Re: Confounders and Coincidences

[BERNARD L COHEN <blc+@PITT.EDU>]

On Mon, 5 May 2003, Otto G. Raabe wrote:



> case-control and cohort studies. In these studies unknown confounders are

> considered by proper control selection procedures and sufficient sample

> sizes.



	 --How can you say this? As a counter-example, suppose people of a

certain ethnicity have a tradition of keeping their homes well sealed,

which would give higher radon levels. Suppose further that people of

that ethnicity have a genetic disposition to getting lung cancer. Then

ethnicity would be an important confounding factor, indicating that high

radon causes lung cancer.



> In ecological studies there are no rigorous control selection

> procedures since we don't know who is exposed to what.

>

	--Section 3 of Item #7 on my web site covers my response to this.

If you do not agree, please let me know specifically what you question in

that section, or in the rest of that paper.



	--I have given above a specific example of how an unrecognized

confounder can negate the conclusions of a case-control study. Can you

suggest a specific example of how an unrecognized confounder can negate

the conclusions of my study? Before offering it, you might look at the

short section 3.2 of Item #7 on my web site.







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

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/