Health and Safety Special Report
Bulletin Article from Issue No.: 47 dated: 01 May 2000
HEALTH EFFECTS OF MOBILE PHONES
Dr Stan Barnett, Senior Principal Research Scientist, Commonwealth Scientific and Industrial Research Association
INTRODUCTION
The use of mobile telephones has increased at an astounding rate in recent years - in 1998 the mobile phone market grew by more than 50%.
There are currently approximately 25 million users of mobile telephones in the UK, representing almost 30% of the population, while the number of subscribers in the USA is currently estimated at around 85 million. The worldwide distribution of mobile telephone technology has created significant changes in the workplace and in lifestyle and behaviour patterns. Some of these behavioural changes have caused increased stress in the community in both users (constantly checking for messages, driving while using mobile phones) and their neighbours (privacy invaded,
disruption of meetings). From a public health perspective it is important to determine the risk of adverse consequences of increased use of mobile telephones and the exposure to radio-frequency radiation (RFR) throughout life.
Many workers now use mobile phones as part of their jobs and there is, therefore, concern both amongst employees and employers over the potential effects on health. This Special Report reviews the research that has been
conducted to determine the potential health effects from using mobile phones.
The advent of mobile telephony has brought about a unique situation where users deliberately place against the side of their head an electronic device with an antenna emitting electromagnetic radiation. The Federal Communications Commission in the USA regulates output of hand-held mobile phones and other communication devices to levels where the specific absorption rate (SAR) does not exceed 1.6 W/kg for any 1 gm of tissue when averaged over 30 min. For some mobile phone models, the calculated SAR in 1 gm of tissue in the head approaches this upper limit. Digital GSM telephony delivers radio-frequency energy in pulses (217 Hz repetition rate) so that the peak level within each pulse is considerably higher than the time-averaged field strength. Energy deposition in the human brain is
complex and continues as the subject of detailed computer analysis. The exposure levels received by the general community from mobile telephone base stations (cell sites) are thousands of times lower than individual exposures from mobile telephones, yet some people report sensitivity to such low levels of electromagnetic radiation (EMR). Whether or not this level of radiation is a threat to human health, the presence of telecommunications towers provides a visual reminder and can lead to stress-related health complaints in susceptible individuals. Any potential health impact related to the use of mobile telephones should be seriously considered. However, the implications of low level long-term exposure to such radio-frequency radiation (RFR) have not yet been adequately examined. The problem is exacerbated as new and different pulsing and frequency conditions are developed. Proper assessment of risk can only be achieved through a comprehensive programme of organised research that examines subtle biological responses to the type of pulsed microwave radiation that is currently used in digital mobile telephones. Because of the uncertainty of so-called non-thermal effects radiation, safety standards/guidelines are largely based on data from robust studies using high field strengths, ie where the SAR (4 W/kg) is sufficient as to heat the entire body by 1°C. Such thermal effects result in obvious alteration of normal physiology and behaviour which is relatively easy to
detect and, therefore, consistently repeatable. On the other hand, the detection of biological responses to low level exposure requires the design of sophisticated sensitive test systems. Their very sensitivity creates a greater likelihood of producing contradictory results. Such studies depend critically on the skill and experience of the researcher and it is essential that results are compared with careful verification in independent laboratories with equivalent expertise. Human studies can suffer limitations due to subjectivity bias and small numbers of subjects. When considering health implications of mobile telephones it is important to understand that the absence of data cannot be taken as evidence of no adverse effect, and that safety of RF exposure can never be proven. This opinion is supported by the conclusions of the Expert Group of the European Commission (McKinlay 1997) which stated: "No study or series of studies producing negative results can prove that an effect does not exist. However, an accumulation of well-performed studies producing negative results provides increasing confidence in the absence of a significant
adverse health effect".
Absence of demonstrated biological effects can simply result from inappropriate study selection or insensitive study design. On the other hand, it is important to avoid placing too much emphasis on response "trends" that are not statistically significant. Standard statistical tests are designed to exclude occasional differences from the norm that may happen by chance, ie are unrelated to the exposure conditions. Premature media reports of yet-to-be-published papers have recently implied a correlation between mobile telephone use and risk of brain tumours. However, the interpretation of "increased risk" is not supported by scientific data. Reports of such small apparent effects should be viewed with interest as a starting point for more detailed research to evaluate the actual scientific validity.
