[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
RE: Cancer deficiency clusters
Philippe,
Thank you for the information. I should mention one factor that may be
considered which may go to the question of bias. A special medical history
was required of "radiation workers," and two questions concerned a history
of cancer or a family history of cancer. If the response was "yes" to
either, physician was to review the record and, if necessary, the medical
information was sent to a review board to ensure that there was not any
additional risk dealing with radiation sources. (Read this to mean "we will
not employ you to work around radioactive sources if there is a chance that
you will get cancer and sue us.") My guess is that if one or more members
in a worker's family had leukemia, they were not hired to work in radiation
areas.
A very good review of bias in epidemiology can be found at
http://ipsapp002.lwwonline.com/content/getfile/64/48/19/fulltext.htm
I think the "health worker effect" should only be considered when you are
comparing some worker group with the general population. This was a cohort
study of individuals doing the same types of work, except some where being
exposed to a "toxin," ionizing radiation.
-- John
John Jacobus, MS
Certified Health Physicist
3050 Traymore Lane
Bowie, MD 20715-2024
E-mail: jenday1@email.msn.com (H)
-----Original Message-----
From: Philippe Duport [mailto:pduport@uottawa.ca]
Sent: Tuesday, August 13, 2002 3:14 PM
To: Jacobus, John (OD/ORS); 'RadSafe'
Subject: RE: Cancer deficiency clusters
Reduced cancer risk in US Shipyard workers
John and colleagues,
Notwithstanding the abstract written by G. Matanoski, anyone can
calculate the odds ratio for all cancers in the nuclear shipyard
workers, compared to their non-nuclear counterparts. Look at the
numbers in tables 3.6.A to 3.6.D (pages 359 to 366 of the June 1991 DOE
Report).
There are 848 cancer cases in 38220 nuclear workers and 878 cases in
32510 non-nuclear workers. The odds ratio for cancer death in nuclear
workers is 0.8175 (95% Confidence Interval: 0.7430 to 0.8995, using the
approximation of Woolf), with a two-sided p value <0.0001, that is, near
certainty. The Odds Ratio was calculated with INSTAT(R) (GraphPad
Software Inc., 5755 Oberlin Dr. # 110, San Diego CA 92121). Anyone can
check the calculation from the numbers given in the report. Odds ratios
are easy to calculate but correct confidence intervals are more
difficult to obtain without specialized software.
Similarly, the histograms of age at entry and age distribution in
nuclear and non-nuclear workers, built form Tables 3.1.A1 and 3.1.A2 of
the report, show that the two populations are quite comparable. The
reduced risk of cancer risk cannot be attributed to a healthy worker
effect in that study. Similar calculations can be done for specific
cancers and any other cause of death examined by the authors.
Since anyone can replicate the above calculations, this should put an
end to speculations and to unfounded affirmations on what the DOE report
really contains.
Reference
Matanoski G: Health effects of low-level radiation in shipyard workers
final report. Baltimore, MD, DOE DE-AC02-79 EV10095, (1991)
Philippe Duport
-----Original Message-----
From: owner-radsafe@list.vanderbilt.edu
[mailto:owner-radsafe@list.vanderbilt.edu] On Behalf Of Jacobus, John
(OD/ORS)
Sent: July 19, 2002 9:43 AM
To: 'RadSafe'
Subject: RE: Cancer deficiency clusters
Howard,
But if we question the validity of statistical cluster that show a
higher
increase of cancer near a power plant, it should be, and is,
statistically
probable that studies will show a statistical cluster of healthful
effects
of radiation.
You noted the Nuclear Shipyard Worker Study. It you like, I will send a
copy of the abstract written by one of the authors of the study, Dr.
Genevieve Matanoski. It appeared in Radiation Research 133, 126-127
(1993).
(I think it is important to supply references rather than speculations,
don't you?) Quoting the abstract:
"The data clearly indicate that both nuclear worker groups have a lower
mortality from leukemia and lymphatic and hematopoietic cancers than
does
the nonnuclear group. All three groups have lower rates than the
general
population. However, if the NW<5.0 or the 5.0-9.0 mSv group is used for
comparison, than all dose groups 10 mSv and above in the NW>/= 5.0 group
have higher mortality rates that the NW<5.0 group for both leukemia and
all
lymphatic hematopoietic neoplasms. There is no consistent dose response
with radiation, which would suggest that radiation is not the factor
associated with the increase. . . . The SMRs are very sensitive to any
changes, such as lagging, due to small numbers, so these within-group
observations may simply represent chance variations."
A couple of simplistic observations, since I am not an epidemiologist
but
feel that I can understand the basics. First, this is a cohort study,
which
may not have any relevance to public exposures. I guess that if you
want to
reduce your risk of cancer, you should work in Naval shipyard, since
"All
three groups have lower rates than the general population." Second, the
risk mortality increases at occupational exposures above 10 mSv.
Third,
other factors that were not considered, such as asbestos exposures, may
influence the results as they may be more important than the factor,
radiation, that is being studied. For example, the study indicates that
the
mortality from lung cancer for all groups was higher than that of the
general population. But only the non-nuclear workers had a
statistically
significant increase. I guess you can conclude that radiation reduces
the
risk of lung cancer in this cohort. Of course, I would rather a more
qualified epidemiologist than I review the data. While I admire John
Cameron for his work, he is not an epidemiologist either.
-- John
John Jacobus, MS
Certified Health Physicist
3050 Traymore Lane
Bowie, MD 20715-2024
E-mail: jenday1@email.msn.com (H)
-----Original Message-----
From: hflong@postoffice.pacbell.net
[mailto:hflong@postoffice.pacbell.net]
Sent: Friday, July 19, 2002 7:03 AM
To: jjcohen
Cc: Gibbs, S Julian; Jacobus, John (OD/ORS); Radsafe Mail list
Subject: Re: Cancer deficiency clusters
Yes, Jerry,
Such a study has been done on 27,872 nuclear shipyard workers - but
until
recently only reported as not showing expected increase in cancer.
John Cameron, one of 8 members of the technical advisory committee of
the
Nuclear Shipyard Workers Study reports, "The cancer death rate of the
NW>0.5
group [those receiving an extra 0.5 rem] was over 4 std.dev. lower than
the
NNW control group [non-nuclear workers of similar ages and jobs]. This
good
news is not mentioned but the data are available in the final report."
http://www.aps.org/units/fps/oct01/a5oct01.html
. . .
************************************************************************
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/