Re: EMF CONTROVERSY, > CONTROVERSY

["John Moulder" <jmoulder@post.its.mcw.edu>]

> We are distressed by the statement of the EMF panel and are at a loss 
> to understand how they could come to their statement in the face of 
> the experimental evidence and of the carefully considered analyses 
> which are readily available.  

Its fairly easy to understand how they came to their "possible carcinogen" 
conclusion when you examine the process they used.

For reasons that are unclear, the working group used the IARC classification 
scheme for carcinogens.  Under the IARC scheme, there are only five possible 
designations for an agent:

--------

Group 1:  The agent (or mixture) is carcinogenic to humans, and the exposure 
circumstance entails exposures that are carcinogenic to humans.  
  Supporting data required:  Sufficient epidemiologic and sufficient animal  
    data
  Examples:  Asbestos, benzene, diethylstilbesterol, radon
  Substances so classified as of early 1997:  74

Group 2A:  The agent (or mixture) is probably carcinogenic to humans and the 
exposure circumstance entails exposures that are probably carcinogenic to 
humans
  Supporting data required:  Limited epidemiologic evidence and sufficient
    animal data
  Examples:  Benzo(a)pyrene, creosote, formaldehyde, ethylene oxide
  Substances so classified as of early 1997:  56

Group 2B:  The agent (or mixture) is possibly carcinogenic to humans and the 
exposure circumstance entails exposures that are possibly carcinogenic to 
humans
  Supporting data required:  Inadequate epidemiologic evidence and
    insufficient animal data
  Examples:  Chloroform, coffee, phenobarbital, saccharin
  Substances so classified as of early 1997:  225

Group 3:  The agent (or mixture) is unclassifiable as to carcinogenicity in 
humans
  Supporting data required:  Agents do not fall into other groups (e.g., no
     data available)
  Examples:  Caffeine, mercury, jet fuel
  Substances so classified as of early 1997:  480

Group 4: The agent (or mixture) is probably not carcinogenic to humans.
  Supporting data required:  Evidence suggesting lack of carcinogenicity in
     humans and animals
  Examples:  Caprolactan
  Substances so classified as of early 1997:  1!

---------

For classification in group 2B (which is what the working group meant by 
"possible carcinogen"), all that is required is an epidemiologic association.  
The epidemiology does not have to show a strong association, it does not have 
to be particularly consistent, and it does not have to show an exposure 
response trend.  Positive laboratory evidence can be used to upgrade from 
group 2B to 2A, but negative laboratory evidence (like there is for power-
frequency fields) cannot be used do downgrade from 2B to 5.  Biophysical 
implausibility plays no role in this scheme.

There is some weak epidemiological evidence for an association between 
residence near powerlines and childhood leukemia, and there is some weak 
evidence for an association between work in "electrical occupations" and adult 
leukemia.  The epidemiological evidence is very weak, the laboratory evidence 
is extensive and unsupportive of a leukemia link, and as discussed in the 
previous post, the association is physically implausible.  But this weak 
epidemiological evidence is all that is required for a "possible carcinogen" 
designation under the IARC scheme.

If they had used a weight-of-evidence approach that considered all lines of 
evidence, or even if they had used a Hill criteria approach, they might have 
come to a conclusion that sounded rather different.  

But much of what the NIEHS working group actually said is not that different 
from what previous panels have said.  They say that the data does not indicate 
that power-frequency fields are "probable carcinogens", but that the data does 
not prove they are not carcinogens. 

BTW:  Much of the above is lifted from:
KR Foster, LS Erdreich, JE Moulder: Weak electromagnetic fields and cancer in 
the context of risk assessment.  Proc IEEE, 85:733-746, 1997.
John Moulder (jmoulder@its.mcw.edu)