EFFECTS OF MOBILE TELEPHONE USE
Immediate effects of mobile telephones are easier to characterise and relate to usage patterns than a slowly developing disease. There have been various reports of users experiencing headaches or warming and itching
sensations on the side of the face. The most convincing data comes from occupational users of mobile phones where a correlation exists between duration/frequency of phone use and reported effects. As with over-use of any device, there is a risk of developing adverse health effects related to repetitive strain injuries. A recent anecdotal report cited a case of a doctor suffering a stroke as a result of his habit of cradling the phone between his head and shoulder. Whilst these effects may be unrelated to RF exposure, they demonstrate the need for awareness of risk.
An important immediate effect that is caused by RF emission from mobile telephones is that of electromagnetic interference (Clifford et al. 1994). The stronger peak electromagnetic fields in pulse-modulated digital mobile
telephones has been shown to interfere with normal working of electronic medical devices including drug infusion pumps, ECG monitors and telemetry systems, and cardiac pacemakers. This can have immediate health
consequences, and the use of mobile telephones is prohibited in susceptible regions in many hospitals.
EVIDENCE OF BIOEFFECTS
Examples of critical cell studies are those involving alteration to mammalian DNA or expression of stress proteins or specific genes. A study that attracted much attention (Lai and Singh 1995) reported an increase in single strand breaks in DNA in brains of rats following a brief (2 hr) exposure to pulsed RFR at a SAR level (0.6 W/kg), ie below that capable of producing significant temperature increase in biological tissues. A subsequent study, sponsored by Motorola Ltd, failed to reproduce the effect on single strand breaks using a different version of the DNA comet assay and using continuous wave rather than pulsed RFR (Malyapa et al. 1998). Further study by this group has reported on increased expression of a stress protein (Goswami et al. 1999). Fos mRNA proto-oncogene levels were
increased in C3H101/2 mouse embryo cells after exposure to a low level (0.6 W/kg) emitted as both continuous wave 836 MHz (FMCW) and as pulse modulated 848 MHz (CDMA). Other stress proteins were unaffected. The induction of stress proteins is an interesting area of research that may provide new information on the sensitivity at the cellular level to insults from various physical agents, including RF radiation. This expression of a specific gene supports the concept of individual sensitivity to RFR. Whilst the production of stress responses at the biochemical level may not necessarily extrapolate to a human health hazard, changes in DNA synthesis or repair can lead to the development of disease, such as cancer. The possibility of cancer developing from long term exposure to RFR is of
global concern. The scientific database remains inconclusive with both positive and negative effects reported from long term animal exposures. In fact, different interpretations have been placed on the same set of data from one study (Chou et al. 1992) in rats which were exposed throughout life to pulsed 2450 MHz. A statistically significant increase in the total number of spontaneous cancers was shown in RF-exposed compared to sham-irradiated controls. However, the data were negative when analysed by tumour type, there being no significant increase in incidence of each type of cancer.
The weight of evidence from animal studies is that cancer is not initiated by RFR. Therefore, it may be appropriate for relevant modern research to develop animal models that are sensitised in some way before being exposed
to RFR at pulsing conditions used in digital telecommunications. Two such examples are given, with different results. A carefully executed study exposed rats to pulsed fields according to the North American Digital Cellular standard, 836 MHz throughout life after injection of the carcinogen, n-ethyl-N-nitrosourea that induced tumours in the nervous
system (Adey et al. 1999). There was no increased development of tumours, however, a statistically significant reduction in tumour incidence in exposed animals is suggested by the authors as evidence of an non-thermal
biomolecular effect of RFR, possibly interacting with key growth-regulating enzyme systems.
The cancer debate was recently re-ignited by the publication of a significant finding from a study using Eu-Pim-1 transgenic mice which have a predisposition to develop lymphoblastic lymphomas. Exposure to RF from a
mobile telephone for an hour each day throughout life resulted in a significant (2.4 times) increase in the incidence of lymphomas (Repacholi et al. 1997) when compared with non-irradiated mice. While these results are interesting, it should be recognised that the experimental protocol was unconventional, largely untested and contained some significant limitations. The majority of the cancers were not the lymphoblastic lymphomas expected from the Pim-1 genetic manipulation. Consequently, this report has generated many questions which may not be answered by simple
repeat studies.
