Jeremy and group,
Gentner and Osborne give a good review of what
LNT is. However, in their paper, the attribute the reduced cancer risk
in the US shipyard workers, compared to the general population, to the
Healthy Worker Effect (HWE). The argument for the HWE can be made for
cohorts whose cancer incidence is compared to that in the general
population. For the US shipyard workers, however, what is interesting
and available is a comparison of identical groups of workers, one exposed, the
other not. In that case, the HWE, if it exists, cancels
out.
I tried to put that comparison in a table, that I
sent to myself to check if it arrives in a readable form. It does for me
(it is also attached as a MSWord document, just in case).
Here it is :
|
|
|
NW>0.5 |
NW<0.5 |
NNW |
All
NW |
NW vs
NNW |
|
|
Code |
O |
E |
O/E |
|
|
|
|
|
|
O |
E |
O/E |
[O/E(NW)]/
[O/E(NNW)] |
|
all causes |
0 |
2215 |
2975.91 |
0.74 |
973 |
1173.59 |
0.83 |
3745 |
3685.41 |
1.016 |
3188 |
4149.5 |
0.768 |
0.76 |
|
all neoplasms |
1 |
603 |
632.3 |
0.95 |
243 |
254.23 |
0.96 |
878 |
784.6 |
1.119 |
846 |
886.53 |
0.954 |
0.85 |
|
leukemias &
aleukemias |
204 |
21 |
24.2 |
0.87 |
4 |
9.87 |
0.41 |
29 |
30.96 |
0.937 |
25 |
34.07 |
0.734 |
0.78 |
|
all non-neoplasms |
(#0-#1) |
1612 |
2343.61 |
0.69 |
730 |
919.36 |
0.79 |
2867 |
2900.81 |
0.99 |
2342 |
3263 |
0.72 |
0.73 |
NW are the nuclear
workers
NNW are the Non-Nuclear
Workers
O is the observed number of
cases (death from all causes, from all cancers, from
leukemia)
E is the expected number of
cases, based on the US general population
O/E is the ratio
obs/exp. (O/E<1
means reduced risk)
It is clear that the NNW have
a lower risk of death, at the time of the study, than the NNW, and this
cannot be attributed to the HWE.
The same is true for the British
Radiologists, with non-radiologists MDs a the control group (No possible
HWE) - The apparent risk reduction is about the same in the US shipyard
workers and the radiologists.
Philippe Duport
Centre for Low-dose Radiation Research
Institute of
the Environment
University of Ottawa
P.O. Box 450, Stn. A
Ottawa, ON,
Canada K1N 6N5
Tel: (613) 562 5800, ext. 1270
Fax: 613) 562 5863
pduport@uottawa.ca
----- Original Message -----
Sent: Friday, June 07, 2002 9:05
AM
Subject: RE: [cdn-nucl-l] LNT
model
Edward,
If your understanding of the LNT model doesn't seem kosher
that's because it isn't -- the theory does not assume "that the effects of
radiation at low doses can be extrapolated in a linear fashion from the
effects of high-level radiation". The LNT model is a reflection of the
gross uncertainty in knowledge of the effects at low doses, and establishes
an upper administrative bound on such effects (if any exist) for use in
health physics planning. It is this upper bound that is extrapolated
linearly from high-dose cohort studies, and even then a reduction in slope
is applied to account for the beneficial effects of lower dose rate.
The LNT hypothesis was never meant, as it is often misused,
to predict the health effect in large populations of small doses of
radiation, and nor was it ever meant as a statement of confidence in any
particular dose-effect relationship at low doses. At the very least,
if one DOES use the LNT model to quantify the health effect of low doses,
the lower bound should always be included as "zero" -- i.e., there could be
no effect at all, and this is consistent with the LNT hypothesis. You
could even extend the lower risk bound to the negative region (i.e. a
positive health effect), and this is also consistent, given that we're just
talking about bounds of uncertainty (and there is evidence that suggests a
positive health effect, also known as "radiation hormesis").
A very good summary of the current understanding was written
by Dr. Norman Gentner and Dr. Richard Osborne, both international experts
(Gentner is currently the Secretary of UNSCEAR, the United Nations agency
charged with studying the effects of nuclear radiation and the global
authority on the subject). Their paper, "Linear vs. Non-Linear:
A Perspective from Health Physics and Biology" (Pacific Basin Nuclear
Conference 1998) is available in PDF format at the link below:
www.magma.ca/~whitlock/Gentner-Osborne_PBNC98.pdf
Cheers,
Jeremy Whitlock
> -----Original Message-----
>
From: Edward Oleen [mailto:eoleen@earthlink.net]
> Sent: Thursday June 06, 2002 4:01 PM
> To: cdn-nucl-l@informer2.cis.McMaster.CA
> Subject: [cdn-nucl-l] LNT model
>
>
> Is there someone
who can point me to a site which
> will explain
exactly what the "LNT" model is and
> what it
predicts? I gather that it assumes that
> the
effects of radiation at low doses can be
>
extrapolated in a LINEAR fashion from the effects
> of a high level of radiation.
>
> This does not seem kosher to me, based upon
the
> effects of pharmaceuticals, etc...
>
>
_______________________________________________
>
cdn-nucl-l mailing list
>
cdn-nucl-l@mailman.McMaster.CA
> http://mailman.McMaster.CA/mailman/listinfo/cdn-nucl-l
>