[Date Prev][Date Next] [Chronological] [Thread] [Archive Top]

RE: [cdn-nucl-l] Low level radiation and cancer: BAD NEWS

Title: Message
Thanks very much Jim.
 
Sorry if my post made it appear like I was trying to compare the two graphs -- that's not the case at all since, as you noted, they are "apples & oranges."
FYI, the Hydro-Quebec presentation (posted at http://www.bape.gouv.qc.ca/sections/mandats/gentilly-2/documents/DA17.pdf ) has similar cancer ERR per Sv graphs near the end (actually should be identical for the Canadian cohort, although the BMJ paper doesn't reference the more recent Zablotska et-al paper, preferring to cite some older one instead).
 
I just posted those two because I found them to be an interesting contrast.
 
Also, I find the "healthy worker" monicker somewhat misleading -- its pretty obvious that cancer is an old-age disease, and that by definition retired folks aren't "workers" - nuclear or otherwise.
 
It does raise an issue though, w.r.t. trying to get a dose-response line to fit the small number of cancers in a relatively youthful population (the workers) versus..... what ? .....a large population that includes a lot of elderly folks and a great number of cancers, but also very little dose.
To me that also sounds a bit like trying to compare apples & oranges -- I guess you need to be an epidemiologist to do that without too much blinking.

 Jaro
^^^^^^^^^^^^^^^^^^^^^

 
 
-----Original Message-----
From: cdn-nucl-l-admin@mailman1.cis.McMaster.CA [mailto:cdn-nucl-l-admin@mailman1.cis.McMaster.CA]On Behalf Of Dukelow, James S Jr
Sent: Wednesday, July 06, 2005 5:39 PM
To: Jaro; cdn-nucl-l@mailman1.cis.mcmaster.ca
Subject: RE: [cdn-nucl-l] Low level radiation and cancer: BAD NEWS

Jaro wrote:
-----Original Message-----
From: cdn-nucl-l-admin@mailman1.cis.McMaster.CA [mailto:cdn-nucl-l-admin@mailman1.cis.McMaster.CA] On Behalf Of Jaro
Sent: Wednesday, June 29, 2005 3:58 PM
To: cdn-nucl-l@mailman1.cis.mcmaster.ca
Subject: RE: [cdn-nucl-l] Low level radiation and cancer: BAD NEWS

Better yet : http://bmj.bmjjournals.com/cgi/rapidpdf/bmj.38499.599861.E0v1.pdf

The paper uses - in part - the same Canadian ARW data set that was cited in this report : http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15161357&query_hl=1

Analysis of mortality among Canadian nuclear power industry workers after chronic low-dose exposure to ionizing radiation, in Radiat Res. 2004 Jun;161(6):633-41.

by Zablotska LB, Ashmore JP, Howe GR.

 

.....whose findings have been summarised by others in the graph (image) I posted at http://www.cns-snc.ca/branches/quebec/reactions/CFCF_vanderHeyden_2005_May_16.html

....it comes from a Hydro-Quebec presentation (in French) to the public hearings of Quebec's BAPE ("Bureau d'audiences publiques sur l'environnement"), posted at http://www.bape.gouv.qc.ca/sections/mandats/gentilly-2/documents/DA17.pdf

.....obviously, the cancer risk for Canadian ARWs is low, below the 1.0 reference for the entire population, but presumably it will catch up, as the group ages.

Having done the age-adjustments, etc., the new international study came up with their results, illustrated in the graph below - in which the Canadian cohort appears to skew the overall results in a noticeable way :

Comments welcome ! (I notice that BH is one of the co-authors)

 Jaro
^^^^^^^^^^^^^^^^^^^^^

=================
 
The difference between the two graphs is that one is apples and the other is oranges.
 
The upper graph is the SMR = standard mortality risk, comparing age-adjusted mortality risk in the Canadian nuclear workers against a control population, probably of the general public.  It shows a strong healthy worker effect and/or the benefits of moderate levels of radiation exposure.
 
The lower graph, from the BMJ paper, is giving the excess relative risk per Sv of radiation exposure -- that is, the slope of the dose-response line.  So ... computing a dose response coefficient for the Canadian workers (probably using the linear regression line -- with no diagnostics provided and no assurance that the assumptions  required to legitimately compute a regression line are, in fact, satisfied) we have the assertion that a Canadian worker receiving a Sv of occupational radiation exposure would have 8 times the risk of cancer (compared to a worker receiving zero occupation exposure).   The risk for a worker receiving 0.1 Sv would be 1.7 times that  of the control worker.
 
I speculated in my posting on the BMJ paper that exactly this situation would hold for some of the worker populations -- healthy worker effect (i.e., a lower mortality rate) combined with a positive dose-response coefficient.
 
Regarding the ORNL data and Steve Wing et al., one of the BMJ co-authors is David Richardson, a Wing co-author on a number of papers, including the recent paper of Hanford worker radiation exposure and health.
 
Best regards.
 
Jim Dukelow
Pacific Northwest National Laboratory
Richland, WA
jim.dukelow@pnl.gov
 
These comments are mine and have not been reviewed and/or approved by my management or by the U.S. Department of Energy.