Radon and lung cancer

["R. William Field" <bill-field@uiowa.edu>]

Cohen stated 

All I ask is that they show that it is possible to concoct a not implausible
explanation for my data. If their position is correct, that should not be
difficult.   
			
Field Response ----------------

Dr. Cohen,

I am not surprised you could not find an analytical explanation for your
findings.
Greenland and Morgenstern showed in 1989 that it is not always possible to
identify empirical sources of ecologic bias from aggregate data alone.  In
order to explain Dr. Cohen's findings in definitive quantitative terms, a
nationwide study at the individual level would be required.  This
individual level study would need to gather information on individual
smoking habits (pack-years, pack-year rate, etc.) and other risk factors.
Once this nationwide study was completed, the individual level data could
be compared to the ecologic data and the empirical sources of  ecologic
bias could then be described in detail.  Obviously, such an undertaking
would require a cost exceeding the $5,000.00 you are offering.   

The individual level surveys are performed on a smaller scale, they are
called case-control studies.

Bill Field
bill-field@uiowa.edu

******************************
R. William Field, Ph.D.
Division of Epidemiology
Department of Preventive Medicine
  and Environmental Health
N222 Oakdale Hall
University of Iowa
Iowa City, Iowa 52242
319-335-4413 (phone)
319-335-4747 (fax)
mailto:bill-field@uiowa.edu
******************************