Re: Dumb questions and comments on ecological/case studies

[epirad@mchsi.com]

Dr. Cohen,



Since counties themselves do not smoke, then please 

explain to me how you can address this problem and 

covariates WITHIN each county associated with lung 

cancer such as socioeconomic status and other within 

county covariates?  The slope is just related to 

smoking, not all the covariates associated with smoking.



WITHIN COUNTY CONSIDERATIONS



1.  socioeconomic status (SES)

2. related exposure to increased occupational 

carcinogens that often correlates with low SES jobs

3. Lower SES also correlates with increased home 

asbestos exposure

4. Lower SES coorelates with increased evironmental 

tobacco smoke

5. SES correlates with poorer quality and avialabilty of 

healthcare

6. Urban pollution correlates with SES

etc....



These are all individual level confounding factors. Some 

group level confounders can not even be identified at 

the group level. Cross level confounding of these within 

county factors likely explain your findings.  How can 

you treat the mix of all these factors and their 

interactions (including additive effects)at one time? 

> > When we regressed Cohen’s adjusted smoking percentages

> > for males and females on Cohen's radon levels.  In both

> > instances there was a significant (p < 0.00001) negative

> > association between smoking and radon.  The slopes were

> > 2.698 for males and 1.981 for females.  Apparently,

> > there are more smokers in counties with low radon

> > levels.

> 

> 	--This has been mentioned in my papers many times. If it were not

> for this correlation, the discrepancy with LNT would be even larger. But

> still this correlation is not nearly large enough to make an important

> difference. That is what I meant when I said that this correlation could

> not be "very large".

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

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/