Re: Mr. Ford and Iowa Radon Lung Cancer Study

[Ted de Castro <tdc@XRAYTED.COM>]

Yes - I got the point that it was not uncommon AND I presumed it was

done professionally and properly.  I just do not find it credible that

it can be modeled to a high enough level of accuracy AND precision that

the residual error from an overwhelming confounder will allow a smaller

effect to be discerned.



Rad health wrote:

> 

> >From: Ted de Castro <tdc@XRAYTED.COM>

> >I am having trouble with this - I don't see that it matters that the

> >mathematical correction is a "common" or "standard" practice of

> >"correction" - THAT claim does not speak to its precision or accuracy.

> >I find it hard to fathom that medical science knows the relationship

> >between smoking and lung cancer well enough to reduce it to a

> >mathematical formula - let alone a formulation so accurate and precise

> >that it can be used to correct an overwhelming confounder (small errors

> >with a correction of a large number will allow large errors in the

> >assessment of a smaller number).  It is hard to accept.

> 

> Ted, I think the point he was making was that it is not that uncommon to do

> this in everyday practice.  It is not a matter of medical science

> understanding it, it is a matter of whether or not smoking can be modeled

> accurately for the subjects.  This modeling can take a week or so to perform

> using standard techniques of multivariate regression.  Just because it is

> standard does not mean it was done correctly.  But looking at who the

> statisiticians were gives me confidence it was done correctly.

> 

> Besides they looked for residual confounding and did not find any.

> 

> http://www.public-health.uiowa.edu/faculty/~woolson.html

> 

> It was my understanding the retrospective recall was based on life events.

> People know when they had kids, worked, were at home, etc.

> 

> Don

> 

> _________________________________________________________________

> Chat with friends online, try MSN Messenger: http://messenger.msn.com

> 

> ************************************************************************

> 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/

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

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/