re: Self-selection bias

["Vesper, Bruce E" <BEVesper@MAIL.BHI-ERC.COM>]

RADSAFERS,



Having participated in some of the studies of the "Atomic Veterans" from

congressional hearings back in the mid 1980's, I seem to recall that the

total task force at Bikini for the Crosroads test was on the order of 30,000

military and civilian persons.  59% of approximately 1572 gives 927 cancer

deaths, which averaged over 30,000 gives something just over a 3% cancer

rate.  The interesting part of the study was the cause of death of these

veterans in the mid-1980's was that most had died from traffic accidents,

rather than other organic causes.  Admittedly, there has been some gap in

the study of these individuals that could have skewed the numbers some what,

but I doubt that it was more than a few percent.



Even more interesting was looking at the volunteers who were present in

advanced post, and who gained significant exposures in the 5+ rem range.  At

the time I attended hearings in 1986, 3 of almost 100 had died to that

point.  Causes of death were auto accident, heart attack, and a suicide with

a hand gun.  All the rest had been extensively tested and now showed any

evidence of ill health, aside from normal age related deterioration, and

even this was significantly less that their unexposed control population.  I

was appalled by one senate staffer who claimed that all should be included

in radiation deaths, since he could come up with an explanation as to why

each one could have died related to the exposure (committed suicide learning

he was exposed, shock cold have caused the heart attack, and the traffic

accident could have been another suicide disguised as a traffic accident.)



I should also point out that after 12 studies by the GAO that concluded

there was very little evidence that the "atomic veterans" were affected by

the radiation at Crosroads, the 'findings' of the 13th report totally

ignored the previous evidence and focused on the sources of inaccuracy of

the available dosimetry.  While not great, it was probably good within the

limits of the testing, and has since been totally ignored.



My opinion is that there is either an extreme healthy worker effect here

(which I strongly doubt), or hormeisis may have some other strengthening

evidence.



Bruce Vesper

Bechtel Hanford, Inc.

Richland, WA



The opinions and recollections are my own, and do not reflect those of my

corporation, though they would probably have wide support.





"If ignorance is bliss, why do we have so many unhappy people in the world

today!"



------------------------------------------------------------

Date: Mon, 16 Jul 2001 08:36:12 -0600

From: Brian Rees <brees@LANL.GOV>

Subject: Self-selection bias



Noted in the original article was a figure of "Tens of thousands" of 

personnel involved - sounds like a reasonable figure to me.  A 59% cancer 

mortality  rate is high in the smaller number of folks who joined the 

association, but since they chose to join the association (National

Association of Atomic Veterans) I'd suspect that there's a significant 

self-selection bias in the data.



(Obviously) my own personal opinion



Brian Rees

brees@lanl.gov









Bruce Vesper



"If pro is the opposite of con, does that mean Congress is the opposite of

Progress?"

(Source Unknown)



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

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.