RE: Cancer clusters

["Dov Brickner" <brickner@zahav.net.il>]

DEar Colleages please calm down

don't panic when a new cluster is published , and don't dismiss a cluster

apriori, it becomes a bad habit. Some clusters are important, and a careful

analysis do  reveal  causation once in fifty years or so...

I am not an epidemiologist but my job keeps me in touch with the

epidemiologist community in Israel (I may say that some epidemiologists are

my best friends...:-)) and with cancer epidemiology.



 It is true that  most epi works consider the alpha to be 0.05. It doesn't

mean that 5% of these studies are false positive - on the contrary. Since

you look for cancer in many sites , every study examine more then 20

parameters and therfore carry agood chnace to find at least one site to be

(falsely) positive (i.e. significant). So I am not surprised that a

research like the one of Chris Busby , found a statistically significant

correlation. I think that a NEGATIVE association , like the association in

the Finnish Nuke workers (Auvinen A et al , CAncer incidence among Finnish

nuclear reactor workers, J Occ Environ Med44:634, 2002) study is much more

meaningful.

Dov  (Dubi)Brickner   MD

Beer Sheva    ISRAEL



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

> From: owner-radsafe@list.vanderbilt.edu

> [mailto:owner-radsafe@list.vanderbilt.edu]On Behalf Of Gibbs, S Julian

> Sent: Tuesday, July 16, 2002 6:41 PM

> To: Jacobus, John (OD/ORS); Radsafe Mail list

> Subject: RE: Cancer clusters

>

>

> Unfortunately it is not just this issue that provides studies for

> epidemiologists (Note I did not say income; most are salaried and

> get their money anyway.)  The multitudes of associations reported

> in the scientific literature and the lay press (e.g., eating peas

> of color x causes or prevents cancer in organ y).  Most of these

> studies use an alpha of 0.05.  That says that 5% of the studies are

> false positives!

> Note: I am not a statistician, but am aware of how one can distort

> the truth (I did not say lie) with statistics.

>

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

> S. Julian Gibbs, DDS, PhD               Voice: 615-322-1477

> Professor, Emeritus

> Dept. of Radiology & Radiological Sciences

> Vanderbilt University Medical Center     Home: 615-356-3615

> 209 Oxford House        Email:s.julian.gibbs@vanderbilt.edu

> Nashville TN 37232-4245     or alias:j.gibbs@vanderbilt.edu

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

>

>





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

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/