Elektromagnetische Felder Aktualisierung 2009

4. Can mobile phones or base stations trigger headaches or other health effects?

• 4.1 Have headaches and other symptoms been linked to mobile phones?
• 4.2 Can mobile phones affect the brain?
• 4.3 Have effects of mobile phones on reproduction and development been reported?
• 4.4 Are children more vulnerable to possible effects of mobile phones?

4.1 Have headaches and other symptoms been linked to mobile phones?

The SCENIHR opinion states:

3.3.3. Symptoms

What was already known on this subject?

The evaluation of the scientific data at the time of the 2007 opinion suggested that symptoms are not correlated to RF field exposure, but few studies had addressed this issue directly. The 2007 opinion concluded that scientific studies had failed to provide consistent support for a causal relationship between RF field exposure and self-reported symptoms (e.g. headache, fatigue, dizziness and concentration difficulties or well-being), sometimes referred to as electromagnetic hypersensitivity (EHS, see also chapter 3.5.3).

What has been achieved since then?

Several studies on symptoms and RF field exposure relevant to the mobile phone user situation have been published since the last opinion. Two studies (Oftedal et al. 2007; Hillert et al. 2008) included subjects who reported having experienced symptoms (headache, vertigo etc) in relation to mobile phone use, but no ill health attributed to other types of electrical equipment. Oftedal et al. (2007) included only subjects who reported headache during an initial open provocation when the subjects knew that they were exposed to RF fields. However, no effect on headache or physiological reactions such as heart rate and blood pressure was observed in the double blind tests. Hillert et al. (2008) observed an increase in reported headache (OR=2.49, 95% CI: 1.16-5.38) at the end of a three hour RF field exposure (double blind tests, time-averaged SAR 10 g 1.4 W/kg). Although more subjects in the group with a history of mobile phone related symptoms reported headache after exposure, the difference between sham and RF exposure was mainly due to a difference between the two exposure conditions in the group without a history of symptoms in relation to mobile phone use. The percentage of subjects in the non-symptom group who reported headache was 45.4 after RF exposure and 25.8 after sham; the corresponding percentages in the symptom group were 48.6 and 57.9. Thus, the results do not support a higher sensitivity in the group who reported symptoms in relation to their every day mobile phone use. No effect was observed with regard to any other symptom (including fatigue, vertigo, nausea and difficulties concentrating). Cinel et al. (2008) analysed a possible effect of RF fields with regard to reported symptoms in three separate studies primarily focusing on cognitive functions in healthy volunteers. In one of the studies the effect of RF exposure was observed on dizziness (p=0.001), but the two other studies by the same research group did not support such a relationship. Further analyses revealed that the significant difference appeared to be due to higher scores on dizziness in males at the end of RF exposure. Headache, fatigue and skin symptoms were unaffected in all three studies, and no other significant gender differences were observed.

In a Swedish cross-sectional study Söderqvist et al. (2008) reported that regular mobile phone users had headaches, difficulties concentrating and asthmatic symptoms more often than less frequent users. An Egyptian study on symptoms and base stations was published in 2007 (Abdel-Rassoul et al. 2007). Participants who lived close to base stations reported more symptoms. Another cross-sectional study, initiated by public concern of health effects from a military antenna system, compared reported health in two villages classified as “exposed” (based on RF measurements in the villages) to an “unexposed” village in Cyprus (Preece et al. 2007). The percentage of responders who reported migraine, headache, dizziness and depression was higher in exposed villages. These cross-sectional studies suffer from the same methodological limitations as earlier cross-sectional epidemiological studies addressing the same question and do not provide any firm basis for a conclusion on a possible causal relationship (see SCENIHR 2007 and chapter 3.8.3.2). In a German study the subjects were provided with personal exposimeters for 24 hours to investigate the association between exposure to mobile phone related RF fields and well-being (Thomas et al. 2008). Three mobile phone frequency ranges were assessed. Participants were randomly chosen from registration offices in four Bavarian cities. Acute symptoms as well as five groups of chronic symptoms were reported in a diary. No statistically significant association between exposure and symptoms was observed (all p values>0.05).

