FEB - The Swedish Association for the ElectroSensitive


EMFguru #6-98, Milham on Carcinogenicity of EMF ... or ...

"The weight of the evidence".....

(Sep 98)

Hi everybody:

In the following document I am forwarding the "summary" and some other quotations from the most recent paper published by Dr. Sam Milham (Washington State Department of Health, USA). (European Journal of Oncology, vol. 3 n. 2, pp. 93-100, 1998)

None of what follows is my interpretation.  All of what follows are the direct quotes from the paper, including the Summary -- which appears first.  I am doing it this way, as I did the Liburdy paper on Melatonin last April, in order that you may have the author's words without any filter.  The paper is too long for me to type it all, so you will see that there are skips, thus ///skip///.

However, the suggestion in the 'subject title' above -- that the paper might be named "The Weight of the Evidence" -- is the only interpretation or comment I am offering.  If you have been reading much of what I have written over the last year or two on this network, you know that "the weight of the evidence" is what I regard as the strongest case for the fact that EMF can be hazardous.  I suggest that in this paper is the proof; Dr. Milham has documented it......


Roy Beavers (EMFguru)
It is better to light a single candle ...
than to curse the darkness!

In the twenty years since publication of the original Wertheimer-Leeper study associating magnetic fields with childhood cancers, about 40 residential and 100 occupational epidemiology studies have been published with nearly 500 separate risk ratios.  For every lowered risk ratio about six are elevated.  A number of these studies also show a significant dose-response between magnetic field and cancer incidence.  The low relative risks seen in these studies may be due to the fact that there are no unexposed control groups available, and that the magnetic field meters used in exposure assessment may have failed to detect very low frequency magnetic fields, further compromising risk calculation.  The electromagnetic field environment in offices, cars, and in special environments like tanning beds should be better characterized.


The early EMF studies were criticized for lack of field measurements (use of job titles and exposure surrogates like wire code), for having low risk ratios, and for failing to show a dose-response relationship.  The next generation of epidemiologic studies, both residential and occupational, took field measurements, and in some of the studies, showed dose-response relationships.  The Feychting-Ahlbom study (1993) (table 12), and the Floderus study (1993) (table 13) are examples.

I think that the best explanation for the low risk ratios found in the EMF studies is that the controls are exposed.  In table 14, I present the basic data of the Doll and Hill (1956) British physicians smoking- link cancer study.  Notice that a high relative risk is achieved only when heavy smokers are compared to non-smokers.  Comparing heavy smokers to light or moderate smokers gives risks a lot like what we see in EMF studies.  I submit that the EMF equivalent of non-smoker does not exist in the industrialized world.


In 1995, I was briefly involved in a law suit in which office workers claimed that they had developed cancers due to strong magnetic fields in the office where they worked (tables 15, 16) (Milham, 1996).  These workers were commercial real estate brokers working in a first floor office above a three transformer 12 kV substation in the basement directly underneath their office.  Magnetic fields were as high as 190 mG at floor level and 90 mG four feet above the floor.  Experts for the employer, building owner and electric utility had done a cohort study of the cancer incidence in these workers and concluded that the cancer incidence in this work force was not increased.

However, an analysis of cancer incidence by duration of employment shows a positive trend of cancer incidence with duration of employment and an odds ratio of 15.1 for workers employed over 5 years.  Since many buildings have a similar electrical infrastructure, I predict that hundreds of thousands of office workers are similarly exposed.


Last year, I purchased a gauss meter which measured down to 25 Hz or below. It registered high magnetic fields in my car, and these fields increased with speed.  Last June at the World Congress for Electricity and Magnetism in Biology and Medicine, held in Bologna, a paper was presented which explains the high magnetic fields in my car (Vedholm and Hamnerius, 1997). It seems that permanently magnetized rotating metal parts like the steel belts in tires generate magnetic fields inside the passenger compartment. A little geometry suggests that the fields generated by the tire cords will be of a frequency too low to be detected by most of the meters used in the EMF-cancer studies.  Power lineman spend a lot more time riding around in their trucks than they spend working on energized lines.  In the electrical utility worker cancer studies, a very different exposure picture would emerge if lower frequencies were captured.

One other population with high exposure to magnetic fields are people who use tanning beds.  I have measured magnetic fields of 50-100 mG on these beds.  This is not surprising since these things contain closely packed fluorescent tubes, ballasts (transformers) and fans.  I wonder if the recent increase in malignant melanoma incidence could be related to the combination of ultraviolet light and magnetic field exposure which these beds generate?


Since the original Wertheimer-Leeper study in 1979, about 40 residential and 100 occupational EMF epidemiological studies have been published. Nearly 500 separate risk ratios have been published in these studies.  I find it interesting that for every risk ratio below 1, there are about 6 over 1.  A visual example of this is seen in fig. 3 (Keifets _et al_, 1995), which shows risk estimates for brain cancer from a meta analysis of 29 occupational studies.

In summary, I believe that there is ample evidence that EMF exposure is associated with increased cancer in humans.

Sam Milham..........

Thanks, Sam......

Roy Beavers (EMFguru)

Back to home page http://www.feb.se