<< Return to previous page | Senate Environment, Communications, Information Technology and the Arts Committee

Navigation: Previous Page | Contents | Next Page

Chapter 2 - Research on the health effects of electromagnetic radiation

Introduction

2.1        While radio waves and other forms of electromagnetic energy have been in use for decades, the recent dramatic increase in the use of mobile phones, the visible proliferation of mobile phone towers and antennas and accompanying anecdotal and scientific studies showing biological and possibly health effects associated with these structures, have led to increased public concern about the safety of mobile phones and other telecommunications technologies.  Many studies have been conducted to examine the relationship between radiofrequency radiation and biological and health effects, however to date, the results have been inconclusive.

2.2        Several recent expert reviews provide an analysis of the relevant scientific literature, with last year’s UK Stewart Report considered the most comprehensive so far.  Other reviews include those conducted by the CSIRO in 1994, the European Commission in 1996, the International Commission on Non-Ionizing Radiation Protection (ICNIRP) in 1996 and 1998, the World Health Organization in 1998, and the Royal Society of Canada and the UK House of Commons Select Committee on Science and Technology in 1999.  The conclusions and recommendations from these reviews will be referred to throughout this chapter.

2.3        The Committee received submissions and evidence from a number of scientists and health professionals, as well as community organisations and individuals.  Some claimed that there is ample evidence of biological and/or adverse health effects associated with non-thermal levels of exposure to electromagnetic radiation, while others concluded that no clear relationship has been established.

2.4        This chapter provides a summary of the scientific research covered by recent major reviews, as part of a discussion of the evidence presented to this Committee based on the observations and research of witnesses and submitters to this inquiry.  It concludes with an overview of current Australian and international research in this field.

Exposure to electromagnetic radiation – if biological effects are shown, what are the health implications?

2.5        Exposure to non-ionising radiation, at exposure levels sufficient to cause heating above 1ºC, is known to cause adverse health effects.[1]  Knowledge about and acceptance of the effects of non-thermal exposure to electromagnetic radiation remains limited and contentious.

2.6                  As stated earlier, a number of expert reviews of the literature have been conducted, which have drawn the following conclusions in relation to the health effects of non-ionising radiation, including radiofrequency radiation:

CSIRO, 1994[2]

This report concluded that there was insufficient reliable scientific evidence on which to base sound conclusions about safety of radio frequency (RF) exposures in telecommunications.  It stated that ‘because of its equivocal nature, the data base for RF emissions has limited value.  It may be dangerous to make general statements on safety based on lack of evidence of harmful effects when so little relevant research has been carried out’.

International Commission on Non-ionizing Radiation Protection (ICNIRP), 1996[3]

Most of the established biological effects of exposure to RF fields are consistent with responses to induced heating resulting in rises in tissue or body temperature of greater than 1°C ...  In contrast, non-thermal effects are not well established and currently do not form a scientifically acceptable basis for restricting human exposure for frequencies used by hand-held radio telephones and base stations.

European Commission, 1996[4]

Overall, the existing scientific literature encompassing toxicology, epidemiology and other data relevant to risk assessment, while providing useful information, provides no convincing evidence that radiotelephones[5] pose a long-term public health hazard.

World Health Organization, 1998[6]

... no known health hazards were associated with exposure to RF sources emitting fields too low to cause a significant temperature rise in tissue.

ICNIRP, 1998[7]

Epidemiological studies on exposed workers and the general public have shown no major health effects associated with typical exposure environments.  This is consistent with the results of laboratory research on cellular and animal models, which have demonstrated neither teratogenic[8] nor carcinogenic effects of exposure to athermal levels of high-frequency.

Royal Society of Canada, 1999

The Royal Society Expert Panel on Radiofrequency Fields noted that there were ‘a number of observed biological effects of exposure of cells or animals to non-thermal levels of exposure to RF fields’, but had found ‘no evidence of documented health effects in animals or humans’ relating to this exposure.  However, it also expressed the view that ‘many of the studies in humans and animals addressing the potential for adverse health effects do not have sufficient power to rule out completely any possibility of such effects existing’.[9]

UK Independent Group on Mobile Phones Report (Stewart Report), 2000

The Stewart Report (Mobile Phones and Health) noted that while there has been little research into the safety of mobile phone and base station emissions, there was some peer-reviewed literature from human and animal studies and substantial non-peer-reviewed information, which refer to the potential health effects caused by exposure to RF radiation from mobile phone technology.  It concluded that the balance of evidence suggests that exposure to radiofrequency radiation below National Radiological Protection Board (NRPB)[10] and International Commission on Non-Ionizing Radiation Protection (ICNIRP) guidelines ‘do not cause adverse health effects to the general population’, but noted that ‘[t]here is now scientific evidence ... which suggests that there may be biological effects occurring at exposure levels below these guidelines’.  The Stewart Report concluded that ‘it is not possible at present to say that exposure to RF radiation ... is totally without potential adverse health effects, and that the gaps in knowledge are sufficient to justify a precautionary approach’.[11]

2.7        Animal studies have provided evidence of significant responses to radiofrequency radiation, including changes in temperature regulation, endocrine function, cardiovascular function, immune response, nervous system activity, and behaviour; however, the significance of biological responses at low exposure levels and their relationship to health effects are either not agreed with or not well understood.

2.8        The Telstra Repacholi et al study in Adelaide is one of those which has shown a significant increase in cancer incidence for mice genetically predisposed to lymphoma, and this study is currently being ‘confirmed’ and is referred to later.

2.9        The Committee was informed that a growing body of research provides evidence of biological effects.  This was the conclusion of the Royal Society of Canada Report, which said:

It is clear to the panel that there are a number of observed biological effects of exposure of cells or animals to non-thermal levels of exposure to RF fields.  These observed biological effects meet the common standards for scientific observation in that the experiments were well-designed, had appropriate positive and/or negative controls, contained valid RF exposure parameters, included appropriate statistical evaluation of the significance of the data, and have been observed to occur by more than one investigator ...[12]

2.10      Despite this, the Australian Communications Authority stated that ‘the evidence for production of harmful biological effects at relatively low levels of exposure (that is, field intensities lower than those that would produce measurable heating) is ambiguous and unproven.[13]

2.11      The World Health Organization (WHO) draws a distinction between effects on health, which it defines as ‘the state of complete physical, mental and social well-being and not merely the absence of disease or infirmity’[14] and biological effects which are ‘a physiological response that may or may not be perceptible to the exposed organism’.[15]  In his paper on exposure to low level radiofrequency fields, Dr Michael Repacholi, Coordinator, Occupational and Environmental Health, WHO, stated:

Biological systems respond to many stimuli as part of the normal process of living.  Such responses are examples of biological effects.  It is questionable whether reported ‘effects’, even if substantiated, can be considered to represent evidence of a hazard simply because the significance of the effect for the organism is not understood.[16]

2.12      Professor Litovitz, Professor Emeritus of Physics at the Catholic University of America, said on the question of whether or not electromagnetic fields caused health effects:

If they cause biologic effects, there is the possibility – not necessarily, but there is the possibility – that there will be health effects.  A biologic effect does not mean a health effect, but you cannot get a health effect without a biologic effect.[17]

2.13      Approaches to interpreting experimental results and determining when a biological response should be considered to constitute a health hazard include:

• any field-induced response is undesirable and should be avoided;
• exposure should be avoided if a physiological response in an organism is measurable; and
• where no discomfort or pain is experienced, the stimulus producing a response should be considered harmless.[18]

2.14      To establish that a biological response has health implications, Dr Repacholi says a number of conditions need to be satisfied, including determining whether the biological or psychological changes are reversible, whether effects are additive, or whether there are adequate compensation mechanisms to respond to the effects.[19]  Dr Repacholi offered the view that where dose-response relationships have not been established, it is difficult to extrapolate results between different frequency ranges and exposure levels, making it important to repeat experiments at different exposures.[20]  Dose assessment is also important in epidemiological and human studies, because of differences between ‘near field’ and ‘far field’ exposure.[21]

The role of epidemiology, in vitro and in vivo studies

2.15      When assessing the literature, it is worth noting that in vitro studies provide insights into the mechanisms underlying biological effects, whereas in vivo studies of animals and humans are considered to provide more convincing evidence of biological effects that may have implications for adverse health consequences for people.[22]  However, the most direct information on the risks of adverse human health effects come from epidemiological studies.  Dr Repacholi commented:

Most of the known human carcinogens were first identified as such by epidemiological studies; for this reason such evidence should not be taken lightly, even if the findings are unexpected or are inconsistent with other evidence ...  Epidemiological studies are important for monitoring public health impact of exposure, particularly from new technologies.[23]

2.16      This view is supported by medical practitioner and specialist in occupational medicine, Dr David Black, who noted that ‘[e]pidemiology is frequently misunderstood, and often wrongly criticised as being limited to showing associations but never proving causation’.[24]

2.17      In his submission, Dr Black describes some of the criteria of causation for epidemiological studies.  It also identifies the different types of evidence relevant to human health studies.  These range from experimental studies, which he says while providing some of the strongest evidence of cause and effect, could not be applied to human populations when the effect is harmful, and have limitations when the results from animal studies are applied to humans because of species differences; cohort and case-control studies, which compare groups which do and do not exhibit the effect, considered to be less precise than experimental studies and requiring a number of consistent studies before a conclusion can be drawn; ecological studies which are considered weaker than the two previously described because they study exposure between population groups rather than individuals, and are generally used for formulating or refining hypotheses for case-control or cohort studies; and finally, individual case studies, descriptive studies, anecdotal evidence etc, which are rarely proof of a definitive relationship but may suggest the need for further research.[25]