Transgenic animal models have been proposed as an alternative to whole-life animal studies, however the concept has not been approved by national toxicology programmes. During a Wireless Technologies Research (WTR)
workshop (Bioelectromagnetics Society Conference, Florida 1998) a panel of experts agreed that there was insufficient reliable data to justify developing transgenic models for studies on mobile phone safety. The major
problem is that while they are sensitive to effects on a specific gene the test is actually less sensitive for a wide range of unknown effects. In other words, one needs to know what specific cancer would respond to RFR before doing the study. It would, therefore, seem more appropriate to test for a specific brain tumour that might be associated with prolonged exposure from mobile phones.
Long-term animal studies are expensive and will, doubtlessly, consume the bulk of research funds leaving precious little funding for essential studies on mechanisms of interactions. Without fundamental scientific information on mechanisms it is difficult for science to progress in a sensibly structured manner. A conclusive argument for the cancer issue awaits the development of a plausible biological mechanism. In the absence of systematic study it is unlikely to be achieved for decades, if trends from past research are any indication.
The existence of non-thermal biological responses to RFR is now generally accepted. The EC expert committee conclusions (McKinlay 1997) stated: "a substantial body of data exists describing biological responses to
amplitude-modulated radiofrequency (including microwave) fields at SARs too low to involve any response to heating".
Interestingly, the strongest evidence for non-thermal biological effects comes from a therapeutic application rather than from directed research. An FDA-approved treatment for chronic psychophysiological insomnia uses ELF
modulated 27 MHz carrier wave RF. The treatment has been demonstrated by clinical evidence as being effective, although the maximum SAR used is below upper limits of safe exposure prescribed by the American National
Standards Institute/Institute of Electrical and Electronics Engineers or the International Commission on Non-Ionizing Radiation Protection (ICNIRP), as based on thermal substantiated biological effects. The effect, in this
case is beneficial and is unlikely to have adverse health consequences. Nevertheless, it demonstrates a sensitive biological endpoint that is not fully understood and which occurs below so-called safety thresholds. Research is required to understand the mechanisms of these low level effects and their possible relationship to hypersensitivity to EMR.
ESSENTIAL RESEARCH
Human adult cancer studies cannot give conclusive data for decades because of long latency periods for disease processes. Many published papers reporting an association of increased incidence of various cancers with RFR
have been criticised as being anecdotal in nature and limited by poor dosimetry information. There have also been some extraordinary claims of absence of adverse effects in populations that used data from subscribers with usage history of as little as one billing period. The obvious limitations of these studies can cause epidemiological findings to appear weak from a scientific perspective. On the other hand, laboratory studies can provide essential information on biological interaction with RFR without which it is impossible to predict probable health outcome with any
degree of confidence. The essential element of proper scientific method is the formation of a testable hypothesis. Sensitive cell responses can create the starting point for appropriate studies into potential disease processes. Fundamental scientific knowledge of cellular interactions is vital for science to progress in a sensibly structured manner.
INTERNATIONAL RESEARCH EFFORT
In view of the worldwide demand for international collaboration, what has been achieved to date? The Wireless Technology Research programme in the USA expended approximately $27 million of industry money over six years.
Its lack of support for bioeffects research has been widely criticised. Its major achievement seems to have been post-market surveillance and compilation of documents on EMI effects of mobile phones. After apparently
little interest in undertaking novel research, the WTR programme closed in 1998 and its web-based information is no longer accessible. By contrast, the Australian programme has a modest budget (approximately US$3million) which is expected to pay for a range of activities. Funding was obtained from the telecommunications industry. Most of the research funds will be consumed by a repeat of a study (Repacholi et al. 1997) on mobile phone radiation and cancer in transgenic mice. The research programme is administered by the National Health and Medical Research
Council and a strict condition is that research must be done in Australia, effectively preventing international collaboration. The EC Expert Group formed its recommendations in 1996 but the programme remains under
consideration. Successful research proposals will be funded to 50% of costs, and while international collaboration is encouraged no funds will be available for non-European partners. It is anticipated that the telecommunications industry will help provide the essential additional support for selected laboratories. It is difficult to see how selection
bias can truly be avoided. This would not satisfy the credibility criterion of "arms length" involvement by industry.
