Re: Radon's linear dependency

[RuthWeiner@AOL.COM]

The problem, as I see it, is that in making and enforcing regulations, the federal agencies apply a simple (and simple-minded) linear conversion factor to "convert" dose (any kind: collective, average individual, external, CEDE) to "latent cancer fatalities" (which is how they refer to cancer incidence, and the correct term would be "latent fatal cancers"anyway, but that's another story).  Moreover, they assume that the simple linear conversion is good down to zero; i.e., no threshold.  This is then published as "cancer risk " -- nothing half as sophisticated as confounding factors, non-linearity, ecological vs case-control studies, ecological fallacies, etc., is even considered.  What is presented to the public, if I may paraphrase, is simply " if your dose is X rem, your risk of latent cancer fatality is 0.0005X"  or, for populations, "if the collective dose is Y person-rem, you can expect !

0.0005Y latent cancer fatalities in that population."


Dr. Cohen's papers show, above all else, that this simple dose-risk relationship is fallacious (they show a lot more, but I just wanted to make this one point).  Even Field. et al, shows that this simple application is fallacious (and shows a lot more as well).  It is not even a question of "disproving" the LNT, but of convincing, or forcing, or whatever, the regulatory agencies not to publish this kind of simplistic stuff.  Should we not try to convince DOE, EPA, NRC to just publish collective doses?  As far as I can see, Field et al doesn't support the LNT, and Cohen et al show its general inapplicability. (This is ENTIRELYaside from hormesis considerations.)  Even BEIR IV and BEIR V waffle on the question of a strictly linear relationship.


Finally, when I was on the EPA NAPCTAC committee, we really looked at documented air pollution effects when advising the Adminsitrator on standards (but that was pre-1982...).  Couldn't we get there again?


Ruth Weiner, Ph. D.

ruthweiner@aol.com