Provocation studies relevant to base station RF field exposure published since the last opinion have given additional support for the lack of a causal relationship proposed by the study by Regel et al. (2006) which followed-up the initial finding of an effect on well- being in the TNO study (Zwamborn et al. 2003). Eltiti et al. (2007a) performed open tests before the double blind experiment. In the open tests the possibly sensitive group (EHS) reported more symptoms and lower well-being during GSM as well as UMTS exposure as compared to sham (Eltiti et al. 2007a). During double blind testing, no difference in symptoms was observed between actual exposure and sham, neither in the EHS group nor in the reference group. During UMTS exposure the EHS group reported elevated levels of arousal which according to the authors may be due to an effect of sequence of exposure administration rather than the exposure itself. More subjects in the EHS group were exposed to UMTS during the first session. No effect on physiological functions was observed. A Danish study by Riddervold et al. (2008) investigated cognitive functions and self-reported subjective symptoms in adults and adolescents in relation to RF field exposure from UMTS base stations. No effects were seen on any of the cognitive tasks that were performed. A subjective headache rating did not reveal any difference between exposure conditions and sham for the two separate groups, but when data from the two groups were combined, a significant increase in headache was observed during UMTS exposure. The authors suggest that this finding may be due to differences at baseline (higher scores were reported for headache before UMTS exposure than before sham). A number of studies investigated the ability of the participants to detect GSM RF fields (e.g. Eltiti et al. 2007a, Bamiou et al. 2008, Furubayashi et al. 2008, Hillert et al. 2008, Kwon et al. 2008). The studies by Eltiti et al. (2007a) and Furubayashi et al. (2008) tested a base station-like signal (including UMTS exposure), while the three other tested GSM signals relevant to the user situation. Possibly sensitive subjects, reporting a sensitivity to EMF from mobile phones (Kwon et al. 2008), mobile phone related symptoms (Bamiou et al. 2008, Furubayashi et al. 2008, Hillert et al. 2008) or EHS (Eltiti et al. 2007a) were also included. The participants could not report correct exposure conditions better than by chance, and the possibly more sensitive groups could not do this better than the control groups. In the study by Kwon et al. (2008) two subjects had initially correct response rates over 90%. However, when tested again one month later these two subjects could not detect the true exposure conditions better than could be expected just by chance. In a meta-analysis Röösli (2008) investigated the ability to discriminate between sham and actual RF field exposure. Seven studies were included in the meta-analysis, including 182 subjects reporting symptoms attributed to electromagnetic fields and 332 non-symptomatic subjects. The results showed numerically a slightly better ability to detect the true exposure conditions than expected by chance. This finding is, however, within the limits of uncertainty of the analysis (Effect size, i.e. the relative difference between observed and expected correct answers, 0.04, 95%CI -0.02 - 0.11). The ability to detect the true exposure conditions did not differ between study groups (participants with EMF-related symptoms versus non-symptomatic participants), exposure sources (mobile phones versus base stations) or longer versus shorter exposure duration.

There is a discrepancy between results in open and double blind tests in studies of subjects with mobile phone related symptoms. In open tests, where the true exposure condition is known to the participant, the participants reported that they do react to the exposure under study. In double blind tests, there is a lack of consistent association between exposure and symptoms. Furthermore, subjects could not correctly detect exposure conditions during double blind tests. It is noted that symptoms are also triggered during sham exposure. These observations suggest that other factors, e.g. expectations of symptoms to be triggered based on prior experiences, i.e. a “nocebo” effect, may play a role in triggering symptoms (Rubin et al. 2006; Landgrebe et al. 2008; Stovner et al. 2008).

Discussion

In the previous opinion, it was concluded that scientific studies had failed to provide support for a relationship between RF exposure and self-reported symptoms. The 2007 opinion also stated that the knowledge at that time suggested that symptoms are not correlated to RF field exposure. Although an association between RF exposure and single symptoms was indicated in some new studies, taken together, there is a lack of consistency in the findings. Therefore, the conclusion that scientific studies have failed to provide support for an effect of RF on symptoms still holds.

