Re: Alternate explanations

[Andy Hull <hull@mail.sep.bnl.gov>]

At 02:58 PM 3/3/98 -0600, R. William Field wrote:
>Dale,
>
>Now that you waded in, you stated "    ",
>
>
>"Dr. Cohen's data show the bodies are missing to a high degree of precision."
>
>How does his data do that?  What do you mean by precision?
>
>
>"Random confounding factors can only dilute the effect not invert it."
>
>Why do you think the confounding factors are random?  
>
>Dale, I would suggest you get a copy of a paper by Hal Morgenstern,
>American Journal of Public Health, Uses of Ecologic Analysis in
>Epidemiologic Research, Volume 72(12)1337-1343, 1982.  
>
>"A long time ago I took age adjusted cancer deaths from the 1985
>American Cancer Society pamplet and plotted them against mean
>state altitude.  To a very great degree the cancer death rates were anti-
>correlated with mean altitude and therefore background exposure."
>
>Did you adjust the rates for smoking?
>
>Regards, Bill Field
>bill-field@uiowa.edu
>
Bill,
It may be of interest that Frigerio et al, found a similar anti-correlation
in their study "Carcinogenic Hazard from Low-Level, Low-rate Radiation"
ANL/ES-26, Part 1
(1973).  Their study compared the state by state mortality rates for 56 ICD
classifications of cancer for the years 1950-67   with the state by state
background
radiation levels as reported by Minx et al , USEPA ORP/CSD72-1 (1972), and
found a negative correlation for all cancer types except leukemia (for
which no correlation was found). While this was admittedly an ecologic
study (in today's parlance), it raised questions which in my mind merited
further investigation (which was not made).
Without belaboring the point, other comparable studies of populations in
high- and low-background areas in other countries have shown the same.  

While they may not rigorously disprove he applicability of the LN-T to
low-level, low dose-rate radiation, they seem to me suggestive of its
non-applicability.


Andy Hull
S&EP Div-BNL
Upton, NY 11973
Ph   516344-4210
Fax  516-344-3105
e-mail: hull@mail.sep.bnl.gov