Re: Radon, Smoking,& LNT

[epirad@mchsi.com]

One of the potential problems with such an analysis is that you are not 

adjusting for the covariates of smoking at the same time or considering within 

county non linearities.  This is a problem with ecologic data, no theoretical 

concrete example is needed to support limitations of ecologic data.  In 

addition, the smoking treatments you use do not account for smoking intensity 

or duration effects. 

> 

> On Tue, 24 Jun 2003, Jerry Cohen wrote:

> 

> >     I would certainly agree with Puskin, Lubin & Fields that a negative

> > correlation between smoking and radon levels would account for Bernie

> > Cohen's observed negative relationship between radon levels and lung cancer.

> 

> 	--I have shown (item #15 on my web site) that, with the Puskin

> type analysis, there is no set of smoking prevalences in counties that

> will explain the Puskin observation.

> 	Using the BEIR-IV type analysis my papers (item #1 as well as item

> #7 on my web site) show that even a perfect negative correlation between

> radon and smoking does not resolve the discrepancy between LNT and my

> data, if best estimates of the width of the distribution of smoking

> prevalence are used. Increasing the width of this distribution can resolve

> the discrepancy only with very highly implausible assumptions.

> 	By "perfect" I mean infinitely strong.

> 

> 

************************************************************************

You are currently subscribed to the Radsafe mailing list. To unsubscribe,

send an e-mail to Majordomo@list.vanderbilt.edu  Put the text "unsubscribe

radsafe" (no quote marks) in the body of the e-mail, with no subject line.

You can view the Radsafe archives at http://www.vanderbilt.edu/radsafe/