Doi:10.1016/j.biopha.2007.12.004

Available online at www.sciencedirect.com Biomedicine & Pharmacotherapy 62 (2008) 104e109 Biological effects from electromagnetic field exposure and a Department of Oncology, University Hospital, SE-701 85 O¨rebro, Sweden b Sage Associates, Santa Barbara, CA, USA Received 6 December 2007; accepted 12 December 2007 During recent years there has been increasing public concern on potential health risks from power-frequency fields (extremely low frequency electromagnetic fields; ELF) and from radiofrequency/microwave radiation emissions (RF) from wireless communications. Non-thermal (low-intensity) biological effects have not been considered for regulation of microwave exposure, although numerous scientific reports indicate sucheffects. The BioInitiative Report is based on an international research and public policy initiative to give an overview of what is known of bi-ological effects that occur at low-intensity electromagnetic fields (EMFs) exposure. Health endpoints reported to be associated with ELF and/orRF include childhood leukaemia, brain tumours, genotoxic effects, neurological effects and neurodegenerative diseases, immune system dereg-ulation, allergic and inflammatory responses, breast cancer, miscarriage and some cardiovascular effects. The BioInitiative Report concluded thata reasonable suspicion of risk exists based on clear evidence of bioeffects at environmentally relevant levels, which, with prolonged exposuresmay reasonably be presumed to result in health impacts. Regarding ELF a new lower public safety limit for habitable space adjacent to all new orupgraded power lines and for all other new constructions should be applied. A new lower limit should also be used for existing habitable spacefor children and/or women who are pregnant. A precautionary limit should be adopted for outdoor, cumulative RF exposure and for cumulativeindoor RF fields with considerably lower limits than existing guidelines, see the BioInitiative Report. The current guidelines for the US andEuropean microwave exposure from mobile phones, for the brain are 1.6 W/Kg and 2 W/Kg, respectively. Since use of mobile phones is asso-ciated with an increased risk for brain tumour after 10 years, a new biologically based guideline is warranted. Other health impacts associatedwith exposure to electromagnetic fields not summarized here may be found in the BioInitiative Report at .
Ó 2007 Elsevier Masson SAS. All rights reserved.
Keywords: Electromagnetic fields (EMFs); Extremely low frequency electromagnetic fields (ELF); Radiofrequency fields (RF); Carcinogenesis; Public health;Standard setting electromagnetic fields; ELF) and from radiofrequency/micro-wave radiation emissions (RF) from wireless communications During recent years there has been increasing scientific and data transmission. So far, guidelines for exposure to evidence for, and public concern on potential health risks microwaves have been based on thermal (heating) effects.
from power-frequency fields (extremely low frequency Non-thermal (low-intensity) effects have not been consideredfor regulation of exposure. Recently a more comprehensivereport was published at Internet that documents consider- Note: the views expressed in this paper are not necessarily those of the able scientific evidence for bioeffects and adverse health im- whole BioInitiative Report group. Some parts of this article rely on chapters pacts at exposure levels far below current public safety by different authors in the BioInitiative Report.
** standards. The purpose of that report was to assess scientific This manuscript is a part of the dossier ‘‘Cancer: Influence of environ- evidence on health impacts from electromagnetic radiation ment’’, Biomedicine & Pharmacotherapy 2007;61:10.
below current public exposure limits and evaluate what * Corresponding author. Tel.: þ46 19 602 10 00; fax: þ46 19 10 17 68.
changes in these limits are warranted now to reduce possible 0753-3322/$ - see front matter Ó 2007 Elsevier Masson SAS. All rights reserved.
doi:10.1016/j.biopha.2007.12.004 L. Hardell, C. Sage / Biomedicine & Pharmacotherapy 62 (2008) 104e109 public health risks in the future. This report was written by for exposure to microwaves during mobile phone use is not 14 scientists, public health and public policy experts to doc- safe for long-term exposure and needs to be revised.
