Re: Mechanisms are Needed to Explain Cohen's Data

[Kai Kaletsch <info@eic.nu>]

----- Original Message -----

From: "BERNARD L COHEN" <blc+@pitt.edu>

To: "Kai Kaletsch" <info@eic.nu>

Cc: "RadSafe" <radsafe@list.vanderbilt.edu>

Sent: Friday, January 11, 2002 11:09 AM

Subject: Re: Mechanisms are Needed to Explain Cohen's Data



> > > In case-control studies, they don't ask whether there are other

> > > people in the house who smoked.

> >

> > They don't have to. They actually measure the radon concentration of the

> > house and assign it to the occupants.

>

> --I thought the problem you raised was that the radon progeny are

> different for smokers and non-smokers.



3 separate mechanisms were proposed (all related to smoking): 1. The

presence of smokers systematically decreasing radon gas levels for

non-smokers. 2. The effect of smoking on the conversion between radon gas

and WLM. 3. The effect of smoking on the conversion between radon gas and

lung dose. (Numbers 2 and 3 deal with the conversion between a measured

quantity and a quantity of interest.)



> > You are trying to use the total amount

> > of radon in the county and divide it up between smokers and non-smokers.

>

> --I am not trying to divide it up. I normally assume the same

> average radon levels for smokers and non-smokers. In special treatments, I

> assume there to be differences and investigate the consequences.



It is the special treatments that I was referring to (latter part of Section

D in your paper on treatment of confounding factors). I think these special

treatments are necessary for any study that tries to investigate the link

between lung cancer and anything else. Especially, if the "anything else" is

systematically related to smoking.



> --My treatments of plausibility of correlation are based on the

> assumption that there is no strong direct mechanism,



If everything but radon concentration and lung cancer is random and if your

sample is big enough, everything "comes out in the wash". The question is

what is a "strong" direct mechanism. The argument being made by others is

that small systematic mechanisms (they call them biases) can have a large

effect on the shape of your graph. My position is that they must identify,

or at least propose, those mechanisms before I buy the argument.



Best Regards,

Kai Kaletsch



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

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/