Approximately four years after the EC expert group made its recommendations for RF research, the International Agency for Research on Cancer (IARC) proposed a multi-centre (approximately 10 countries including Australia,
UK, USA, Sweden, Canada and Germany) case-control study of head and neck cancers in people using mobile telephones. The Australian programme plans to study 100 cases of brain tumour patients compared with matched normal healthy residents in Sydney. The criteria for selection include a 10 year history of mobile phone use. As digital GSM technology has not been in use for that length of time it is probable that users will have been exposed to
both analog and digital pulsed RF fields. For GSM use, the observation time may be still too short for definite conclusions.
The UK and Australia have similar government supported prestigious research centres that are responsible for providing expert advice on issues like possible health effects of radio-frequency radiation; the National Radiological Protection Board (NRPB) and Commonwealth Scientific and Industrial Research Organisation (CSIRO), respectively. Both organisations are inadequately funded for research in this area; the UK Department of
Health annual budget for RFR research is approximately £100,000 while the CSIRO in Australia does not have resources available for research on biological effects of RFR. Most worldwide research is funded by industry.
Motorola Ltd provides substantial support for the Australian research programme including providing the RF exposure system for the transgenic mouse cancer study.
The UK House of Commons Select Committee on Science and Technology investigation into health effects of mobile telephones concluded that the level of publicly funded research was inadequate for a programme of research into public health implications. The committee recommended that the industry and the NRPB explore ways in which the design of mobile telephones might limit personal exposure to RF radiation. As governments appear reluctant to provide funding, or even incentives for health related research, and telecommunications industries have their own
agendas and priorities, a simple solution to the funding problem has been suggested. A small levy applied to each subscriber to mobile telephone service would readily produce funding at a level far greater than any government effort has initiated. If applied at a minimal rate of only 20 pence per month per user, this would produce an annual amount of £10 million in the UK. This magnitude of resources could make a difference to the otherwise inadequate level of non-commercial funding for research on RFR.
WHAT REGULATIONS?
In recent years the development of appropriate health related standards for RF exposure has been driven by both commercial interests and the need to protect the public from risk. There are two separate issues involved. As
this technology is globally applied, there are sound commercial and economic reasons to create a universally consistent technical standard. This can help lower international trade barriers and also ensure that all manufacturers use the same methods of measuring and describing RF output. The advent of mobile telephony has introduced concern about the potential risk of causing adverse health effects from long term, whole of life, exposure to RF energy. Attempts to apply the RF technical standard as a health standard has created a far more complex situation. In order to properly protect against risk of potential adverse health consequences, regulatory standards need to include a margin of safety to allow for uncertainty. This is contrary to the requirements of a simple technical
standard that has to be based on measurable quantities. The problem that now faces various national standards setting committees is, how to deal with the uncertainty and inadequacy of existing knowledge on biological
effects of RF radiation.
