Re: Deaths from fossil fuel burning air pollution

[Bernard L Cohen <blc+@pitt.edu>]

On Wed, 29 Nov 2000, ruth_weiner wrote:

> Re: definition of air pollution: I was responding to the question of how air
> pollution is defined, and I responded with a rather standard legislative and
> textbook definition, namely (briefly) that a pollutant is something not
> found in "clean air," or not found in high concentrations therein.  It makes
> no sense for each of us to make up our own definition.  There are certainly
> sources of air pollutants in addition to fossil fuel burning power plants
> (and indeed, natural gas plants emit fewer and different pollutants than
> coal plants -- no particulates, for example).  Automobile and truck exhaust,
> all kinds of smelters, coke ovens, agricultural and road dust, crop
> spraying, chemical plants are just a few. One source used by Pope, et al,
> is, I believe, a steel plant and not a power plant.

	--I meant the definition for purposes of my message. The people
who do these studies have concluded that fine particulate (<2.5 microns)
is a suitable surrogate for fossil fuel burning pollution. There is a long
history of how this was decided on, and I am not enough of an expert to
defend it here. I believe agricultural and road dust particles are coarser
than 2.5 microns.Most of the other things you mention are from fossil fuel
burning.

> 
> Re reference to the LNT: a linear conversion factor implies a linear
> dose-response relationship, does it not?  The objections to applying a
> linear conversion factor in the case of radiation health effects are, in
> part, that it involves an extrapolation outside of the data, but data on
> health effects from large doses of ionizing radiation certainly exist (e.g.,
> the radium dial painters study).  The citation of  the large number of
> "deaths" or "premature deaths"  postulated in the air pollution case also
> involves extrapolation and a linear conversion factor.

	--There is a huge difference. For radiation, there are no data on
effects below 20 rem, but people use it for much lower levels, truly an
extrapolation. For air pollution there are lots of data points throughout
the region where it is used, so there is no extrapolation involved


> Finally, to reiterate: years of life lost would appear to be a much better
> metric than death, or even premature death.  Controlling the confounding
> factors in putative deaths from inhaling air pollutants is extraordinarily
> difficult, if not impossible.

	--EPA and others do estimate the loss of life expectancy for
victims of air pollution - I have been using 7 years - and this allows a
quantitative comparison with nuclear victims for whom loss of life
expectancy averages about 16 years. Dozens of investigators have faced the
problem of controlling for confounding factors so this issue has been
worked over extensively. The early work on this was done by Lave and
Seskin and they wrote a book on the subject.

	--Granted that there are difficult problems, does that mean that
we should ignore the health effects of air pollution? That is the game
played by the anti-nukes. They say that nuclear power can kill people (<10
per year treated probabilistically, even with LNT), whereas they ignore
the 10,000 or so deaths per year from the competing source of
electricity, air pollution from coal burning. Basically, that is the case
for nuclear power if public health is the issue.
	How do you respond to the fears that nuclear power can kill
people?


 > 
> >
> >On Tue, 28 Nov 2000, ruth_weiner wrote:
> >>
> >> This type of conversion factor is now being applied to inhaled air
> >> pollutants (and this is in fact an application of the LNT theory), which
> is
> >> quite a stretch, and which I myself do not agree with.
> >>
> >> An air pollutant is defined in a number of laws and regulations as a
> >> substance other than nitrogen, oxygen, carbon dioxide, water vapor,
> ozone,
> >> and argon, or a significant quantities of a  substance like CO and some
> >> terpenes that may in very small quantities be constituents of clean air.
> >> Particulate matter is a pollutant whether it comes from a stack, is
> crustal
> >> dust, or comes from a volcanoic eruption.
> >
> > --Here I define air pollution as things emitted from fossil fuel
> >burning, with very fine particulate (<1.5 miicron) serving as a surrogate.
> >How do you explain the fact that there is a statistically robust tendency
> >for areas with high air pollution to have higher mortality rates than
> >areas with low air pollution, after considering other factors that may be
> >relevant? No linear-no threshold assumption is involved; these are
> >straightforward data. Also, how do you explain the fact that in a given
> >city, mortality rates are higher when pollution is higher? Dozens of
> >studies have corroborated these findings.
> >
> >
> >************************************************************************
> >The RADSAFE Frequently Asked Questions list, archives and subscription
> >information can be accessed at http://www.ehs.uiuc.edu/~rad/radsafe.html
> 
> 
> 
> ************************************************************************
> The RADSAFE Frequently Asked Questions list, archives and subscription
> information can be accessed at http://www.ehs.uiuc.edu/~rad/radsafe.html
> 

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