ument the scientific evidence on electromagnetic fields. Thecurrent short review is based on the BioInitiative Report 2.2. RF fields other than from mobile phones and and gives summaries of relevant topics. For more details in- epidemiological evidence for brain tumours cluding complete reference list, see that document at It is concluded that only few studies of long-term exposure Everyone is exposed to two types of electromagnetic fields to low levels of RF fields and brain tumours exist, all of (EMFs): (a) ELF fields from electrical and electronic appli- which have methodological shortcomings including lack of ances and power lines, and (b) RF radiation from wireless quantitative exposure assessment. Given the crude exposure devices such as cell phones and cordless phones, cellular an- categories and the likelihood of a bias towards the null hy- tennas and towers, and broadcast transmission towers. In this pothesis of no association, the body of evidence is consistent report we will use the term EMFs when referring to all electro- with a moderately elevated risk. Occupational studies indicate magnetic fields in general, and the terms ELF and RF when that long-term exposure at workplaces may be associated referring to the specific type of exposure. They are both types of non-ionizing radiation, which means that they do not have Although in some occupations (especially in military jobs) sufficient energy to break off electrons from their orbits current exposure guidelines may have sometimes been reached around atoms and ionize (charge) the atoms, as ionizing or exceeded, overall the evidence suggests that long-term exposure to levels generally lying below current guidelinelevels still carry the risk of increasing the incidence of brain Despite a rather low population attributable risk (likely be- 2.1. Mobile phone use and evidence for brain low 4%), still more than 1000 cases per year in the US can be attributed to RF exposure at workplaces alone.
We made a review including 18 studies, two cohort studies 2.3. Evidence for childhood cancers and leukaemia and 16 case-control studies. Most studies have published datawith rather short latency period and limited information on The only endpoint studied so far in sufficient detail is child- long-term users. Thus, a meta-analysis of the risk for acoustic hood leukaemia. Brain and nervous system tumours were also neuroma, glioma and meningioma was performed for mobile studied in some detail but due to the diversity of these tumours phone use with a latency period of 10 years or more . Over- no conclusions can be drawn. Childhood leukaemia is the most all OR ¼ 1.3, 95% CI ¼ 0.6e2.8 was obtained increasing to frequent childhood malignancy that peaks in the age group of OR ¼ 2.4, 95% CI ¼ 1.1e5.3 for ipsilateral mobile phone 2 to about 5 years. This peak seems to have been newly use. For glioma OR ¼ 1.2, 95% CI ¼ 0.8e1.9 was calculated.
evolved in the early quarter of the 20th century and may be Ipsilateral use yielded OR ¼ 2.0, 95% CI ¼ 1.2e3.4. In total due to electrification This assumption is supported by OR ¼ 1.3, 95% CI ¼ 0.9e1.8 was found for meningioma in- the absence of this peak or it being much less pronounced in creasing to OR ¼ 1.7, 95% CI ¼ 0.99e3.1 for ipsilateral use.
Only two studies have been published since then. Both were An overview of existing evidence from epidemiological on acoustic neuroma They were small and included no studies indicates that there is a continuous increase of risk cases with a latency period of at least 10 years. Furthermore, with increasing levels of average magnetic field exposure.
most ORs were <1.0 in these two studies indicating serious Risk estimates reach statistical significance at levels of 3e methodological problems. The final results on this topic 4 mG (0.3e0.4 microTesla or mT). The overall odds ratio in from the Interphone study led by the International Agency nine studies was 2.1, 95% confidence limit 1.3e3.3. A low for Research on Cancer (IARC) are expected during 2008.
number of children are exposed at these or higher levels.
