Re: New Steve Wing Study

[stabin@esper.com (Mike Stabin)]

I don't know if this will be helpful at all, but a while back I wrote up the
following parody of bad epidemiology.  Couldn't get the Journal of
Irreproducible Results to publish it.  Maybe some Rad-Safer's will find it
interesting in the current dialog.  If it would ever be of any help
(doubtful), feel free to quote it, just attribute the source.

Mike Stabin
Oak Ridge


In 1985, Voelz et al. (Voelz et al. 1985) published results of their 37-year
epidemiologic followup of Mannhattan Project plutonium workers.  While some
comfort is afforded by the results regarding the incidence of neoplasms and
other diseases in this small population, I find it somewhat disturbing to
note the increased incidence of fatalities due to auto/pedestrian accidents
(1 observed vs. 0.96 expected).  From the data, it appears that the
increased risk of such fatalaties is about 4% in the observed population.
The authors neglected to derive a dose-response model for these data,
believing, quite naively, that "None of the positive findings...can be
attributed to the Pu depositions."  In this enlightened age in which all
reasonable people recognize the clearly defined link between small amounts
of radiation and effects such as stillbirths, cancer, etc., it is difficult
to understand why the authors overlooked this obvious effect.

To derive a dose-response model, I will limit my discussion to dose the
bone, liver, and lungs, because these organs account for almost 100% of the
total transformations of Pu-239 in a class Y material (ICRP 1979).  From
Table 4 of Voelz et al., and from tables in ICRP 30, I estimate roughly a
collective committed dose of 37 person-Sv to the red marrow, 460 person-Sv
to the bone, 100 person-Sv to the liver, and 150 person-Sv to the lungs in
the exposed population.  This would result in a collective effective dose
equivalent of 43 person-Sv.  Therefore, the excess increased risk of
radiation-induced traffic fatalities of this type should be on the order of
9.2E-4 per person-Sv.  Assuming a linear, non-threshold model, this would
indicate that 870 auto/pedestrian fatalities in the U.S. each year may be
attributed to background radiation, based on a population of 270 million
people receiving an average of 3.5 mSv per year.  I think that it is a clear
and present danger that we face here, and I feel that we should make every
effort to immediately halt nuclear power production, coal power production
(because the lung dose from this pathway is comparable to or higher than
that from nuclear power production), natural gas consumption (due to the
Rn-222 component), and above all, we must find some way to completely
eliminate the public's exposure to cosmic and terrestrial radiation.

References

ICRP 1979 - International Commission on Radiological Protection, 1979.
Limits for Intakes of Radionuclides by Workers.  ICRP Publication 30,
Oxford, Pergammon Press.

Voelz 1985 - G. Voelz, R. Grier, and L. Hempelmann, 1985.  A 37-year medical
followup of Mannhattan Project Pu workers.  Health Physics, 48:249-260.