A general site mainly for journalists.

Stewart Fist, journalist, columnist and film-maker.

    NOTE: In republishing this article in 2019 I need to point out that the problems that emerged with cellular mobile phones in the 1990s applied to GSM pulse-power mobiles of the European standard known as Groupe Speciale Mobile. This standard was very similar to the American TDMA (often called Digital AMPS) because both used a strobe-technique, correctly called "Time Division Multiple Access". This allowed a number of mobiles to share a single frequency, with each taking a turn to transmit during its own time slot.

    The damage to DNA from these phones was discovered by Henry Lai and Narendra Singh at Washington University in the USA, and the Adelaide Hospital study was an attempt to replicate their work.

    The unconscionable two year delay before Telstra released the findings was caused by the cellphone industry agreeing to a new standard and completely changing over to a non-pulsing form which they called "G3" (trying to pretend in was a third-generation of GSM). It used a version of the CDMA transmission system and at a much lower peak-power level, but still used the old GSM call-handling and messaging techniques.   SF

Are Mobile Phones Good For You?

The Work at Adelaide Hospital...

By Stewart Fist

Source:- Electronics Australia Magazine March 1999

Back in early 1993 there was a lot of discussion in the Australian press about the potential adverse health effects of GSM cellular mobile phones. Eventually a research project was undertaken by a group of cancer scientists working through Adelaide hospital.

Actually, there were two parallel research projects - one funded by the electricity supply authority of Australia (ESAA) to evaluate whether exposure to mains-power electric fields could cause or promote cancers (mainly leukemia), and another funded by Telstra to look at the possibility that GSM digital cellular phones could do the same.

The concern about power lines stemmed from a few epidemiological reports which suggested that people (mainly kids) living under power lines had a higher rate of leukemia than normal. There were parallel fears that cellphone handsets could do the same.

Leukemia is a disease of the immune system which involves the lymphatic nodes and white blood cells, so the researchers decided to use a special breed of lymphoma-prone mice as sensitive (or rather, discriminatory) detectors of the effects. The studies were conducted over a period of 18 months, beginning late 1993 with 1400 mice.

The key to understanding and interpreting these experiments is to understand how meticulously the experiments are designed to avoid any possible extraneous influences.

The basic experimental protocol calls on the scientists to use two identical groups of mice, maintained in clean and identical environments. One group is exposed to the suspect toxic substance while the other is not exposed, and is called the 'control'.

Many decades of experience have convinced the biomedical research industry that is a must be extraordinarily careful to avoid extraneous influences. So a second requirement is that the research is to be conducted 'blind': no one able to influence the results should know which mice are exposed and which are not.

I don't think is is generally understood how meticulously planned these experiments really are ... or how costly such research can be.

In the case of the radio-frequency experiment there were 100 mice in each group. They were held five at a time in small plastic cages in two identical rooms which were shielded from external radio signals by a millimetre of aluminium sheeting.

For the technically-minded: Both rooms were equipped with vertical ground planes (2.5 x 2.2m) with a 1/4 wave monopole at the centre, and the exposed group received two 30-minute doses of RF (Radio Frequency energy) per day, with computer in control of both the time and duration. The radio signal was of the GSM (mobile phone) type pulsed (217Hz) 900 Mhz (microwave) signals with a pulse width of 0.6ms - calibrated so as to match normal human-head/handset power densities.

    [Note: this means that the handset against the head was generating 217 power pulses per second, each carrying a signal of 900MHz radio frequency. The base stations are quite different and less able to have biological impacts.]

Where the experimental design can be criticised is that it failed to place the antenna against the side of the mouse head. The mice had distributed absorption over the whole of their body, while human exposure is localised [in the ear during GSM times]. Both the mouse body-size and orientation will change the characteristics of the RF absorption, but thermal resonance effects weren't detected.

    [There were claims made by the industry that any increase in cancer rates could be due to microwave heating, rather than direct RF effects on the DNA. This said "NO".]

It can also be said that the exposed mice were in the far-field rather than in the near-field, and also that the mice were exposed to RF but not inductive effects, which, in cellphone, can be more significant in energy transfer than the antenna radiation.

    [It was claimed that, at distances over a wave-length (the far field) radio-signals act differently on biological tissue. Inductive effects are direct influences within any varying magnetic field. The cellphone industry was just looking for arguments to discount the significance.]

But no experimental design is perfect.

The animal house staff were never told which group was exposed and which was the control. Sympathetic animal-loving staff might treat the mice differently if they knew, which is why the antenna systems were duplicated. In the scientific literature, this is referred to as 'sham' exposure of the control group.

Whenever animals died or developed tumours to the point of discomfort they were killed [humanely to prevent fear complications] and sent for autopsy. Dr Alan Harris, who performed the microscopic examination of the corpses at the Walter Eliza Institute in Melbourne only had identification numbers attached to the mice: this ensured that he had no knowledge of whether a particular mouse had been exposed or not.

The first person in the research chain able to associate the incidence of tumours with either of the two groups [exposed or control], was the statistician Val Gebski at Sydney University. He began to notice an excessive rise in tumour rates among the exposed group from about 9 months, and it continued to rise for the duration of the 18-month experiment.