No other studies than from the Hardell group has published The balance of evidence suggests that childhood leukaemia results for use of cordless phones (DECT) As we have is associated with exposure to power-frequency ELFs either discussed in our publications it is pertinent to include also during pregnancy or early life. Considering only average such use in this type of studies. Cordless phones are an impor- MF flux densities the population attributable risk is low to tant source of exposure to radiofrequency microwaves and moderate. However, there is a possibility that other exposure they are usually used for a longer time period on daily basis metrics are much stronger related to childhood leukaemia as compared with mobile phones. Thus, to exclude such use, and may account for a substantial proportion of cases, perhaps as was done in e.g. the Interphone studies, could lead to an un- up to 80% of all cases. The population attributable fraction derestimation of the risk for brain tumours from use of wire- ranges between 1 and 4% assuming only exposures above In summary our review yielded a consistent pattern of an Other childhood cancers except leukaemia have not been increased risk for acoustic neuroma and glioma after 10 studied in sufficient detail to allow conclusions about the exis- years mobile phone use. We conclude that current standard L. Hardell, C. Sage / Biomedicine & Pharmacotherapy 62 (2008) 104e109 The International Commission for Non-ionizing Radiation conclusive evidence is untenable given the enormous costs Protection (ICNIRP) and Institute of Electric and Electronics and societal and personal burdens caused by this disease.
Engineers, Inc. (IEEE) guideline levels are designed to protectfrom short-term immediate effects only, but not chronic expo- 2.5. Changes in the nervous system and brain function sures. Long-term effects such as cancer are evoked by expo-sure several orders of magnitudes below current guideline Exposure to electromagnetic fields has been studied in con- levels. The BioInitiative Report concludes that the evidence nection with Alzheimer’s disease, motor neuron disease and for increased risk of childhood leukaemia with chronic expo- Parkinson’s disease. There is evidence that high level of amy- sure to ELFs is sufficient to warrant revision of ELF public loid beta is a risk factor for Alzheimer’s disease, and exposure to ELF can increase this substance in the brain. There is con-siderable evidence that melatonin can protect the brain against damage leading to Alzheimer’s disease, and also strong evi-dence that exposure to ELF can reduce melatonin levels.
There is evidence from multiple areas of scientific investiga- Thus it is hypothesized that one of the body’s main protections tions that ELF is related to breast cancer. Over the last two de- against developing Alzheimer’s disease (melatonin) is less cades there have been numerous epidemiological studies on available to the body when people are exposed to ELF. Pro- breast cancer in both men and women, although this relationship longed exposure to ELF fields could alter calcium (Ca2þ) remains controversial. Many of these studies, however, report levels in neurons and induce oxidative stress. Concern has that ELF exposures are related to increased risk of breast cancer.
also been raised that humans with epileptic disorders could The evidence from studies on women in workplaces sug- gests that ELF is a risk factor for breast cancer for women Laboratory studies show that the nervous system of both with long-term exposures of 10 mG (1.0 mT) and higher.
humans and animals is sensitive to both ELF and RF. Mea- Laboratory studies that examine human breast cancer cells surable changes in brain function and behaviour occur at have shown that ELF exposure between 6 mG and 12 mG levels associated with new technologies including cell phone (0.6e1.2 mT) can interfere with protective effects of melatonin use. Exposing humans to cell phone radiation can change for the growth of these breast cancer cells. For a decade, there brainwave activity at levels as low as 0.1 watt per kilogram has been evidence that human breast cancer cells grow faster if (W/Kg) specific absorption rate (SAR) in comparison to the exposed to ELF at low environmental levels. This is thought to US allowable level of 1.6 W/Kg (in 1 g of tissue) and IC- be because ELF exposure can reduce melatonin levels in the NIRP allowable level of 2.0 W/Kg (in 10 g of tissue). Cell phone radiation can affect memory and learning.
