterans who were not
deployed to Vietnam."

How about a biased sample as the most likely cause of the observation?  I
know physicians who were doing work in that field years ago, and they found
nothing.


> > If it doesn't its
> > unlikely that someone will break in.  If they do,
> > maybe a few people would
> > die; maybe not.  But, if civilization fails,
> > billions would die.  Why have
> > nuclear power held to this standard when there are
> > zillions of other things
> > that are less important that would also fail this
> > criterion without anyone worrying about it.
>
> I don't think that continuation of our civilisation
> depends upon nuclear power,

It doesn't.  But, if you held every industry to the same standards that you
hold the nuclear power industry, we wouldn't have much.  For example, we
wouldn't have buildings because of the dangers of people falling during
construction.  We wouldn't have fossil fuel because of global warming. We
wouldn't have cars because of fatal accidents.

> I also hold the logging industry responsible
> WRT loss of old-growth forest that will not be
> duplicated/replaced until ~ 1000 years (coastal
> redwood forest), and the loss of the many species
> dependent upon those ecosystem (frex various salmon
> populations).

The reason that forests shrunk so is due more to your profession than any
other profession. <grin>.  What stopped the loss of forests was the advent
of tractors and the improvement of the technology of farming. I have no
argument with limiting or eliminating the harvest of old growth redwoods
now, but that's only a small fraction of the reason the forests in the US
are much smaller than 400 years ago.

Nuclear power is the only know technology that can be a cost effective
substitute for fossil fuels.  The risks associated with nuclear power are
much lower. I'd be willing to bet that the loss of life per MWh associated
with wind and solar power will be greater than nuclear, too.  Yet, nuclear
is the one everyone is afraid of.

Dan M.


_______________________________________________
http://www.mccmedia.com/mailman/listinfo/brin-l

Reply via email to