2.18      Dr Black also said the use of statistical significance to describe scientific results is also defined as indicating ‘the way the data has fallen but does not take into account reasons for this that are not related to true cause and effect, such as bias, confounding or statistical variation’, and therefore ‘statistical significance’ per se should not be confused with ‘causation’.[26]

2.19      Dr John Moulder, Professor of Radiation Oncology at the Medical College of Wisconsin, USA, when discussing cancer risk assessment, observed:

When the epidemiological evidence for an association between a physical agent and cancer is weak and/or the link is biophysically implausible, laboratory studies are critical for risk evaluation.  If there is strong cellular (in vitro) and/or animal (in vivo) evidence that an agent is carcinogenic, it can make even weak epidemiology evidence for an association credible.  Conversely, if appropriate laboratory studies are done and these studies fail to show any consistent evidence for carcinogenic activity, then we tend to dismiss weak epidemiological evidence, particularly if the association is biophysically implausible.[27]

Replication

2.20      One of the most contentious issues with regard to the way in which evidence from scientific studies is interpreted and afforded credibility is the question of replication, confirmation or verification.

2.21      The Mobile Manufacturers Forum argued:

... the results of any individual study cannot be considered sufficient to establish or refute a possible human health risk.  Individual studies must be validated and replicated before they can be relied on, and the determination of whether a potential health hazard exists requires a weight of evidence that evaluates all relevant, credible and valid data.[28]

2.22      Professor Mark Elwood, epidemiologist and public health expert, stated:

I want to emphasise only one methodological principle relating to most of these studies, and that is a general principle of epidemiology and, indeed, of science; that is, when you do a study which finds an unexpected and new finding which has not been reported before, it is very difficult within that study to assess whether that finding is meaningful or whether it is due to chance variation.  The only real way to assess it is to set up a second, independent study to test it.[29]

2.23             Dr Moulder argued that the failure to replicate results may be indicative of flaws in the original study:

... [the fact] that you cannot confirm and replicate it implies that there is something at least slightly wrong with the original – not necessarily totally wrong but something did not happen the way the authors think it happened.  At the first stage of an attempt to confirm, where you have somebody reporting something and somebody else saying they cannot confirm it, you really cannot necessarily believe either study ...  Sometimes it is not clear and you basically have to wait for more people to attempt to do it and you end up making what is basically a weight of evidence argument.[30]

2.24      Dr Neil Cherry from Lincoln University, New Zealand, reported in his submission that Dr Repacholi had informed an industry sponsored press conference that there was no evidence that GSM cellphones were hazardous to health:

At the conference he [Dr Repacholi] presented his paper on the Telstra funded project that showed that GSM cellphone radiation at quite low non-thermal levels, doubled the cancer in mice.  When challenged by the conference chairman, Dr Michael Kundi, Dr Repacholi said that a study is not evidence until it is replicated.  The conference rejected this.  A study is evidence.  Replication provides confirmation and establishment.[31]

2.25      Dr Cherry also pointed out that in replication work there can be unforseen variables:

It was shown in the calcium ion efflux work of Dr Blackman that biological effects in the laboratory can vary with the local magnetic fields, with temperature and with a number of other factors.[32]

2.26      Professor Litovitz advised the Committee:

There have been a large number of publications, and certainly over 100 have reported non-thermal biologic effects at exposure levels below that considered safe by most government standards.  If there have been that many publications, you can ask the question: why is there controversy?  If all of these papers are out there and every scientist is correct, why is there such a controversy and why is there so much argument?  The answer is that the papers do not all agree.  For almost every paper you see on biologic effect, you will see papers that say ‘I didn’t see anything.  I see a big effect, but I didn’t see anything.’

... So I ask myself: is this field of biomagnetics a junk science field?  Are these scientists out there who see effects at low levels all incompetent, or worse?  The answer is that lack of replication – that is to say, two scientists disagreeing – is not limited to bioelectric magnetics but rather it is a general problem in toxicity, it is a general problem in biology. ... Let us take drug X, whose name is not important.  We ask this question: does this drug induce deformed limbs in Norway rats?  The results are as follows.  In one set of experiments, those treated with the drug show 60 percent deformed limbs, those untreated eight per cent.  You have to conclude from that experiment that this drug is a teratogen, that is to say it causes abnormal embryos. ... This is not a story, this is a publication.

The difference between these experiments is that they were both using Norway rats, but there are all kinds of Norway rats – just like we are all people but we are genetically enormously different, and we are genetically enormously different in our susceptibility to various kinds of stress.  So even though you go out and buy these rats that does not mean you have identical rats.  The drug that was used in this experiment was called thalidomide, which, as you well know, was an enormous disaster.  It was a disaster because it was only studied in one strain and was not studied in the other.

The difference in genetic susceptibility of the test animals was never taken into account, and this experiment was only done after 10,000 children were born without limbs.  So this lack of replication does not mean that there is no scientific validity.  It means that science is complicated; it means that biology is complicated, and that the human system is complicated – and even rats are complicated.[33]

2.27      Professor Litovitz also cited an experiment in the US in which six laboratories with identical equipment tested chick embryos to see if magnetic fields caused abnormalities:

... When these six laboratories’ results came back, two said yes, two said absolutely no, and four said, ‘We might see something.’ ...  Six months later we made a measurement again and found no effect. ...  As we went through the three-year period, we found an enormous genetic compound in the response of chick embryos to electromagnetic fields.  ...  It is not that you [the laboratory] did something wrong; it is the genetics.  They were working with different genetic material.[34]

2.28      The Committee queried whether the Vernon-Roberts study (see Australian research below) could be considered a true replication of the 1997 Adelaide mouse study, given the modifications that have been made to the original methodology. Dr Repacholi, from the World Health Organization and member of the Adelaide mouse study team advised:

... in initial studies they may have done something that is not particularly helpful or there is a better way of doing it.  If the result is a true result it should still occur in the animal.  There is no reason to expect that you are still exposing the animal to radiofrequency fields using the same pulsing regimes, maybe different times, different orientations, but if there is going to be an effect it should still occur.  We were very careful in reviewing the follow-up study in Adelaide, and there is another study being done in Europe, to make sure that, yes, what was done in the original study is going to be either confirmed or not confirmed in these studies.[35]

2.29      In referring to the Adelaide mouse confirmation study, Dr Thomas Magnussen, CEO of the EMX Corporation, said:

... but there are significant differences between the two experiments.  For instance, Repacholi’s first experiment ran for 18 months.  The new one is going to run for 24 months.  The way the animals are exposed is quite different in the two experiments.  The genetics can never be the same.  When we are talking about biological experiments, it is virtually impossible to make a replication.[36]

2.30      The Consumers’ Telecommunications Network commented that there was insufficient evidence to conclude that there are no potential health risks associated with radiofrequency radiation.[37]

2.31      Dr Black said that in science it is impossible to prove a negative, and thus it will not be possible to claim that there are no health effects, only that the evidence suggests that such a scenario would be highly unlikely, as illustrated by the following statements:

... it is frequently stated by people who are concerned that the application of [radiofrequency] technology should not proceed until there is proof of the absence of any adverse effect.  The answer to this can only be that there will never be such proof about RF, or for that matter anything else ...

It is also equally true that it is theoretically impossible to provide absolute unarguable proof of an association.

The only conclusion which can be drawn from an understanding of the principles of epidemiology and of the assessment of scientific data is that whilst it is possible to prove an association with substantial and convincing certainty, it is impossible to prove an absence of an association in such a compelling way.[38]

2.32      Before outlining the research that is currently under-way both in Australia and overseas into electromagnetic radiation and its effects as it relates to telecommunications equipment, this section summarises what is known so far about the biological and health effects of electromagnetic radiation.

2.33      Expert reviews referred to at the beginning of this chapter have relied upon existing literature and a number of witnesses have concluded from scientific abstracts that there are potential health effects of EMR.

2.34      Mr Stewart Fist, journalist, claims to have the largest website collection of abstracts of scientific research publications and says that about 60 per cent of them show effects from non-ionising radiation.[39]

2.35      The World Health Organization website includes a database of current and published research into the biological and health effects of radiofrequency radiation.[40]

2.36      Some witnesses expressed the view that while this information is a valuable resource in understanding the science, it was an inadequate substitute for a working knowledge of the material.  The CSIRO’s submission to this inquiry commented on its own limitations in relying on research by others:

CSIRO is maintaining a watching brief, although it appreciates the limitations of attempting to evaluate research without the benefit of involvement and participation.  Independent, authoritative scientific information is provided in response to enquiries from Government and the community.

The absence of involvement in scientific research into biological effects of EMR is a recognised limitation in any assessment of the state of research.  It is only possible to fully understand the complexities of sophisticated biological procedures through experience gained from working at the bench.  Unfortunately, this level of expertise and understanding is lacking, or indeed absent, in many of the participants of committees or working groups that try to make assessments of the veracity of scientific research.[41]

Is the scientific evidence inconclusive?

