Re: Glass-Based Radon Measurements

[Kai Kaletsch <info@eic.nu>]

John,



Both the RR and the case and control numbers, from which I get the OR, come

from the same lines in the Table. As I understand it, the Bq/m3 are the

average over a 32 year assessment for both.



I have an idea of what happened. Its only a guess, but here goes:



It looks like the authors fitted their data to a linear excess risk model

and calculated a single excess risk per 100 Bq/m3 for the entire data set.

They then applied that value to the radon levels in each category. In other

words, the reported RR for a category has nothing to do with the data in the

category, but is a forced lineraization of the entire data set.



The casual reader looks at the RR values vs. radon concentration categories

and concludes: "Ooooo, look at what a nice, smooth, increasing function the

RR makes with radon levels!", when in fact these attributes are mathematical

artifacts and the raw odds don't support that conclusion.



Anyway, I'm not trying to convince anyone. Everyone can read the paper at:

http://www.nature.com/cgi-taf/DynaPage.taf?file=/jea/journal/v12/n5/full/750

0236a.html&filetype=pdf

(free registration required) and come up with their own explanations, or

they can Email the authors.



Best Regards,

Kai



From: "Jacobus, John (NIH/OD/ORS)" <jacobusj@ors.od.nih.gov>

To: "'Kai Kaletsch'" <info@eic.nu>; "RadSafe" <radsafe@list.vanderbilt.edu>

Sent: Friday, October 11, 2002 7:15 AM

Subject: RE: Glass-Based Radon Measurements





> Kai,

> Excuse by ignorance, but could it be that in calculating the relative

risk,

> value is based on the number of cases per person-years of exposure.  Thus,

> if the number of years exposed to that concentration is small, the

relative

> risk is higher.

>

> On the other hand, the odds ratio is just the number of cases presented,

> without any consideration as to how long the person was exposed to that

> concentration.

>

> -- John

> John Jacobus, MS

> Certified Health Physicist

> 3050 Traymore Lane

> Bowie, MD  20715-2024

>

> E-mail:  jenday1@email.msn.com (H)

>

> -----Original Message-----

> From: Kai Kaletsch [mailto:info@eic.nu]

> Sent: Thursday, October 10, 2002 5:17 PM

> To: RadSafe

> Subject: Re: Glass-Based Radon Measurements

>

>

> Friends,

>

> For those interested in Radon, I have calculated the odds ratios from the

> data given in the first 2 rows of Table 4 of the Lagarde et al. paper and

> graphed them. These rows give the number of cases and controls as a

function

> of radon concentrations as assessed by air based and surface based

> measurements respectively. The graph can be seen at

> http://members.shaw.ca/eic/odds.pdf

> Also shown on the graph (dashed lines) are the relative risks as reported

by

> the authors.

>

> To me at least, it seems that the raw odds tell a different story than the

> RR values.

>

> Please let me know if I made any mistakes in calculating the odds ratios.

> . . .

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

> 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/