The background for symptoms reported to be triggered by RF fields in everyday life has been discussed. There is a discrepancy between open exposures to RF fields where symptoms are triggered when the subjects are aware of the exposure, and double-blind provocations studies where there is no consistent association between RF and symptoms when subjects do not know if they are exposed to RF or not. These results indicate that a nocebo effect plays a role in symptom formation. This does not exclude the possibility of a RF field effect, but so far the support from scientific studies is stronger for a nocebo effect.

With regard to detection of fields, scientific studies have not provided any evidence that either so-called sensitive groups or healthy control groups can detect RF fields better than expected by chance.

Source & ©: SCENIHR,  Health Effects of Exposure to EMF (2009),
3.3.3. Symptoms, p.26-28

4.2 Can mobile phones affect the brain?

The SCENIHR opinion states:

3.3.4. Nervous system effects

What was already known on this subject?

SCENIHR concluded in the previous opinion that no clear neurotoxic effects due to RF exposure were seen (SCENIHR 2007). Certain changes in electrical activity and neurotransmitter biochemistry were noted but did not suggest any pathological hazard. It was also suggested that care must be taken in future cognitive animal experiments to reduce the stress effects related to restraint of animals.

What has been achieved since then?

A number of studies on human volunteers as well as on various animal species have been published since the previous opinion. They can mainly be divided into studies focusing on behaviour and cognition, electrophysiological measurements, sensory related functions, and studies focusing on cell and tissue integrity, including the blood-brain-barrier. Exposures have mostly been to GSM-related signals and UMTS-signals.

Human studies

Several studies employing healthy human volunteers have investigated possible effects on various behaviours and cognitive functions after acute exposures to GSM and/or UMTS signals. Most of the studies were randomised cross-over studies, and have used double- blind protocols. Several of these studies did not find any effects of exposure (Kleinlogel et al. 2008a, Kleinlogel et al. 2008b, Thomas et al. 2008, Riddervold et al. 2008, Unterlechner et al. 2008). Also the work by Curcio et al. (2008) was essentially negative, although they found a trend of shorter reaction time in a finger tapping test of subjects exposed to a 900 MHz GSM signal (SAR 0.5 W/kg). Hung et al. (2007) reported that GSM in “talk-mode” delayed sleep latency. Recently, Wiholm et al. (2009) found that subjects with self-reported symptoms who were performing a virtual spatial navigation test scored better after exposure to 884 MHz at an average SAR of 1.4 W/kg. Augner et al. (2009) studied psychological symptoms (good mood, alertness, calmness) in subjects exposed to GSM base station signals for 50 minutes. Exposure levels were 5.2 µW/cm² (“low”), 153.6 µW/cm² (“medium”), and 2126.8 µW/cm² (“high”). None of the exposure situations had any effect on mood or alertness, but calmness was increased during medium and high exposure. This finding could be due to chance, since multiple endpoints were investigated.

Electrophysiological measurements such as EEG allow for studying effects on specific parts of the central nervous system and thus specific functions in a non-invasive manner. Since the previous opinion, several papers have tried to replicate the study of Huber et al. (2002) which found that GSM-exposure enhanced the power of the so-called alpha- band. Recent studies show contradictory results. Perentos et al. (2007) exposed 12 subjects for 15 minutes to both modulated and non-modulated GSM signals, without finding any effects on any EEG component. In contrast, Croft et al. (2008) exposed 120 subjects for 30 min and measured EEG signals before, during, and after the exposure. This study found an alpha-power enhancement during exposure, which disappeared when exposure was terminated. Vecchio et al. (2007) also reported similar findings (exposure for 45 min to a telephone signal at maximal power which generated a calculated SAR of 2 W/kg). They found modulations of both alpha-1 and alpha-2 bands and also enhanced interhemispheral connectivity. In another study, Inomata-Terada et al. (2007) investigated whether mobile phone signals influenced cortical motor evoked potentials that were triggered by transcranial magnetic stimulation (TMS) but did not find any effects. An in vitro study employing rat cortical neurons, using the patch-clamp technique, did not show any changes in voltage-gated Ca2+-channels that could explain effects of modulated and un-modulated GSM signals on EEG (Platano et al. 2007).