2.37      The most recent expert reviews of the relevant electromagnetic radiation literature suggest that the results in this area are inconclusive.[42]

2.38      Industry submissions generally argued too that the science was inconclusive.  Hutchison Telecommunications, said in its submission:

... the world’s leading experts and key health advisory bodies state that there is no substantiated evidence to suggest a link between the use of mobile phones and long term public health risks, but we acknowledge there is public concern on this issue.[43]

2.39      Nokia Mobile Phones, Australia, said:

... a substantial amount of scientific research conducted all over the world over many years, demonstrates that radio signals within established safety levels emitted from mobile telephone[s] and their base stations present no adverse effects to human health.[44]

2.40      Motorola Australia, said:

... the scientific evidence does not demonstrate a risk to public health from wireless phones.[45]

2.41      In his submission, Mr Neil Boucher, said:

Most of the ‘research’ that has been carried out on the health effects of electromagnetism are top down studies.  That is people are assembled, with largely medical and statistical qualifications (and usually with little or no knowledge of electromagnetism itself), to look for epidemiological evidence of some health effect.  The fact that nothing conclusive has been found to date testifies both to the relative insignificance of any effect (if it exists) and to the futility of the methods employed.[46]

2.42      The Australian Communications Authority (ACA) submitted that radiofrequency devices that operate in accordance with recognised human exposure standards do not pose a health risk.[47]

2.43      The Committee notes the observations in the Stewart Report:

We were struck by certain inconsistencies and inadequacies in the scientific literature on the biological effects of RF radiation.  Many studies in this field have been exploratory and preliminary in nature, and claims of effects have sometimes been based on single experiments rather than a consistent series of hypothesis-driven investigations.  In some cases, study design and statistical analysis have been inadequate, and apparent effects may have been artefactual or due to random variation.  Indeed, the field is troubled by failures to replicate previous studies and by a lack of theoretical explanation of some effects that have been claimed.  There may also be biases arising from selective publication and non-publication of results. 

Finally, even for effects that appear to be well substantiated, the biological significance and the implications for health are often unclear.[48]

2.44      Not all witnesses were of the view that the evidence was inconclusive.  Dr Neil Cherry told the Committee that his work in preparing for a tribunal hearing for the first mobile phone base station in NZ in 1995 had led him to examine epidemiological and biological research from around the world:

I was very surprised there is so much published evidence in reputable, peer review journals that has not been sighted, summarised or integrated.  The more I received the more solid the evidence seemed to be and the more consistent it seemed to be.  And so when I heard people saying that the evidence was weak and inconsistent, I decided I should debate this with people and go to conferences and talk to them about it. ... This culminated, I believe, in a climax last year at the conference at the European Parliament where I was asked to look particularly at low level effects and epidemiological studies with those response relationships of low level effects.  ...  Over 20 studies show that radiofrequency microwave radiation damages the genes, damages the chromosomes, damages the DNA, and therefore indicates genotoxicity.  I am also aware that many studies only use small samples – they are epidemiological studies or laboratory samples.  They find elevated levels but they are not specifically significant and they are often described as showing no effects.  But I have supplied with my evidence a summary of brain tumour studies, and I have characterised them as studies showing elevated effects, studies showing significantly elevated effects and studies showing dose response effects.  And that is a classical way, I believe, at looking at the evidence trail and asking: was it elevated, was it significantly elevated and have we found dose response elevation? ...

... Following those principles, I come to totally different conclusions than Dr Moulder, Dr Black, Dr Elwood and Dr Repacholi.[49]

2.45      Mrs McLean of Electromagnetic Radiation Alliance of Australia (EMRAA), said that many studies are showing a range of effects, including brain tumours, leukaemia, heart problems, neurological problems, neuro-degenerative diseases, breast cancer and affects on the immune system, as well as affecting melatonin levels, enzymes, hormones, genes and signal transduction in cells[50].  These are discussed later in this chapter.

Anecdotal and non-peer-reviewed evidence

2.46      A number of submissions to this inquiry referred anecdotally to cases of brain tumours,[51] headaches,[52] hyperactivity in children and nausea,[53] skin growths protruding from the ear against which the mobile phone was held,[54] chronic fatigue,[55] nose bleeds,[56] and other health effects,[57] which they linked to mobile phone use.

2.47      Submissions also noted that expert panels, such as the Independent Expert Group on Mobile Phones (the Stewart Group), had been presented with anecdotal evidence of adverse health effects from mobile phones and their base stations, which were claimed to be related to non-thermal effects of radiofrequency radiation.[58]  Reference was also made to reports of ‘microwave sickness’ from mobile phones, including headaches, fatigue, impotence, blood pressure changes, chest pain and sleep disturbance.[59]  One submission raised the possibility of a link between legionnaires disease outbreaks with the presence of mobile phone towers and high voltage power lines in the vicinity of cooling towers.[60] 

2.48      The Committee notes the conclusions of the Royal Society of Canada Report:

Headache and fatigue are nonspecific symptoms. ... Headache is not an indicator of ‘brain activity’ and in general headaches occur in the absence of structural abnormalities of either the brain or the blood-brain barrier. ...  Although there is need to consider the possibility of [microwave-induced] symptoms such as headache and fatigue, existing data do not support the conclusion that [microwave fields] can induce headaches.[61]

The panel did not find persuasive evidence of the existence of radiofrequency radiation sickness syndrome, however, some individuals may be able to sense when they are exposed to radiofrequency fields.[62]

2.49      The Report recommended further research into this area.

2.50      While the EMR Safety Network International argued that anecdotal evidence should be heeded,[63] Dr Repacholi argued that this type of evidence is more valuable in establishing a hypothesis, rather than as proof of causal effect:

When reviewing the scientific literature, only independently confirmed effects can be considered when assessing health risk.  For establishing research needs, effects which have not been confirmed, but are possible and could have implications for health, should be considered because they may ultimately be established.[64]

2.51      The Committee notes that the Stewart Group included evidence from sources other than peer-reviewed scientific journals as part of its assessment of the potential health risks associated with exposure to radiofrequency fields.[65]  The Committee was advised that material that has not been peer-reviewed can suffer from several shortcomings, including deficiencies in methodology, analysis and conclusions.

2.52      Dr Repacholi said that the quality of peer review can vary and that the results of many studies need to be compared and evaluated before a conclusion can be drawn.[66]

2.53      Dr John Moulder mentioned difficulties in selecting suitable independent candidates to undertake peer review, particularly in small and highly specialised fields such as dosimetry:

What I do is look for people who are involved in the specific field but who have no direct connections, either positive or negative, with the authors of the study.  Sometimes that is in fact impossible. I will explain what I would do if I could not find the perfect person by taking the example of radiofrequency radiation and cancer in animals.  If everybody who is in that field is conflicted, I might look for someone who is an expert in RF dosimetry, even though they knew nothing about cancer, and then look for someone who was into carcinogenesis in animal models, even if they knew nothing about radiofrequency radiation, and then possibly back that up with a statistician who would not necessarily be familiar with either, but statistics is statistics.[67]

Publication and research bias

2.54      Dr David Black, in his submission, also drew the Committee’s attention to what he described as ‘publication bias’, whereby journals may prefer to publish a paper where the study has produced ‘novel’ results rather than one ‘simply reiterating a well accepted status quo’.  A similar bias was suggested in relation to difficulties in attracting funding for studies considered ‘likely to be simply reiterating well established fact’, and that these two biases need to be considered when undertaking a literature survey.[68]  The Committee also notes the comments of Dr Stan Barnett, CSIRO:

One of the biggest difficulties that we have in this particular area of research is that there are all sorts of biases in research generally.  That is a given.  You have to take adequate controls to make sure that you do not allow those biases – the experimental biases, the observer biases and the biases in the statistical analysis program that you use.  All of those things are biases which researchers are familiar with and which we understand ... but before you even start the research one of the biggest biases that exists generally is that of selection bias. ...  Selection bias is simply that the person who has the money ... has the resources and therefore has the ability to select, firstly, the type of research that they want to spend their money on; secondly, the facility where they would like to have it done ... and, thirdly, they can select whomever they wish to do that research, whether it is somebody who has the necessary experience in the area or somebody who has a high profile.  There may be issues other than the essential science that determine the selection of the research that is undertaken.[69]

2.55      Concerns raised about the difficulties in obtaining funding for replication studies are referred to in Chapter 3.

Biological effects

2.56      A number of studies have linked exposure to electromagnetic radiation with a range of biological and health conditions including: high blood pressure in humans; severe depression of the immunological and endocrinological responses of young chickens; increases in the permeability of the blood-brain barrier; calcium efflux from brain tissue; effects on the dopamine-opiate system considered to be involved in headaches; influences on epileptic activity; and increases in the mortality of chick embryos.  Studies have also found evidence of chromosome aberrations and increases in double and single strand DNA breakages, and increases in the promotion of certain cancers in genetically predisposed mice.[70]

2.57      Biological effects that have been specifically linked to radiofrequencies include changes to calcium ion mobility in the brains of cats and rabbits as well as isolated cells and tissues, changes to the proliferation rate of cells, alterations to enzyme activity, and affects on genes.[71]

The search for a mechanism

2.58      Various mechanisms have been proposed for the way in which radiofrequency fields interact with biological systems, generally involving the induction of movement of molecules.

