|
Howard,
there are a lot of ways to make the ecological data
not contradict the case-control data, but they require a bit of "out of the box"
thinking. My favorite goes like:
Exposure to radon is bad for you when you are sick
or sleeping (at home). Exposure to radon is good for you when you are healthy
and active (out of the house and exposed to the county average).
That way people with high levels in their homes are
expected to have higher levels of lc than their neighbors, but the total number
of lc in the neighborhood decreases with higher average Rn levels.
Why do we assume that the effects of very low doses
of radiation are related to cumulative dose?
Kai
----- Original Message -----
Sent: Thursday, June 06, 2002 5:20
AM
Subject: Re: Cohen's Ecologic
Studies
Dear Bill Field,
I do conceive of a way both you and Cohen
could be right about your own studies.
What if c 4 to 6 pCi/L were a "threshold"? If below that, down to c 1.0
pCi/L there was a lower risk of cancer than either more OR LESS radon
concentration, both could be correct.
Do you have data to compare lung cancer rates where radon was 2-4 pCi/L
with locations having <1.0 pCi/L? You wrote that most Iowa controls as well
as cases had exposure to > 5.0 pCi/L radon, far above the USA average
for Cohen-studied counties.
Howard Long
,
|