Additional studies that have investigated if there are effects on sleep and sleep EEG have been published recently. Regel et al. (2007) exposed healthy male subjects to GSM handset-like signals (sham, 0.2 W/kg, 5 W/kg) for 30 minutes and found dose-dependent effects on sleep EEG and on cognitive tasks. Hung et al. (2007) also exposed volunteers for 30 minutes to various modulated GSM signals. The only modulation that influenced sleep EEG was the “talk-mode” signal (simulating the exposure when talking into a GSM handset), where sleep latency was delayed compared to the other exposure situations. In contrast, Fritzer et al. (2007) found neither short- nor long-term effects on sleep or on a battery of cognitive tasks when subjects were exposed to GSM signals (approximately 1 W/kg) during entire nights.

Due to the vicinity of the mobile phones to various head and face structures during use, there is interest in finding out if sensory functions are affected by exposure to mobile phone signals and specifically their RF components. In particular, properties of the acoustic and visual senses are of interest. Since the previous opinion, several papers have been published that experimentally investigated if electric potentials and muscle activity concomitant with hearing and vison are influenced by exposure. No effects relating to exposure were found in any of the studies investigating hearing parameters (Stefanics et al. 2007, Parazzini et al. 2007, Bamiou et al. 2008, Paglialonga et al. 2007, Cinel et al. 2007, Kleinlogel et al. 2008a) or visual parameters (Irlenbusch et al. 2007, Bamiou et al. 2008, Terao et al. 2007, Kleinlogel et al. 2008b).

Animal studies

There are recognised animal murine and rodent models that are very suitable for studies of various kinds of cognitive functions and behaviour. However, few studies have been performed recently that address the question if RF exposure typical for mobile phones has any neurological effects in animals. The exceptions have used long-term exposure to GSM signals in protocols for studies of learning and memory in rats. Nittby et al. (2008) found that 2 h exposure per week for 55 weeks at SAR levels as low as 0.6 and 60 mW/kg could impair object memory, whereas exploratory behaviour and spatial memory was not affected. At considerably higher SAR values (0.3 and 3.0 W/kg), Kumlin et al. (2007) found that an exposure for 2 h/day, 5 days/week, for 5 weeks did not have any effects on several behaviour parameters or on the cellular/histological appearance of the brain. However, two end-points, i.e. memory and learning, as evidenced by the water maze test, improved after exposure to these RF fields. Interestingly, this study employed young male rats, and thus investigated the developing brain.

It was previously suggested that RF exposure at very low levels could irreversibly damage the blood-brain barrier, causing albumin leakage into the brain from the vasculature and also to changes in neuronal appearance (“dark neurons”) (Salford et al. 2003). Such effects, if substantiated by replication studies and follow up experiments, could possibly indicate very detrimental consequences of mobile phone use. It was noted in the previous opinion that other groups did not find such effects. Since then, a few papers have studied blood-brain-barrier functions and brain histology. In most studies, no effects have been noted, even at relatively high SAR values (up to 4.8 W/kg) (Masuda et al. 2007a, Masuda et al. 2007b, Kumlin et al. 2007, Grafström et al. 2008). An exception is a study by Eberhardt et al. (2008), where rats were exposed for 2 h to a 900 MHz GSM signal yielding average whole body SAR values of 0.12, 1.2, 12, and 120 mW/kg. After exposure, rats were left until day 14 or 28 post exposure after which they were sacrificed and brain sections were investigated for parameters indicating blood- brain-barrier damage (albumin extravasation, albumin in neurons, dark neurons). The evaluation of the brain slices was performed as a subjective analysis/assessment of the investigated parameters in a blinded manner. Albumin extravasation was reversibly increased in the rats 14 days after exposure, at 120 mW/kg (p<0.01). Albumin uptake into neurons was also found at day 14 after exposure, but not at day 28. The effect was most pronounced at the lowest intensity, 0.12 mW/kg and also at 1.2 and 12 W/kg. Finally, dark neurons were significantly more numerous at 28 days post exposure, at 0.12 mW/kg and 1.2 W/kg. These remarkable findings regarding strongest effects at the lowest SAR values are in line with a previous paper from the group (Persson et al. 1997), but contradict their findings reported by Salford et al. (2003) where the strongest effects on dark neuron appearance was at the highest SAR value (200 mW/kg). It is difficult to evaluate these findings, particularly as the authors themselves do not have an explanation for the surprising results. However, the experiments do not include any positive controls which could show the effect level of blood-brain-barrier opening on the chosen end-points and thus what could be considered to be normal variation and strong effects, respectively. Furthermore, the subjective scoring of the microscopy slides may lead to variation in response determination and could pose evaluation problems.