2.59      Professor Philip Jennings, referred to ferrimagnetic material in human tissue with possible implications for the interaction between electromagnetic radiation, particularly extremely low frequencies, and biological systems.[72]

2.60      Professor Litovitz said:

There are those who believe that only heat can cause an effect and there are those who believe otherwise, whose experiments suggest that it takes only a signal to a cell to cause the cell to do something.  The cell has its own energy; you supply the trigger and the cell proceeds to produce enzymes and proteins, et cetera. ... Let us look at the example of garage door openers ... You are in your car and you press this and your garage door opens.  The question is:  can you believe that this supplied the energy for the garage door to open?  Was it this that supplied the energy for that motor to pick up the garage door?  We are saying no.  We are saying that this is a signal that turned on the energy to the motor.  That is the similarity, that is what athermal effects are all about: cells receive a signal and turn on the engine inside the cell which produces proteins, which produces enzymes necessary for survival.

We have studied in detail the target of the EMF and we now know the number of milliseconds that it takes the cell to be able to say there is a field there. ...  It is well known in biology that this information goes to a process called signal transduction on the surface of the cell or receptors.  They say something and send a signal to the nucleus, which proceeds to undergo various biochemical processes.  This takes seconds.[73]

... We are now working on a possible mechanism which relates EM field exposure to health effects.  We find that EM fields alter the levels of protective proteins.  It turns out that the major effort in my lab today is to use these non-thermal effects to protect against damage due to heart attacks, to treat cancer and to treat inflammation.  These non-thermal effects are remarkably useful, and will be useful in the next few years, in therapy.  The question is:  when are they therapeutic and when might they be harmful?

...  You have a protein that works, you come in with a electromagnetic field stressor, the protein is damaged and unfolded, nature produces protective proteins, goes in and refolds the protein and repairs the damage.  This is one of the most exciting discoveries in the past 30 years in medicine.  These protective proteins, these stress proteins, are being studied by almost every pharmaceutical company in the country because of their potential, because they are the basic repair mechanisms ... and we have found that EM fields can modify the amount of protective proteins that you have.  I say ‘we’ – there is a minimum of four, and I think it is five, labs that have replicated the concept that EM fields can affect protective proteins. ...[74]

There is a theory now that these protective proteins are related to Alzheimer’s and that a reduction in protective proteins means a greater probability of Alzheimer’s.  This is a theory which we have not tested, but there is data out there that appears to relate the incidence of Alzheimer’s to exposure to electromagnetic fields. ... We cannot necessarily say that there is a health effect, but we can say that mechanisms exist for potential health effects.[75]

2.61      Dr Peter French drew a link between evidence of the role of heat shock proteins in cancer and mobile phones:

In plain English, the point is that it has been demonstrated by several researchers that increasing the amount of heat shock proteins in cells results in the increased potential for developing tumours, increased stimulation of metastasis or spread of cancers, the direct development of cancer, de novo, and the decreased effectiveness of anti-cancer drugs.  Any one of these outcomes is obviously undesirable, but there is, within the heat shock protein and medical research literature, evidence for each of these statements.

... where are we with the mobile phone cancer link?  This is a summary of this part of my presentation.  A mobile phone user will experience energy from the radiation of the phone going into the brain.  That can induce some physiological effects, as has been published by Krause et al, but, importantly, it can potentially induce the heat shock response in the brain which can lead to the turning on of heat shock proteins.  For a single event that is fine, because that is the body responding defensively.  Normally it takes four to eight hours for the protein machinery to work after the protein machinery has been activated.  It takes from four to eight hours for the proteins to be secreted, to be made and then ultimately they disappear if they are not needed.  If you continually use a mobile phone, you can imagine that the heat shock proteins would be chronically induced, similar to the over-expression studies which have been described.  Continued regular mobile phone use can result in chronic expression of heat shock proteins, which can lead to – from those findings which are referenced there – increased metastasis, initiation and promotion of cancer and resistance to anti-cancer drugs.

I am not saying mobile phones cause cancer.  I am saying that this is a pathway – which is founded on solid, peer reviewed international science – which provides a mechanism whereby mobile phone radiation could lead to cancer.  Given that that is the case, then I would contend that some action is needed. If this is a possibility, then clearly research is needed to determine whether in fact heat shock proteins are being induced in the brains of mobile phone users; furthermore, we do not need to wait 30 years until that bottom line is confirmed. ...

... The link has been made by me.  Having said that, the mechanism by which microwaves may cause protein unfolding, leading to the heat shock response, has not yet been determined, and there are a couple of possibilities.  De Pomerai’s group says that there may be a resonance of the microwave field with the protein or with the water.  We have published, and it is in the written submission, a hypothesis paper in the Journal of Theoretical Biology which advances those two possibilities as well, for attributing low power as another stressor to activate the heat shock response.[76]

2.62      Associate Professor Olle Johansson from the Karolinska Institutet in Sweden, in discussing the health effects of visual display units, referred to the role of mast cells as a possible mechanism:

Here in Sweden, the problems around different types of electromagnetic devices arose with the introduction of radio in the twenties and thirties but it was much more evident in the late seventies.  When the PC explosion came, all the offices were turned into computer based systems and people were sitting all day long in front of visual display terminals of different types.  At the end of the seventies and at the beginning of the eighties, a growing number of people complained of different symptoms, especially from their face, on their neck, arms and hands after they had been sitting in front of these visual display terminals.  From the very beginning, it was not understood what was going on, but people were searching around in the working environment for different explanations.  Very soon, the ideas focused upon the radiation from the visual display terminals.  With respect to the symptoms, one could mention, for instance, skin problems, facial burning, redness, dry skin, facial heat, swelling, tingling sensations and even blisters.  Also, it was connected with feelings of fatigue and headaches, and memory losses were claimed et cetera.  Of course, as scientists we tried to understand the symptoms.

... In the last years, the focus has been much more on different high frequency devices, which of course include modern computer screens but also include light tubes of high frequency, different kinds of telecommunications systems, such as wireless DEC telephones, different radio alarm based systems and, of course, mobile telephones.  Parallel to this, a number of investigators – some among them having some very interesting data from Australia – have documented the results of experiments at the cellular and tissue level of different animals and humans which show the effects of, for instance, exposure to high frequency signals from mobile telephones. ...

... there are now more and more studies coming out pointing to possible mechanisms, from the cellular and molecular level, all the way up to more macroscopic events.  Our working hypothesis is very simple actually.  For instance, looking at human skin, both from patients claiming these kind of health problems and from normal healthy volunteers who have sat in front of visual display terminals, we see alterations in different cell types.  For instance, the histamine contained in mast cells is identical to what you would see – and it is reported also in the literature – from other irradiation damage sources: for instance, from sunrays, X-rays and radioactivity.  Our very simple and maybe naive working hypothesis that this irradiation damage is of a more long-term type compared to other more energetic irradiation damage.

Of course, the molecular cell biochemistry machinery has to be worked out in detail and this work is, of course, going on.  As I said before, in Australia, you have the research team around Peter French and his collaborators that has been studying these mast cells that have been irradiated using high frequency mobile telephone signals.  From their studies, it is evident that these cells are affected.  You then have to imagine what would happen if you have the same situation in a human being.[77]

2.63      Dr Cherry proposed another mechanism:

... The early studies show that oscillating signals interfere with the brain very significantly and can change the EEG and can change the calcium ions, and these change reaction times.  This is a classical physics approach of resonant absorption.  If a system can oscillate and an oscillating signal comes in, it can resonantly be absorbed.  It is what an aerial does, it is what a cell phone does, it is what is used in telecommunications, ...  It has been demonstrated in many laboratories that it actually does occur.[78]

2.64      But according to Dr John Moulder, in order to induce a biological change, ‘radio-frequency radiation must deposit enough energy to significantly alter some biological structure’.[79]

2.65      In noting some of the current hypotheses about possible biological interactions, Dr Repacholi stated:

These RF field-induced alterations, if they occur, could be anticipated to cause a wide variety of physiological changes in living cells that are only poorly understood at the present time.[80]

2.66      While observing that thermal effects may account for positive results, the Stewart Report considered that reports of epigenetic effects should be taken seriously and further research undertaken.[81]

2.67      The Committee notes that a number of studies cited in submissions as providing evidence of biological or adverse health effects relate to extremely low frequency (ELF) exposure.  Areas of similarity between the effects of radiofrequency radiation and extremely low frequencies include effects on calcium efflux, ODC[82] activity and behaviour associated with the opioid system.  The Royal Society of Canada Expert Panel suggested that ‘many of the efforts now underway to understand the mechanism associated with ELF effects could be used to investigate the mechanisms by which ELF-modulated RF fields elicit non-thermal effects’.[83]

2.68      The importance of determining the biological mechanism(s) responsible for any observed effects, particularly in relation to the setting of safety standards, was highlighted by the CSIRO:

... it is generally agreed by various expert panels that research on mechanisms of interaction is essential.  Without an understanding of how low energy RF fields cause these biological effects, it is difficult to establish safety limits particularly for non-thermal levels.[84]

How important is it to distinguish between frequencies?