Normally T-cell (Thymic) lymphomas are the type most often found among these mice. T-cells are the more mature disease-fighting cells, but here the exposed mice showed a doubling of B-cell (Basal cell) lymphoma rates also. This suggested that the DNA in the immune system was becoming disrupted during early cell growth stages.

Absolute numbers don't mean much here because these were were lymphoma-prone mice. The significance lies in the ratio between the control and the exposed group, and secondly, in the type of lymphomas being created or promoted. This kind of research is also often complicated (to some degree) by mice dying from causes not related to the exposure environment; however nothing of significance was detected in this study.

Eventually, by the time the project ended, the rate of tumours among the exposed group was calculated to be 2.4-times that of the un-exposed group.

However statistical adjustments are necessary to allow for the likelihood (or not) that such a difference could have arisen by chance alone. This is always a possibility when dealing with human or animal genetic variability, no matter how well the research has been conducted.

For reasons too complex to go in here, the final determination from this study was that the the RF exposure at least doubled the lymphoma rate, and this was established to a very high degree of certainty (99.9%).

Such a confidence figure is about 10 times more 'certain' than scientific results which would be classified as 'significant' (99%) and 50 times a rate which is normally accepted as 'proven' (80%).

The researchers state modestly that their findings are "highly significant", and most of the genuine biomedical research industry appears to have accepted this evaluation. Of course, the scientists also issued the standard warning against extrapolating such animal findings to humans, and they also suggested that the results required independent replication to be 100% sure.

Both qualifications, of course, are true. But there is also the element of the "bleedin' obvious" about these ritualistic observations. No research of this kind is ever considered in isolation by the scientific community, and at this time it was well known that there are many other research findings with similar conclusions.

Nor can the non-human associations be lightly dismissed. These findings were highly significant in terms of DNA disruptions and, as one of the scientists admitted to me, "DNA is DNA". Mice DNA and human DNA have almost identical characteristics, even though the outcomes of individual gene disruption might be different.

Summarily Dismissed...

What annoys me most about the reporting of the results in Australia was the way in which human consequences were so summarily dismissed in public. Meanwhile, around the world within the scientific establishment, they were treated as immensely important.

You don't need to be a scientist to point out that this $1 million research program was not designed to ensure the health of mice. There were, surely, some human implications ... yet any such suggestions was dismissed by Telstra and the cellphone industry.

When reporting positive and negative results, the scientists needs to be consistent and even-handed. If a result isn't applicable to humans when it is positive, then it also doesn't apply when it is negative or inconsequential. Yet ....

In the parallel ESAA power-line study, far less convincing associations were found between 50Hz (power line) exposure and lymphoma/DNA, to the point where the scientists were able to dismiss the findings as 'not proven'. (This is probably the case since the experiment was inconclusive.)

    [In fact this half of the research was cut short because of some disease problems in the mice, but the results they did generate were treated as being statistically acceptable, and to have proven safety.]

The parties involved in the power-line study then actively sought out television crews and gave interviews on camera and to the newspapers announcing that the results should "Quell fears that power lines might cause childhood leukemia".

They didn't qualify the claims by saying that such supposedly favourable results (a) required replication before they were accepted, and (b) that these were single isolated findings, which (c) are only applicable to lymphoma-prone mice!

The GSM study which found a potential problem was treated quite differently. When the results of the GSM RF study was released in May '97 (an unconscionable TWO YEARS after completion) the US Cellular Telephone Industry Association (CTIA) jumped into print (mysteriously a day before the research results were released in Australia) with a statement dismissing the research because it used "genetically engineered mice" rather than normal mice.

In the process, of course, they were effectively dismissing the viability of billions of dollars of global research medical research each year. Specially bred mice are constantly used for just this purpose, and for the same reason - they are best able to act as detectors for minute environmental changes. Detection of potential long-term harmful effects must use animals with increased sensitivity if they are to see results within the two-year lifetime of a mouse.

And clearly, as every radio engineer knows, if you want to reduce background noise, then you must select the most discriminating detection system available to you.

    [Meaning only that the identification of the incidence of disease is always confused by the range of other disease over the years. We couldn't afford to wait a human life-time to find out whether these cellphones could be harmful; and even now, we don't know with any degree of certainty. Nothing was 'proved' in the popular sense, it was only established as 'possible' or 'probable': so the abandonment of GSM and American TDMA was precautionary only.]

The CTIA also quoted Dr Basten as saying "Mice and humans both absorb energy from these fields differently, so we cannot conclude from this single study that humans have an increased risk of cancer from the use of digital mobile phones."

In fact, humans in the near-field of a GSM phone held to the ear, absorb between four and ten times more energy than that experienced by the mice, because there are also induction effects as well as radiation. Up to 80% of the energy generated by a cellphone transfers to the human head by induction. Induction may or may not play a part in DNA disruption.


Stewart Fist,
70 Middle Harbour Road,
Lindfield NSW 2070

    Contact: editor@electric-words.com.au