Laboratory studies of animals that have breast cancer tu- Changes in the way in which the brain and nervous system mours have been shown to have more tumours and larger react depend very much on the specific exposures. Most studies tumours when exposed to ELF and a chemical tumour pro- only look at short-term effects, so the long-term consequences moter at the same time. These studies taken together indicate of exposures are not established, but existing scientific docu- that ELF is a likely risk factor for breast cancer, and that mentation of effects is sufficient to warrant preventative action ELF levels of importance are no higher than many people with reduction in exposures, particularly for vulnerable groups are exposed to at home and at work. A reasonable suspicion of risk exists and is sufficient evidence on which to recom- Factors that determine effects can depend on head shape mend new ELF limits; and to warrant preventative action.
and size, the location, size and shape of internal brain struc- Given the very high lifetime risks for developing breast tures, thinness of the head and face, hydration of tissues, thick- cancer in women, and the critical importance of prevention, ness of various tissues, dielectric constant of the tissues and so ELF exposures should be reduced for all people who are in on. Age of the individual and state of health also appear to be high ELF environments for prolonged periods of time. Reduc- ing ELF exposure would be particularly important for people There is large variability in the results of ELF and RF test- who have breast cancer. The recovery environment should ing, which would be expected to be based on the large variabil- have low ELF levels given the evidence for poorer survival ity of factors that can influence test results. However, it is rates as shown for subjects with another malignant disease, clearly demonstrated that under some conditions of exposure, childhood leukaemia patients in ELF fields over 2 mG or the brain and nervous system functions of humans are altered.
The consequence of long-term or prolonged exposures has not Preventative action for those who may be at higher risk for been thoroughly studied in either adults or in children.
breast cancer is also warranted, particularly for those taking ta- The consequence of prolonged exposures to children, whose moxifen during their anti-cancer treatment, since in addition to nervous systems continue to develop until late adolescence, is reducing the effectiveness of melatonin, ELF exposure may unknown at this time, but there are credible, published studies also reduce the effectiveness of tamoxifen at these same low reporting bioeffects and adverse health impacts with exposures exposure levels. There is no excuse for ignoring the substantial at very low levels (far below public safety standards). This body of evidence we already have that supports an association could have serious implications to adult health and functioning between breast cancer and ELF exposure; waiting for in society if years of exposure of the young to both ELF and RF L. Hardell, C. Sage / Biomedicine & Pharmacotherapy 62 (2008) 104e109 result in diminished capacity for thinking, judgment, memory, Factors that may explain the failure of some studies to dem- learning, and control over behaviour.
onstrate effects, while others report clear and reproducibleeffects include (a) which DNA assay is used, (b) the exposure 2.6. Evidence for effects on gene and protein expression parameters of the experiment, and (c) which cell lines areused. Any effect of EMF has to depend on the energy absorbed The effects of RF EMF on global gene and protein expres- by a biological entity and on how the energy is delivered in sion have been investigated in different biological systems, space and time. Frequency, intensity, exposure duration, and and most of the studies were focused on the mobile phone uti- the number of exposure episodes can affect the response, lization frequency (800e2000 MHz) at a relatively low expo- and these factors can interact with each other to produce dif- sure density (average SAR near 2.0 W/Kg). Some studies reported negative results of RF EMF exposure on gene The ‘comet assay’, has been used in most of the EMF stud- ies to determine DNA damage. Different versions of the assay Based on current available literature, it is justified to con- have been developed. These versions have different detection clude that EMF exposure can change gene and/or protein sensitivities and can be used to measure different aspects of expression in certain types of cells, even at intensities lower DNA strand breaks. A comparison of data from experiments than ICNIRP recommended values. However, the biological using different versions of the assay may be misleading, and consequences of most of the changed genes/proteins as based may explain differing study results since some DNA comet on early studies from proteomics and transcriptomics are still assays are far more sensitive in detecting DNA damage than unclear, and need to be further explored. Thus, it is not the time point yet to assess the health impact of EMF based on A plausible biological mechanism to account for carcino- the gene and protein expression data. The IEEE and WHO genesis is via free radical formation inside cells. Free radicals databases do not include the majority of ELF studies; they kill cells by damaging macromolecules, such as DNA, protein do include the majority of the RF studies.