2.69      Dr Moulder argued for the need to clearly distinguish between the evidence for adverse health effects from exposure to radiofrequency radiation as opposed to extremely low frequencies (ELF).  The applicability of ELF research to radiofrequency exposure was referred to by EMF South World Pty Ltd:

... observed bioeffects induced by mobile phone microwave radiation[85] are remarkably similar to bioeffects induced by power-line frequency EMF.[86]  This means that two decades of epidemiological data on power-line frequency EMF can be used in the debate on potential health effects of mobile phone radiation, on which there is virtually no epidemiological data.[87]

2.70      Dr Moulder advised that it was not appropriate to extrapolate the results of exposure to frequencies from different areas of the electromagnetic spectrum:

... the biophysics of the interaction is completely different. I do not want to be absolutist ...  But, in general, if you want to understand the biological effects of radiofrequency radiation, you use radiofrequency radiation.[88]

2.71      Dr Moulder later added:

In general ... most of the effects of radiofrequency radiation that we know of are not strongly dependent on frequency ...  But the bigger the jump you make, the less certain you can be ... if we finally concluded that radiofrequency radiation was safe enough for all practical purposes, that does not tell us whether powerline frequency is safe. ...  But, if you demonstrated that the frequencies used for FM and television were hazardous, then you would certainly worry about cell phone frequency. It would not prove it, but the closer together in frequency your information is, the more likely it is to be relevant.[89]

2.72      The Committee notes, however, the views expressed by Professor Philip Jennings, who stated:

Our society’s experience with ionising radiation should persuade us to take great care ... The original standard set for ionising radiation protection ... has proven to be quite inappropriate and as further research has been performed and evaluated the public limit has been reduced by nearly a factor of a thousand.  This could also happen with EMR.  We are still in the infancy of EMR research and we should learn from the mistakes we made with ionising radiation and introduce a principle of prudent avoidance or ALARA’.[90]

2.73      Professor Litovitz argued that:

The cell’s characteristic response to a mobile phone is the same as that to a power line.  This was beautiful for us, because it meant that all the data out there on powerline problems could be translated to the data on cell phone or mobile phone problems.  That is to say, you could put them together to try to understand what is going on.[91]

2.74      Many of the studies cited during this inquiry relate to extremely low frequency (primarily 50/60Hz) exposure, which report observed effects on the reproductive system, blood changes, ECG[92], heart rate, blood pressure and body temperature, melatonin and cancer.[93]  Studies have also been conducted into the health implications of exposure to radars, which operate at radiofrequencies ranging from 300 MHz to 15 GHz.

2.75      Submissions and evidence to this inquiry have referred to biological and health effects associated with powerlines, radio and television towers and video display units (see below); however, this inquiry is concerned with electromagnetic radiation associated with telecommunications technologies.

2.76      Dr Neil Cherry reported in his submission that:

Ten epidemiological studies have found significant miscarriage from EMR exposure across the spectrum from ELF, SW, to RF/MW.  The Scandinavian physiotherapist studies, Kallén et al. (1982) and Larsen et al. (1991) also found significant prematurity, congenital malformation, still birth and cot death.  Ouellet-Hellstrom and Stewart (1993) confirm the causal relationship with a highly significant dose-response relationship.[94]

2.77      Dr Cherry said it was also important to note that if an effect is seen with low frequency signals, such as an ELF 50 Hz or 60 Hz signal, or the Schumann Resonance ELF signals, then it is more likely and likely to be worse for modulated or pulsed RF/MW: 

This is because an ELF signal has a very long wavelength and generally passes easily right through the body.  Unless there is a resonant oscillator, such as for the Schumann Resonances, it induces quite small fields in the body.  On the other hand the RF/MW signals have wavelengths closer to the dimensions of bodies and body parts, they are more strongly absorbed in human bodies through the aerial effect.[95]

2.78      The Committee notes that the World Health Organization draws a distinction between radio and TV broadcasting and telecommunications facilities.  While for the most part the Committee has confined its comments to telecommunications technologies, in acknowledgment of concerns raised in relation to electromagnetic radiation generally, the Committee has digressed into other frequency ranges and technological applications in its review.

Observed biological and health effects of radiofrequency radiation

 Movement of substances across cell membranes

2.79      Studies have examined the effect of radiofrequency radiation on the movement of substances across cell membranes.  The role of calcium in the functioning of brain and other cells has prompted research into calcium movement in brain tissue.  While some studies have shown that low levels of RF exposure cause an increase in calcium efflux from brain tissue, according to the Stewart Report results are contradictory, and evidence of an amplitude modulated response at extremely low frequencies does not appear to be relevant to mobile phone technology, ‘where the amplitude modulation within the critical frequency band is very small’.[96]  The Stewart Report further concluded that ‘[i]f such effects occur as a result of exposure to mobile phones, their implications for cell function are unclear and no obvious health risk has been suggested. Nevertheless, as a precautionary measure, amplitude modulation around 16 Hz should be avoided, if possible, in future developments in signal coding’.[97]

Exciting neurons

2.80      The Stewart Report found evidence that exposure to high intensity radiofrequency fields, sufficient to result in a temperature rise in tissue, can reduce the excitability of neurons.  However, exposure at non-thermal levels does not appear to have an effect.[98]

2.81      It also reported that various studies have examined the potential of radiofrequency radiation to affect gene expression and produced inconsistent results.  While the well publicised study showing an increase in the lifecycle of nematodes may be suggestive of a non-thermal effect, the report said that there was little evidence to support the proposition that mobile phone radiation causes a stress response in mammalian cells.

ODC activation

2.82             The enzyme ornithine decarboxylase (ODC) plays a role in the synthesis of polyamines which can trigger DNA synthesis, cell growth and cell differentiation.  Activation of ODC has been related to the late, ‘promotional’ phase of cancer production, which is usually (but not always) correlated with an increase in the rate of cell division in the affected tissue.  Again, the results of studies examining the effects of radiofrequency radiation on ODC activity have been mixed.  Positive findings do not indicate an obvious pattern of dose-response or reveal a mechanism to explain the changes.  The Stewart Report noted that although all carcinogenic factors stimulate ODC, not all stimuli that increase ODC activity promote cancer, and said it was unlikely that the small increases observed from exposure to pulse-modulated radiofrequency fields could, on their own, have a tumour-promoting effect.[99]

2.83      The Royal Society of Canada Report states that:

... the lack of major [cell] proliferative response in the tissue of cell line following ELF exposure does not necessarily mean that ELF is incapable of serving as a tumour promoter, particularly if alterations in ODC activity are involved ....  It is possible that this small change in ODC activity brought about by ELF is unrelated to human cancer risk.[100]

2.84      The Report suggests that further research is warranted.

Heat-shock protein response

2.85      Dr Peter French indicated that the heat-shock protein response which is activated by external stressors such as chemicals, heavy metals, drugs and radiofrequency radiation has been shown in a separate study to be causally linked to cancer formation.  Other research submitted by Dr French suggested a link between RF exposure, cell changes and gene transduction.

Melatonin production

2.86      Submissions referred to studies that had shown that extremely low frequency (ELF) electromagnetic fields reduce melatonin production by the pineal gland, and the magnetic fields prevent melatonin from inhibiting the development of breast cancer.[101]  Circulating levels of this hormone have a strong circadian rhythm with melatonin levels peaking in humans at night.  Melatonin affects the mammalian reproductive system as well as other physiological and biochemical functions.[102]  While it may be hypothesised that similar effects may result from exposure to radiofrequency radiation, the Royal Society of Canada Report said that additional research is required to test the effects of RF radiation on pineal function, circulating melatonin levels, and the utilization of melatonin by target cells and tissues.[103]

2.87      Dr Cherry cited a study from Switzerland on the Schwarzenberg tower:  

... They were sampling melatonin before and after the tower was permanently turned off and they found a significant rise in melatonin after the tower was turned off.  They found a dose response increase in sleep disturbance.  When the tower was turned off experimentally, the sleep quality improved and melatonin rose in animals.[104]

2.88      The Stewart Report commented that part of the brain and the gland involved in melatonin production are further from the surface of the head in humans than in animals and concluded that:

... even if there were an effect on melatonin production in animals resulting from a direct interaction of fields within the brain, it would be much less likely to occur in people.[105]

2.89      In his submission, however, Dr Cherry claims that EMR reduces melatonin and enhances free radical activity in humans and that this is genotoxic, damaging the DNA and chromosomes, enhancing oncogene expression and transforming cells to neoplastic cells and causing cancer in exposed populations.

