Re: Linearity of covariates and radon exposure

[epirad@mchsi.com]

Cohen wrote earlier today:  "I do not assume linearity of lung cancer risk with 

smoking. I use the BEIR treatments. Note that the BEIR Committees include 

epidemiologists."



I sincerely do not see how the above statement can be justified. 



You may not assume that, but that is how you treat it.  



In your analyses: 



- Baseline lung cancer mortality is solely a function of age and smoking

- You assume all smokers have the same increased risk of mortality

- You assume smoking duration and intensity is the same for everyone

- You assume age is the only modifying effect of radon exposure 

and so on ,,,,



In one response in the journal in the past you responded to us that you crudely 

tried to introduce the pack-years concept.  For control of ecologic confounding 

you need more than crude attempts. 



Unlike the Cohen derived LNT model, the BEIR model is quite general and allows 

for any degree of control for smoking.  The use of pack-years or pack-year rate 

would result in a much better approximation of smoking exposure then the 

dichotomous model used by Cohen.  The BEIR model could even include nonlinear 

effects of smoking unlike Cohen's model. 



Bill Field



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

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/