How big should the "safety margin" be to adequately allow for uncertain risk? The ICNIRP has published guidelines with exposure limits based on thermal effects and extrapolated with an arbitrary reduction factor over a
range of frequencies, including those used in telecommunications. Some national standards bodies have chosen to adopt the ICNIRP guidelines, while others have recently chosen to set more conservative limits. In Australia,
the continuing debate over an acceptable safe level for health effects could not be resolved by an expert committee working under the voting requirements of Standards Australia. There was insufficient agreement in
the committee to support a proposed relaxation of existing RF exposure limits. There were also strong differences of opinion on the value of including a precautionary principle in the standard. Meanwhile, the recently formed Australian Communications Authority (ACA) published its intention to use an Australian standard for regulatory purposes. The
purpose was to invoke regulatory arrangements limiting exposure of the general public to radiofrequency electromagnetic radiation from radio-communications transmitters. Since the failure to reach consensus on
the new standard, the ACA is now working with the newly formed Australian Radiation Protection and Nuclear Safety Agency (ARPANSA) to create a new standard. ARPANSA was given authority to develop national standards under the provisions of the Australian Radiation Protection and Nuclear Safety Act, 1998. Whilst Standards Australia expert committees could not reach agreement on relaxation of limits after many years of deliberations, the ARPANSA select group plans to create an Australian standard in a few months. Publication of the completed standard is expected late in 2000. The most likely outcome is that the limits prescribed in the ICNIRP guidelines will be adopted. It is also intended that the technical safety standard will be complemented by a published code of practice document which is intended to address the precautionary approach to the exposure of persons
to EMR resulting from telecommunications services (ACA 2000). Meanwhile, other countries such as Switzerland and China have introduced far more conservative limits on RF exposures. The exposure limit for telecommunications frequencies set by the Chinese health ministry is approximately 100 times lower than that of the Institute of Electrical and Electronics Engineers (IEEE) or ICNIRP. A similar low level was introduced by the Swiss Federal Agency for Environment Forests and Landscape in February 2000 to cover exposures at locations of sensitive areas, eg
schools, playgrounds and hospitals.

CAUTIOUS APPROACH
While inconsistencies and uncertainty remain in the scientific database on biological effects of RF exposures from mobile telephony, it is reasonable to exercise some caution in the use of mobile telephones. If individuals
are concerned about the potential risk of adverse health effects that might result from the application of a RF-emitting device to the side of their head, they can simply choose to avoid or minimise the private usage patterns. Another option is to create a greater distance between the radiating antenna and the user's head. This can be simply achieved by the use of kits that connect an earpiece to the mobile telephone by means of a cable. There have been some suggestions in the popular media that use of these kits might actually increase the exposure to the user. It is
difficult to conceive as to how this might occur as the purpose of the mobile phone is to concentrate maximum RF output from the antenna. It is possible that the use of handpieces remote from the head may alleviate some of the symptoms associated with headaches and skin irritation by removing the source of heat from the side of the face.
In occupational settings it may be more difficult to reduce the exposure from mobile telephones. Where there is a more immediate risk, such as with electromagnetic interference with electronic devices, the use of mobile phones is simply prohibited. This occurs in some hospital critical areas. A major teaching hospital in Sydney, Australia recently advised staff to minimise personal exposure to mobile telephones on the basis of the current uncertainty about health effects of continued exposure to low level RFR. The executive director announced in a media statement, in March 2000, that the hospital has adopted a precautionary approach due to concern about the lack of definitive research and clear outcomes. The precautionary "Mobile Cellular Phone Policy" directs staff to follow recommended practices in the use of mobile telephones, as prescribed by the hospital's occupational health and safety department. The recommended practices include minimising use of mobile phones by using pagers and landlines where possible, and when using mobile phones to ensure that the device is as far from the body as possible. It is also noted that there is no Australian Standard covering the use of shielding devices, and that their effectiveness is still
unproven.
Air travellers are accustomed to being instructed to switch off their mobile telephones until after arrival at their destination. In this case the concern is for the potential interference with aircraft navigation systems; such an effect might result in an immediate and catastrophic health effect. This risk minimisation process is not based on direct
scientific evidence, but merely applies a sensible precautionary approach. Standards for acceptable exposure to RF radiation are based primarily on independently verified evidence of robust, thermally mediated, biological
effects and take little account of a need for precautionary risk minimisation. The standards have adopted a somewhat ad hoc so-called safety factor which sets the upper limit of exposure of 0.4 W/kg (SAR) at a tenth of the value needed to produce adverse physiological effects. This level is applied for occupational (or informed) exposures, while the whole body exposure limit for the general public is reduced by a further factor of five to 0.08 W/kg.
The various models of mobile telephones employ different engineering designs and, therefore, emit different amounts of RF energy that may be absorbed by the head of the user. When measured in terms of the specific
absorption rate (SAR) the levels of electromagnetic radiation emitted by some mobile telephones approach the upper limit prescribed by international standards. Under the present system of labelling there is no requirement
for mobile telephone manufacturers to notify the user of the output of each model.