We have natural EMR-based communication systems in our brains, hearts, cell and bodies.  External natural and artificial EMR resonantly interacts with these communication systems altering hormone balances and damaging organs and cells.  The brain and the heart are especially sensitive because they mediate and regulate primary biological functions that are vital to life, thinking and heart beat, using EMR signals, the EEG and ECG.  When EMR interferes with the EEG this is communicated to the body by neurotransmitters and neurohormones, including the serotonin/melatonin system.  EMR reduces melatonin.  Melatonin is vital for the health of the Immune System, the Brain, The Heart and every cell, because it is the most potent naturally produced antioxidant.  It is a potent free radical scavenger that plays a vital protective role to protect the DNA in every cell.  Reduced melatonin causes cancer, miscarriage, heart disease, neurological diseases, viral and bacterial diseases, etc....[106]

2.90      In his submission, Dr Cherry says:

Cancer is a chronic disease problem from accumulated genetic cell damage.  Latencies for children and soft tissue cancers are as short as a few years, for most cancers they take 10 to 40 years to develop.  Cancer rates rise rapidly with age over 65 years because of the life-time of accumulated cell damage and the drastic reduction in melatonin that occurs after puberty.[107]

This shows how vulnerable very young children are because they have very low melatonin levels and undeveloped immune systems.  It also shows how reduced melatonin makes older people more vulnerable and much more prone to disease and cancer.[109]

2.91      Dr Cherry cited a large epidemiological study of female breast cancer over 24 states in the US which identified several organic solvents, including organochlorines, that significantly increased the incidence of breast cancer and which showed that radiofrequency fields were as dangerous as toxic chemicals and ionising radiation.[110]

2.92      Dr Cherry says this evidence is backed by more than 10 other studies showing that EMR across the spectrum increases breast cancer incidence and 15 studies showing reduced melatonin, including four with dose-response relationships:

... These are sufficient to classify a causal relationship between EMR and breast cancer, with melatonin reduction [a]s the biological mechanism.[112]

2.93      Dr Cherry also cited studies which found that melatonin reduction can be a cause of miscarriage and that microwaves significantly increased the incidence of miscarriage in a dose-response manner in the first trimester and that very young babies are sensitive to variations in the natural EMR at extremely low levels:

One of the most important single studies involved cot death (Sudden Infant Death Syndrome) in Ontario, Canada.  O’Connor and Persinger (1997) were investigating the GMA melatonin hypothesis by seeing if a melatonin-related syndrome (SIDS) varied with GMA.  They found that SIDS incidence significantly increased when GMA >30 nT and GMA <20 nT, - a homeostatic result.  This confirms that GMA causes illness and death in vulnerable people, babies, and involves melatonin homeostasis.[113]

Blood brain barrier

2.94      A number of studies have examined the potential of radiofrequency radiation to affect the permeability of the blood-brain barrier.[114]  While most studies have had negative results, one study did find an increased blood-brain permeability to albumin in RF irradiated rats.  While it has been suggested that blood-brain barrier breakdown following microwave radiation exposure may be due to thermal effects, some researchers have suggested that the disturbance may occur under ‘power window’ conditions where there may be a range of power intensities at which the barrier remains intact.[115]

2.95      The Stewart Report concluded that ‘[t]he available evidence for an effect of RF exposure on the blood-brain barrier is inconsistent and contradictory. Recent, well-conducted studies have not reported any effects’.[116]  In contrast, the Royal Society of Canada Report stated that effects on the blood-brain barrier permeability, calcium efflux and ODC activity ‘occur at exposures not thought to elicit thermal effects, [and] it is likely that these effects, even if they also occur at higher exposure levels, are non-thermal biological effects’.[117]

DNA

2.96      A number of studies also have examined the potential of radiofrequency fields to cause damage to DNA, and some have found no effects at non-thermal levels of exposure.  While radiofrequency fields do not have sufficient energy to break chemical bonds or directly cause DNA strand breaks, several studies have shown an increase in breakages at non-thermal levels of exposure and chromosomal aberrations.  Whilst these studies have not been replicated, they are ‘confirmed’ by the fact that they were similar and carried out in laboratories independent of each other.

2.97      According to Dr Cherry:

The first identified study that showed that pulsed RF radiation cause significant chromosome aberrations was Heller and Teixeira-Pinto (1959).    Garlic roots were exposed to 27 MHz pulsed at 80 to 180 Hz. for 5 mins.  They were examined 24 hrs later.  They concluded that this RF signal mimicked the chromosomal aberration produced by ionizing radiation and c-mitotic substances.  No increased temperature was observed. ...[118]

Garaj-Vrhovac et al. (1990) noted the differences and similarities between the mutagenicity of microwaves and VCM (vinyl chloride monomer).  They studied a group of workers who were exposed to 10 to 50 µW/cm² of radar produced microwaves.  Some were also exposed to about 5 ppm of VCM, a known carcinogen.  Exposure to each of these substances (microwaves and VCM) produced highly significant (p<0.01 to p<0.001) increases in Chromatid breaks, Chromosome breaks, acentric and dicentric breaks in human lymphocytes from blood taken from exposed workers.  The results were consistent across two assays, a micronucleus test and chromosome aberration assay.  Chromosome aberrations and micronuclei are significantly higher than the controls, (p<0.05, p<0.001, p<0.0001), for each of the exposure intensity.[119]

2.98      Dr Cherry also drew the Committee’s attention to studies done of staff in the US Embassy in Moscow that was chronically exposed to radar over a decade and found increased chromosome damage:

... I have found more than 30 studies showing chromosome damage in people exposed to radiofrequency microwave radiation.  This is far more than we have for benzine, which is a carcinogen.[120]

2.99      The results of genotoxic[121] studies were said by the Stewart Report to have been generally negative.  Dr Cherry says the studies he cited in his submission show very strong evidence of genotoxic effects from RF/MW exposures and notes that when chromosomes are damaged, one of the primary protective measures is for the immune system natural killer cells to eliminate the damaged cells.

2.100         The Committee notes that the general public ICNIRP guideline for microwaves above 2 GHz is 1 mW/cm², and for workers is 5 mW/cm². Dr Cherry pointed out that the Garaj-Vrhovac et al (1991) study of Chinese hamster cells in an isothermal exposure system showed that even at exposures 100 times below the public exposure guideline a 60 minute exposure kills 28 per cent of the cells and 30 minutes kills 8 per cent of the cells.

2.101         Garaj-Vrhovac (1999) also found that 12 workers occupationally exposed to microwaves had significantly increased chromosome damage as well as disturbances in the distribution of cells over the first, second and third mitotic divisions.

2.102         Dr Stan Barnett in commenting on the CSIRO’s unsuccessful proposals for NHMRC funding which was to look at cell response to radiation at specific periods in the cell division cycle, said:

... One of the biggest failings of all cellular studies is that, largely, they either use highly transformed cell lines which are very sensitive to almost anything, or they use cell lines which are general laboratory, fairly robust cells like lymphocytes.  Nobody bothers to try to synchronise the cells.  It is well known in radiation biology that cells respond to radiation at specific periods in the cell division cycle.  Our proposal was to use a fairly complex system which would allow us to use what we know as a radiation sensitive cell line and to synchronise it so that we only exposed it in G1, where we know – because of 30 years of background work – this particular cell is highly sensitive to radiation.  It is deficient in DNA repair enzymes, and we know that, if you are going to produce any kind of impairment of DNA repair which would be manifest as single strand breaks as per the Henry Lai study, this would be an opportunity to use the most sensitive available end point that we know of to test that scenario.[122]

2.103         It is also the case that studies have shown an increase in the number of cells with micronuclei, the formation of which are considered to reflect DNA damage, after exposure to RF radiation.  In spite of this, the Stewart Report concluded that implications for human health are unclear as normal tissue can also exhibit a high and variable incidence of micronuclei, making results difficult to interpret.[123]

2.104         Overall, while there have been numerous studies showing a range of biological effects, and while further research is required to satisfy the need to replicate positive results and to establish their implications for human health, the Committee Chair is persuaded that there is cause for concern.

Health effects discussed

2.105         Sleep disturbance, chronic fatigue, immune system impairment and learning difficulties have also been observed in radiofrequency exposed residential populations, and it has been argued that these effects are consistent with observed biological effects including calcium ion alteration and melatonin reduction. Various symptoms such as headaches, dizziness, feelings of discomfort, burning skin, which appear to be highly correlated with ‘warm sensations’ on and behind the ear against which the mobile phone is held, are described by Hocking (1998) and later observed in a survey of over 10,000 mobile phone users in Norway and Sweden.[124]  There have also been newspaper reports of more epileptic seizures in a school since mobile phone use has increased.[125]

Cancer

2.106         Although the development and promotion of cancer ranks in the general public’s mind as a real health risk associated with mobile phone and other telecommunications technologies, and indeed with other artificial sources of electromagnetic emissions, the scientific evidence for this association is said by many to be less definitive.

2.107         One area of contention is whether radiofrequency radiation initiates cancer or whether it may be implicated in the promotion of cancer.[126]  While there is general agreement that the energy in non-ionising radiation emitted by mobile telephones is unlikely to break chemical bonds, thereby inducing alterations in the genome,[127] Dr Cherry informed the Committee that in his view there is now sufficient evidence to show that EMR interacts and interferes with communication systems in our brains, hearts, cell and bodies through neurotransmitters and neurohormones, including the serotonin/melatonin system.

2.108         According to Dr Cherry, both through reducing melatonin and through enhancing free radical activity, EMR is genotoxic, damaging the DNA and chromosomes, enhancing oncogene expression and transforming cells to neoplastic cells and causing cancer in exposed populations.