and membrane. Furthermore, free radicals play an essential Currently, the state of proteomics and transcriptomics is in role in the activation of certain signalling pathways. Several its infancy, with only a few dozen studies reporting results, reports have indicated that EMFs enhance free radical activity some positive and some negative. The EMF research commu- in cells particularly via the Fenton reaction The Fenton nity should pay equal attention to the negative reports as to reaction is a catalytic process of iron to convert hydrogen per- the positive ones. Not only the positive findings need to be oxides, a product of oxidative respiration in the mitochondria, replicated, the negative ones need to be critically assessed into hydroxyl free radical, which is a very potent and toxic free radical. Any exposure, including prolonged low-intensity ELFand RF exposures that result in increased free radical produc- 2.7. Evidence for genotoxic effects e DNA damage tion may be considered a plausible biological mechanism forcarcinogenesis.
From this literature survey, about 50% of the studies re- ported effects. Not every study, however, would be expected to document effects, given the wide range of exposure condi-tions and varying sensitivity of assays. One can conclude that Studies of the stress response in different cells under vari- under certain conditions of exposure, radiofrequency radiation ous conditions have enabled us to characterize the molecular is genotoxic. Data available are mainly applicable only to cell mechanisms by which cells respond to EMF and their effects phone radiation exposure. Other than the study by Phillips on health risk. That information can now correct assumptions et al. there are very few published studies of RF radiation about biological effects of EMF, and establish a scientific basis at levels that one can experience in the vicinity of base stations It is generally agreed that EMF safety standards should be During cell phone use, a relatively constant mass of tissue in based on science, yet recent EMF research has shown that a ba- the brain is exposed to the radiation at relatively high intensity sic assumption used to determine EMF safety is not valid. The (peak SAR of 4e8 W/Kg). Several studies reported DNA dam- safety standard assumes that EMF causes biological damage age at lower intensity than 4 W/Kg. The IEEE has revised its only by heating, but cell damage occurs in the absence of heat- recommended standard for localized tissue exposure, changing ing and well below the safety limits. This has been shown in it from 1.6 W/Kg over 1 g of tissue to 2 W/Kg over 10 g of tis- many studies, including the cellular stress response where cells sue, although the Federal Communications Commission has synthesize stress proteins in reaction to potentially harmful not adopted this change. Since distribution of radiofrequency stimuli in the environment, including EMF. The stress response energy is non-homogenous inside tissue, this change allows to both the power-frequency (ELF) and radiofrequency/micro- a higher peak level of exposure. Furthermore, since critical ge- wave (RF) ranges shows the inadequacy of the thermal SAR netic mutations in one single cell are sufficient to lead to cancer and there are millions of cells in a gram of tissue, it is incon- The stress response is a natural defence mechanism activated ceivable that the base of SAR standard was changed by IEEE by molecular damage caused by environmental forces. The re- from averaged over 1 gm of tissue to 10 gm.
sponse involves activation of DNA, i.e., stimulating stress genes L. Hardell, C. Sage / Biomedicine & Pharmacotherapy 62 (2008) 104e109 as well as genes that sense and repair damage to DNA and pro- acceptable to build new power lines and electrical facilities teins. Scientific research has identified specific segments of that place people in ELF environments that have been associ- DNA that respond to both ELF and RF. It has been possible ated with an increased risk of adverse health effects, levels to move these specific segments of DNA and transfer the sensi- generally at 2 mG (0.2 mT) and above.
tivity to EMF. At high EMF intensities, the interaction with A new, lower planning limit for habitable space adjacent to DNA can lead to DNA strand breaks that could result in muta- all new or upgraded power lines and for all other new con- tion, an initiating step in the development of cancer.
