Re: Cohen's Ecologic Studies

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

On Mon, 3 Jun 2002, Ted de Castro wrote:



> Excuse me - but - it seems to me that analysis using aggregate data MUST

> presuppose that effects are linear and you are using aggregate data for

> analysis of confounders.

>

> If response to confounders is NOT linear - then they can be accounted

> for only by groupings in discrete exposure level groups and NOT ranges

> and such grouping would require individual data.



	--Grouping does not require individual data. For example, consider

annual income as a possible confounder. Suppose the lung cancer risk from

radon is important only for people with incomes between $10,000 and

$15,000 per year. This is certainly a non-linear dependence of risk on

income. But I treat the fraction of the population within this income

range as a confounding factor. It does not affect my results.

	Of course I treat all other income ranges as potential confounding

factors and none of them change my results.

	Of course I do this with other potential confounding factors also.



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

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/