2.109         The 1994 CSIRO report says:

For any biological effect to become significant the body’s homeostatic mechanism has to be overcome.  Homeostatis uses cellular communications via molecules and ions to control the three basic functions of cells: proliferation, differentiation, and activation.  Cancer promotion involves the disruption of cell-to-cell communication.[128]

2.110         There is more agreement and significant evidence to support non-ionising radiation as a cancer promoter.

2.111         Dr John Holt stated that cancer cells were three times as conductive of RF as non-cancer cells, and that non-ionising radiation rendered tumours more sensitive to ionising radiation.[129]

2.112         In its report of 1994, CSIRO said:

However, because a promoting agent requires high doses, must continue for long periods of time, and is reversible, it has been argued that the risks from a promoting agent are less than the risks from an initiating agent.[130]

2.113         Most epidemiological studies[131] that have been published focussed on RF exposure not directly related to cellular phones, and provide primarily indirect evidence from occupational or amateur radio operator radiofrequency exposure, with exposures being ‘more varied in dose, type of signal, and anatomical localisation than exposures from cellular telephones’.  These studies had variable findings.[132]

2.114         Professor Mark Elwood, epidemiologist, concluded:

... overall ... I do not see any consistency in relationships between cancer and radiofrequencies.  There are quite a lot of studies, so there are some positive results which require further assessment.  The studies are limited by lack of information on exposure, lack of control for other factors and, in some studies, biases in the data. ... Very often it is the weaker studies, with much smaller numbers and much weaker study designs, that tend to show unusual results, which therefore need testing.  So, overall, my conclusion is that there is no consistent evidence relating radiofrequency exposures and cancer in humans, in terms of current research.[133]

2.115         The information provided by these studies is considered, by most reviews, to be of limited value because of inherent selection biases and because they incorporate exposure conditions dissimilar to those experienced from cellular phone use.

2.116         The Stewart Report notes that studies of brain cancer have provided ‘inconsistent results’.[134]  The Report also refers to studies of other types of cancer, concluding ‘data on other types of cancer are more sparse and although some have suggested increased risks from RF exposure, their limitations are such that these findings should not be a cause for concern’.[135]  Several studies published since the Stewart Report support this conclusion.[136]

2.117         The recent occupational study of Motorola employees is considered to have dealt with some of the shortcomings of earlier studies.[137]  This extensive study of 195,775 Motorola employees between 1976 and 1996 found that for the nine per cent of employees that had experienced moderate to high levels of RF exposure, there was no increase in brain or lymphatic/haematopoietic[138] cancer mortality than either the general population or employees that had been exposed to lower levels of RF radiation.[139]

2.118         Professor Elwood, in his submission to the Committee, commented that the comparisons of employee mortality with general population mortality in this study were of limited value, but that the analyses of mortality between employees with different levels of exposure were more powerful.[140]  His analyses revealed no increased risk for cancers of the brain, all lymphatic and haemopoetic cancers, leukaemia, non-Hodgkin’s lymphoma and Hodgkin’s disease (although given the small numbers involved, a slight increase or decrease could not be discounted), nor for any general increased mortality risk.

2.119         Professor Elwood noted that an important finding of this study was the lack of association between degree of exposure and the incidence of the cancers studied, and that it also indicated no difference in overall specific risks between the men and women studied.[141]  However, he advised:

... even a study of this size cannot confidently exclude a modest increased risk of specific cancers which occur in relatively small numbers, although it can confidently exclude increases in total mortality or from major causes such as all cancers.[142]

2.120         In evidence to this Committee, Dr Peter French, Principal Scientific Officer, Centre for Immunology, St Vincent’s Hospital, Sydney, advised that there was no ‘definitive evidence’ for a link between mobile phone radiation and cancer.  However, he added that while there apparently was insufficient evidence on the surface, buried within the unsubstantiated assertions, fears, anecdotes and myriad of facts there were clues that point to a link between cancer and mobile phone emissions.[143]

2.121         Professor Elwood, on the other hand, concluded that based on an overall assessment of the research to date, there was ‘no consistent evidence relating radiofrequency exposures and cancer in humans’.

... the better studies ... are the ones that show no association. Very often it is the weaker studies, with much smaller numbers and much weaker study designs, that tend to show unusual results which therefore need testing. So, overall, my conclusion is that there is no consistent evidence relating radiofrequency exposures and cancer in humans, in terms of current research.[144]

2.122         Radiation oncologist, Dr John Moulder, in his submission to the Committee, concluded that:

... the epidemiological evidence for a causal association between cancer and exposure to radio-frequency radiation is weak to non-existent.[145]

... animal carcinogenesis studies conducted to date provide no replicated evidence that exposure of animals to radio-frequency radiation at non-thermal intensities causes or promotes cancer.[146]

...[o]verall, exposure of cells to radio-frequency radiation with an intensity that does not significantly raise cell temperature does not produce any consistent evidence for genotoxic or epigenetic activity.[147]

2.123         The interpretation of the scientific literature by some expert bodies, including the ICNIRP in the preparation of its exposure safety guidelines, has been criticised.[148]  Dr Cherry stated:

They decide that there is no evidence of genotoxicity but they do not cite any studies that have been published that do show that RF microwave damages chromosomes – and that is the classic test of genotoxicity...  Secondly, when I looked at two of their studies on cancer, they said that two recent studies do not show any significant effects.  I have those studies and they do show significant effects.[149]

2.124         Dr Barnett advised that the CSIRO had submitted two projects to the NHMRC, both of which were shortlisted but unsuccessful, related to the potential effects of radiofrequency radiation on DNA and cancer production:

One was an animal system, where we were looking at repeating, I believe, a very important research finding which has been largely ignored, which was finally published in 1992 by Chou and others.  That work was actually undertaken at the Brooks Air Force Base in San Antonio.  That study looked at simply exposing rats to 2450 megahertz of radiation throughout their lives.

When the data was analysed for tumour development in the exposed versus controlled animals, it turned out that, depending on how you chose to analyse the data, you got either a negative or a positive result.  The study had been largely referred to as providing a negative result.  It was only negative if you separated out each type of cancer and then looked at the difference in numbers for each type of cancer.  Clearly, because they only used a couple of hundred animals, when it was broken down into all the different types of cancer, the numbers that were being compared were extremely small, so the statistical power would be pretty poor.  When they compared the incidence of primary malignancies between the two groups there was a fourfold increase in the exposed group.[150]

2.125         Some witnesses to this inquiry referred to anecdotal evidence of people claiming, ‘with hindsight and when prompted’, to suffer from a range of cancer types resulting from chronic exposure to electromagnetic radiation.[151]  While it has been claimed that the involvement of electromagnetic emissions in the proliferation of cancer cells and possibly even as the cause of cancer is ‘beyond doubt’,[152] this view has not been supported by recent reviews on recently published papers.

2.126         The results of a case-control study conducted at five United States academic medical centres between 1994 and 1998 using a structured questionnaire, were published by Muscat et al in 2000.[153]  There were 469 men and women aged between 18 and 80 years with primary brain cancer, with 422 controls.  Details obtained from interviews included the number of years of use, minutes/hours of use per month, year of first use, phone manufacturer, reported average monthly bill, demographics, smoking history, alcohol consumption, exposure to power frequency fields, occupation and medical history.  No assessment was made of participants’ diet.

2.127         The researchers concluded that the study ‘shows no effect with short-term exposure to cellular telephones that operate on (primarily) analog signals’ and recommended that further research is undertaken to account for longer induction periods, particularly for slow-growing tumours, and the differences between analog and digital mobile phones.[154]

2.128         There was no association observed between the duration of cellular phone use and incidence of brain tumours.  In the cases examined, cerebral tumours occurred more frequently on the side of the head to which the phone had been held, however, for patients with temporal lobe cancer, the tumours occurred more frequently on the side opposite to that against which the phone was customarily held.  This contrasts with a Swedish study that found an association between the side of the head a brain tumour occurred and the side of phone use, although this study also did not find an overall association between cell phone use and the risk of brain cancer.[155]

2.129         The Committee received a confidential submission from a person suffering from a growth inside their skull.  The growth was adjacent to the mobile telephone antenna position.  This person was a heavy user of both analogue and digital mobile phones and believes that the excessive microwave radiation resulting from extremely heavy mobile phone use, most probably caused the malformation.[156] 

2.130         Dr Bruce Hocking undertook a survey of 40 people to categorize the types of symptoms exhibited by users of mobile phones.  The symptoms mainly affected the head and, for a few, the waist.  These symptoms included dull pain, an unpleasant warmth or heating, as well as ache, throb, sharp pain and pressure.  All respondents could distinguish the sensations from ordinary headache.  Most respondents felt the sensation less than five minutes after commencing the mobile phone call, but for others the sensation built up as the day progressed.  For some, the sensation lasted less than an hour after ceasing calls, for others it lasted till bed-time, and five respondents felt it the next day.[157]  In addition, Dr Hocking co-authored a paper[158] on a detailed study of a person who had enduring effects on the side of his head where he used his GSM mobile phone.  He experienced persistent unpleasant feelings lasting for more than a year and underwent extensive investigations by neurologists to find out if he had brain tumours or some other odd sort of neurological condition that could have been causing these problems, and nothing had been found.  Dr Hocking informed the Committee:

This is the first time that I am aware of that there has been a clear demonstration of a health effect in humans attributable to a mobile phone.  I agree it is only one case, and before you get too excited you would like to see more.  Nonetheless, I think it is a significant warning when you see it in context with the previous 40 cases that I was reporting that were getting similar sorts of symptoms that there is considerable likelihood that mobile phones, at the low levels of radiofrequency which they are operate on, are causing disturbances of neural function.