struction should be applied. A lower limit should also be Scientific research has shown that ELF and RF fields inter- used for existing habitable space for children and/or women act with DNA to stimulate protein synthesis, and at higher in- who are pregnant. This recommendation is based on the as- tensities to cause DNA damage. The biological thresholds sumption that a higher burden of protection is required for (field strength, duration) are well below current safety limits.
children who cannot protect themselves, and who are at risk To be in line with EMF research, a biologically based standard for childhood leukaemia at rates that are traditionally high must replace the thermal SAR standard, which is fundamen- enough to trigger regulatory action.
tally flawed. EMF research also indicates a need for protection While it is not realistic to reconstruct all existing electrical against the cumulative biological effects stimulated by EMF distributions systems in the short-term, steps to reduce expo- across the electromagnetic spectrum.
sure from these existing systems need to be initiated, espe-cially in places where children spend time, and should be A precautionary limit should be adopted for outdoor, cumu- lative RF exposure and for cumulative indoor RF fields withconsiderably lower limits than existing guidelines. It should Exposure to EMFs has been linked to a variety of adverse reflect the current RF science and prudent public health re- health outcomes. There are other effects not summarized sponse that would reasonably be set for pulsed RF (ambient) here, see the BioInitiative Report Health endpoints that exposures where people live, work and go to school. This level have been reported to be associated with ELF and/or RF in- of RF is experienced as whole-body exposure, and can be clude childhood leukaemia, adult brain tumours, childhood a chronic exposure where there is wireless coverage present brain tumours, genotoxic effects (DNA damage and micronu- for voice and data transmission for cell phones, pagers and cleation), neurological effects and neurodegenerative diseases, personal digital assistants (PDAs) and other sources of radio- immune system deregulation, allergic and inflammatory re- frequency radiation. Although this RF target level does not sponses, breast cancer in men and women, miscarriage and preclude further rollout of WI-FI technologies, wired alterna- tives to WI-FI should be implemented, particularly in schools Effects are not specifically segregated for ELF or RF, since and libraries so that children are not subjected to elevated RF many overlapping exposures occur in daily life, and because levels until more is understood about possible health impacts.
this is an artificial division based on frequencies as defined This recommendation should be seen as an interim precaution- in physics that have little bearing on the biological effects.
ary limit that is intended to guide preventative actions. More Both ELF and RF, for example have been shown to cause cells conservative limits may be needed in the future.
to generate stress proteins, a universal sign of distress in plant, The current guideline for microwave exposure from mobile animal and human cells, and to cause DNA damage and neu- phones in Europe is 2 W/Kg for the brain. This is based on rological impacts at levels far below current safety standards.
thermal effect using cataract development in animal eyesinduced at 100 W/Kg with a safety factor of 50 for standard 3.2. Public health policy recommendations setting. There were also considerations about the relationshipbetween the whole-body SAR and local hot spots and local There are many historical examples of scientifically based SAR in relation to whole-body SAR. Since use of mobile early warnings about potential health effects from environ- phones is associated with an increased risk for brain tumours mental hazards and a long time period until precautionary (glioma, acoustic neuroma) after 10 years a new biologically and preventive measures were undertaken . Vested inter- based guideline should be applied. This new guideline should ests may thereby counteract necessary public health actions be based on non-thermal (low-intensity) effects from micro- The precautionary principle should be used when there wave exposure. It should be added that in toxicology normal is reasonable ground for concern. Based on the BioInitiative practice is to add a safety limit of at least factor 100, which Report , this criterion is fulfilled regarding exposure to is factor 10 from animal to human beings and factor 10 for in- electromagnetic fields, both extremely low frequency electro- Exposure from base stations for DECT phones are not New regulatory limits for ELF based on biologically rele- specifically addressed in the BioInitiative Report. However, vant levels of ELF are warranted, see the BioInitiative Report.
we conclude that indoor exposure to RF should be assessed ELF limits should be set below those exposure levels that have as well as exposure while using DECT phones. There is indi- been linked in childhood leukaemia studies to increased risk of cation of increased brain tumour risk associated with DECT disease, plus an additional safety factor. It is no longer phones and a safety factor is warranted both for these phones L. Hardell, C. Sage / Biomedicine & Pharmacotherapy 62 (2008) 104e109 and indoor base station exposures. The same standard might [3] Klaeboe L, Blaasaas KG, Tynes T. Use of mobile phones in Norway and be applied to cordless phones as for a new guideline for mo- risk of intracranial tumours. Eur J Cancer Prev 2007;16:158e64.