CONCLUSIONS
The level of RF exposure received by the general public from mobile telephone base stations is below that where there is independently verified convincing evidence of directly linked adverse biological effects. Indirect, stress-related effects can have health consequences, particularly for sensitive individuals. Users of mobile telephones have reported a range of irritating effects, including headaches, which may not necessarily result in adverse health consequences. There is evidence that RFR from mobile telephones directly interferes with the operation of some medical electronic equipment.
The issue of bioeffects and safety of mobile telephones remains uncertain and will not be resolved until a convincing mass of credible data is obtained from organised independent research. For the sake of ensuring public acceptance of the results, it is essential that research programmes are not controlled or directly funded by the telecommunications industry. An essential requirement of credible bioeffects research is the elimination of bias from the system. Bias comes from interpretation of data and, more importantly for this area of research, in selection of study type.
Plausibility of results can only be assured by preventing any form of selection bias that might restrict thorough study or concentrate efforts on a particular politically sensitive issue at the expense of more fundamental research.
Industry funded studies prefer to simply repeat studies that report positive effects without fully investigating the possible causes of reported effects. As governments appear reluctant to organise or fund research programmes on possible health effects of RFR, it is unlikely that any systematic examination of the subject will occur in the near future. Therefore, the issue of human health risk associated with prolonged use of mobile telephones is unlikely to be resolved soon.
REFERENCES
1. ACA Connections. EMR Standard - New Developments. The Australian Communications Authority Bulletin. Issue 10; 15. 2000.
2. Adey, WR., Byus, CV., Cain, CD., et al. Spontaneous and Nitrosourea-induced Primary Tumours of the Central Nervous System in Fischer 344 Rats Chronically Exposed to 836 MHz Modulated Microwaves. Rad. Res., 152; 293-302. 1999.
3. Barnett, SB. Status of Research on Biological Effects and Safety of Electromagnetic Radiation: Telecommunications Frequencies. Report to Spectrum Management Agency, Australian Dept. of Communications and the Arts, Canberra ACT. 1994.
4. Chou, C-K., Guy, AW., Kunz, LL., et al. Long-term, Low-level Microwave Irradiation of Rats. Bioelectromagnetics, 13; 469-496. 1992.
5. Clifford, KJ., Joyner, KH., Stroud, DB., et al. Mobile Telephones Interfere with Medical Electrical Equipment. Aust. Physical Engineering, 17; 23-27. 1994
6. Goswami, PC., Albee, LD., Parsian, AJ., et al. Proto-oncogene mRNA Levels and Activities of Multiple Transcription Factors in C3H10T1/2 Murine Embryonic Fibroblasts Exposed to 835.62 and 847.74 MHz Cellular Phone Communication Frequency Radiation. Rad. Res., 151; 300-309. 1999.
7. Lai, H., and Singh, N.P. Acute Low-intensity Microwave Exposure Increase DNA Single-strand Breaks in Rat Brain Cells. Bioelectromagnetics, 16, 207-210. 1995.
8. Malyapa, RS., Ahern, EW., Bi, C., et al. DNA Damage in Rat Brain Cells After in vivo Exposure to 2450 MHz Electromagnetic Radiation and Various Methods of Euthanasia. Rad. Res., 149; 637-645. 1998.
9. McKinlay, A. Radiotelephones and Human Health: A European Research Initiative. Radiation Protection Dosimetry, 72; 313-320. 1997.
10. Repacholi, MH., Basten, A., Gebski, V., et al. Lymphomas in Eu-Pim1 Transgenic Mice Exposed to Pulsed 900 MHz Electromagnetic Fields. Rad. Res., 147, 631-640. 1997.
11. Royal Society of Canada. A Review of the Potential Health Risks of Radiofrequency Fields from Wireless Telecommunications Devices. Expert panel report prepared at the request of the Royal Society of Canada for
Health Canada. ISBN 9 20064 68 X. 1999.
_______________________________

EMFacts Consultancy
PO Box 96,
North Hobart, 7002
Tasmania, Australia
Phone: (03) 62430195
Fax: (03) 62430340
Email: emfacts@trump.net.au
ICQ: 30814841
Web: http://www.tassie.net.au/emfacts/
______________________________