It is also considerable evidence of an athermal effect.  Given that mobile phones operate at low intensity – we are told by government, WHO and industry that mobile phones operate well within safety standards – that to produce this sort of effect we are having effects outside at low levels. [159]

2.131         Since 1994, researchers at the National Cancer Institute (NCI) in the United States have been conducting an adult brain tumour study which includes investigating a range of possible risk factors including: workplace exposures to chemical agents and electromagnetic fields; dietary factors; family history of tumours; genetic factors; home use of selected appliances; reproductive history and hormonal exposures; viruses; and medical and dental exposure to ionising radiation.  Cell phones, as another potential risk factor, were included in the research program in response to public concern about possible links between cellular phones and brain cancer.

2.132         Results from NCI research into cell phones and brain cancer were published early in 2001.  The case-control study of the relationship between cellular/mobile phone use and brain tumours was conducted in three hospitals in the United States between 1994 and 1998.  The study identified 782 patients in these hospitals who had glioma, meningioma or acoustic neuroma; from the same hospitals, 799 patients with non-malignant conditions, were used as the control group.

2.133         The study found no evidence that the risks of glioma, meningioma, acoustic neuroma, or all types of tumours together, was higher among people who used mobile phones for an hour or more a day or regularly for five or more years.  The researchers concluded that the results did not support the hypothesis that the use of mobile phones causes brain tumours, but stated that the results were ‘not sufficient to evaluate the risks among long term, heavy users and for potentially long induction periods’.[160]

2.134         The Committee acknowledges the difficulty of testing long term exposure and notes that the results of this study should be interpreted cautiously for the following reasons:

• widespread use of mobile phones is only a recent phenomenon, with few people in the United States having used mobile phones prior to the 1990s.  Only a small number of study participants had used a mobile phone for over five years.  Consequently, the study would not have been able to detect the risk of brain tumours after a long latency period;
• there was a reliance on interviews and the ability of participants to accurately recall mobile phone use rather than by objective measurements of exposure;
• the study was designed to assess the risk of all types of glioma, and the sample was too small to detect increased risk for glioma subtypes; and
• factors other than duration of use influenced the level of exposure of brain and nervous system tissue in the head to radiofrequency radiation, including distance from the base station, local topography and vegetation, whether the phone is used indoors or outdoors, the design of the phone, and the position of the phone and the antenna in relation to the head.[161]

2.135         In recognition of these limitations, the NCI advised that ‘it would be premature to conclude that use of hand-held cellular telephones does not cause tumors of the brain and nervous system’.[162]  Noting that analog phones were predominantly in use during the study period, contrary to recent years when phones have been increasingly based on digital technology, the NCI nevertheless offers the view that ‘there is no evidence at this time that cancer risk would differ for the two types of phones’.[163]

2.136         The results of a unique Danish study into the relationship between mobile phones and cancer were also published at the beginning of February 2001 in the Journal of the National Cancer Institute.[164]

2.137         A research team, headed by Dr Christoffer Johansen, conducted a retrospective cohort study[165] of cancer incidence in 420,095 Danish users of mobile phones between 1982 and 1995, using telephone subscription lists from two Danish mobile phone operating companies and the Danish Cancer Registry.  The team observed no significant difference between expected and observed incidence of cancers of the brain, nervous system or salivary gland, or of leukaemia.  Risks for these cancers did not vary by duration of cellular telephone use, time since first subscription, age at first subscription, or type of cellular phone used (analog or digital).  The study concluded that the results did not support the hypothesis that there is an association between the use of mobile phones and tumours of the brain, salivary gland, leukaemia or other cancers.[166]

2.138         Dr Johansen is reported as stating that ‘[i]f it is assumed that tumour promotion occurs close to the site of exposure, this finding provides additional evidence against a link between cellphone use and brain cancer’.  However, Dr Johansen indicated that the study results did not rule out a relationship between mobile phones and other health risks such as ringing noises in the head, migraine, headaches, other symptoms of the conditions associated with the central nervous system, Parkinson’s and Alzheimer’s diseases, various types of dementia, and skin diseases.[167]

2.139         Responding to the report, Australia’s Dr Bruce Armstrong, who is undertaking an epidemiological case-control study on the relationship between exposure to radiofrequency radiation and brain and other tumours in adults (see Australian research below), stated that while it was a ‘reassuring study’, it did not ‘give an ultimate assurance of a lack of a hazard’.  A shortcoming of the study was that only a small percentage of the mobile phone service subscribers had used their phones for more than seven years and this ‘raised questions on what links there were between cancer and long term mobile phone users’.[168]

2.140         The Committee Chair considers that there is sufficient doubt as to the association between radiofrequency and cancer to warrant further research before the public can be confident that any risks are adequately safeguarded against through current exposure standards. A discussion of the efficacy of current standards is discussed in Chapter 4.

Other effects

2.141         Although a dominant concern, cancer is only one of the health effects that has been attributed to radiofrequency exposure.  Electromagnetic emissions have also been implicated in many debilitating and/or serious health conditions, often immune system related, including allergies, repeated flu-like episodes and auto-immune diseases.[169]  There is also some evidence of genetic predisposition and age-related factors that may influence sensitivity to potential effects of RF radiation.[170]

2.142         While there have been reports of effects on the cardiovascular system from exposure to electromagnetic radiation, the Stewart Report concluded that ‘on the basis of published evidence, [there is] no basis for concern about effects of mobile phone use on the heart and circulation’.  People subject to chronic electromagnetic energy exposure have also reported suffering heart attacks and high blood pressure.[171]  The Stewart Group said, however, that while normal use of a mobile phone against the head is unlikely to have a direct effect on the human heart, influences on cardiovascular centres in the brainstem and on the carotid body, a body of tissue involved in the regulation of the heartbeat, were more conceivable, and further experimental work on human volunteers was warranted.  Observed effects were said to be attributable to thermal effects from acute exposures to radiofrequency radiation.[172]

2.143         Despite concerns about the possible effects of mobile phone use on cognitive functions such as memory, attention and concentration, relatively few laboratory studies have addressed this issue in people and, of those that have, all have investigated effects from acute rather than chronic exposure.  While exposure to radiofrequency radiation at levels which cause increases in core temperature of 1ºC lead to changes in performance of primates in well-learned tasks or other simple behaviour, on which the current standards are based, the Stewart Report said that results at non-thermal levels are inconsistent and recommended further research.[173]  Most studies which investigated exposure to low levels of RF radiation focussed on physiological measures of brain function, such as the electroencephalogram (EEG), rather than indices of cognitive performance per se.  The Stewart Report noted that the functional significance of different components of the normal, waking EEG is poorly understood, making it difficult to interpret results showing an influence of radiofrequency signals on the EEG.

2.144         This was said to be less of a concern with respect to EEG patterns associated with sleep as these are ‘well characterised and routinely used as indices of the different sleep stages that a typically healthy individual will move between during the night’.  There have been observations of a range of sleep-related disorders, including altered sleep patterns, circadian rhythm and reaction times, from naturally occurring electromagnetic radiation and short-wave radio exposure.[174]  However, these effects have been observed at lower frequencies than what are used for mobile phone transmissions.  In addition, the Stewart Report said that results of work on the neurotransmitter system, which is involved in regulation of emotion, memory and sleep, appear to show temperature-related effects.  To determine the extent to which the results of those studies can be extrapolated across the electromagnetic spectrum requires that these studies should be repeated using radiofrequencies.  The Stewart Report concluded that further research should be conducted in both areas.[175]

Alzheimer’s Disease

2.145         Reference was made to a study that linked exposure to electromagnetic fields with an increase in incidence in Alzheimer’s Disease (AD), which, it is hypothesised, is due to a chain reaction of cellular effects starting with interference to cellular calcium ion homeostasis.[176]  In its report, the Royal Society of Canada acknowledged this and another related hypothesis, but noted that studies aimed at testing these claims had used exposure to extremely low frequency fields (powerlines) rather than radiofrequency radiation.  In addition, methodological shortcomings limited the interpretation of the results.  The report concluded that ‘there are no convincing, reproducible data to suggest a relationship between AD and [microwave] exposure’.[177]

The Immune System

2.146         While it has been suggested that the evidence indicates that an increase in diseases connected with the immune system may be the long term effect of radiofrequency radiation from mobile phone use,[178] other reviews have been more cautious and point to the ambiguous nature of outcomes in this area of research.  The European Commission Report noted that there is a level of adaptability and redundancy built in to the immune system via self-regulation.[179]  Thermal effects that have elicited responses in the immune system have been found to be transitory, with levels returning to normal with the cessation of radiofrequency exposure.  The Stewart Report concluded that, given the inconsistent results from studies using low level radiofrequency radiation exposure, it was difficult to attribute any effects to exposure.[180]

The eyes

2.147         The Stewart Report also referred to various studies that had investigated the effects of high intensity pulsed RF fields on the eye.  Noting that these exposure levels were well above the specific absorption that could occur from the use of current mobile phones, it warned that possible adverse health effects in the eye may be associated with high peak-power pulsed radiofrequency fields.[181]

Reproductive problems

2.148         Some drugs and environmental hazards are known to have damaging effects on a developing embryo at exposure levels which are of little or no risk to the adult animal.  According to the Stewart Report, despite extensive research into the potential effects of radiofrequency fields on fertility and development, studies have failed to show any convincing evidence of effects.[182]  The Stewart Report referred to a 1993 study that showed an increased risk of miscarriage in physiotherapists who reported exposure during the first six months before or three months after pregnancy and a higher risk in those with more frequent exposure and concluded that there w