[4] Schlehofer B, Schlafer K, Blettner M, Berg G, Bo¨hler F, Hettinger I, bile phones based on biological effects. This is a reasonable et al. Environmental risk factors for sporadic acoustic neuroma (Inter- suggestion to address the condition where occupied interior phone Study Group, Germany). Eur J Cancer 2007;43(11):1741e7.
space is affected by DECT phones or other RF-emitting devices installed by the occupants. As with ELF fields also [5] Hardell L, Carlberg M, Hansson Mild K. Pooled analysis of two case- for RF fields different limits may be needed in the future as control studies on the use of cellular and cordless telephones and therisk of benign tumours diagnosed during 1997 [6] Hardell L, Hansson Mild K, Carlberg M. Pooled analysis of two case- control studies on use of cellular and cordless telephones and the risk for malignant brain tumours diagnosed in 1997e2003. Int Arch OccupEnviron Health 2006;79:630e9.
Supported by grants from a private philanthropic organiza- [7] Milham S, Ossiander EM. Historical evidence that residential electrifica- tion administered by Commonweal, Bolinas, California, USA, tion caused the emergence of the childhood leukemia peak. Med Hypoth-eses 2001;56:290e5.
and from Cancer-och Allergifonden, Sweden. The paper is [8] Kheifets L, Afifi AA, Shimkhada R. Public health impact of extremely based on the BioInitiative Report, and its individual chapters by the authors of this article, and by Carl F. Blackman, PhD, Martin Blank, PhD, Professor, David Gee, European Environ- [9] European Environmental Agency. Highlights: European Environmental mental Agency, Michael Kundi, PhD med. habil. Professor, Agency website by Dr. Jacqueline McGlade, Director. Statement ofSeptember David O. Carpenter, MD, Zoreh Davanipour, DVM, PhD, Olle Johansson, PhD, Associate Professor, Henry Lai, PhD, [10] Phillips JL, Ivaschuk O, Ishida-Jones T, Jones RA, Campbell- Kjell Hansson Mild, PhD, Professor, Amy Sage, BS, Research Beachler M, Haggren W. DNA damage in Molt-4 T-lymphoblastoid cells Associate, Eugene L. Sobel, PhD, Zhengping Xu, PhD, and exposed to cellular telephone radiofrequency fields in vitro. Bioelectro- [11] Lai H, Singh NP. Magnetic-field-induced DNA strand breaks in brain cells of the rat. Environ Health Perspect 2004;112(6):687e94.
[12] European Environment Agency. Late lessons from early warnings: the precautionary principle 1896e2000. Copenhagen, Denmark. Environ-mental [1] BioInitiative report: a rationale for a biologically-based public exposure standard for electromagnetic fields (ELF and RF). [13] Hardell L, Walker M, Walhjalt B, Friedman LS, Richter ED. Secret ties to industry and conflicting interests in cancer research. Am J Ind Med [2] Hardell L, Carlberg M, So¨derqvist F, Hansson Mild K, Morgan LL.
Long-term use of cellular phones and brain tumours: increased risk [14] Scinicariello F, De Rosa CT. Genetic heterogeneity and its effect on associated with use for 10 years. Occup Environ Med 2007;64: susceptibility to environmental factors. Eur J Oncol 2007;12(3): 626e32. doi:10.1136/oem.2006.029751.

Source: http://www.mreengenharia.com.br/pdf/biologicaleffects.pdf