Two papers demonstrate activation of glial cells and possible gliosis after 900 MHz GSM exposure, both after a single exposure for 15 min at 6 W/kg (Brillaud et al. 2007), and after chronic exposure (15 minutes a day, 5 days a week for 24 weeks) at the same SAR value (Ammari et al. 2008a), whereas exposure to 1.5 W/kg did not cause any glial cell activation.

Discussion

With the exception of a few findings in otherwise negative studies, there is no evidence that acute or long-term RF exposure at SAR levels relevant for mobile telephony can influence cognitive functions in humans or animals. There is some evidence that RF exposure influences brain activity as seen by EEG studies in humans. Human studies also indicate the possibility of effects on sleep and sleep EEG parameters. However, certain findings are contradictory and are furthermore not substantiated by cellular studies into mechanisms. There is a need for further studies into mechanisms that can explain possible effects on sleep and EEG.

There is no evidence that acute exposures to RF fields at the levels relevant for mobile telephony have effects on hearing or vision. Furthermore, there is no evidence that this kind of exposure has direct neurotoxicological effects. Most studies show lack of effects on supporting structures like the blood-brain-barrier. The positive finding is lacking dose- response relationships and needs independent replication in studies with improved methodology. The findings of activated glial cells at relatively high SAR-values could indicate gliosis and thus subsequent neurodegeneration after exposure, although exposures at lower levels did not reveal any such effects.

Source & ©: SCENIHR,  Health Effects of Exposure to EMF (2009),
3.3.4. Nervous system effets, p. 28-31

4.3 Have effects of mobile phones on reproduction and development been reported?

The SCENIHR opinion states:

3.3.5. Reproduction and development

What was already known on this subject?

The previous opinion of 2007 discussed epidemiological and animal studies on adverse developmental effects of RF fields. Numerous animal studies have clearly shown that RF fields are teratogenic at exposure levels that are sufficiently high to cause significant (>1°C) increase of temperature. There was no consistent evidence of adverse effects at nonthermal exposure levels. The limited number and statistical power of epidemiological studies available at that time, as well as their inconsistent findings precluded any definite conclusions.

What has been achieved since then?

Development

A recent study on a big Danish cohort found that children aged seven whose mothers had used mobile phones either during or after pregnancy had increased overall scores for behavioral problems (Divan et al. 2008). In light of the very low exposure to the children that would occur as a consequence of the mothers’ use of the phone during or after pregnancy it is doubtful that RF exposure from mobile telephony could have anything to do with the observed association. Yet, the explanation for this association is unknown at this time.

Two recent studies have evaluated developmental effects of RF fields in animals. Odaci et al. (2008) investigated effects of prenatal exposure to a 900 MHz field (60 min/day) on the number of granule cells in the dentate gyrus of the hippocampus of young rats. Three pregnant rats were used in both the exposed and the control groups. The brains of six offspring from the exposed group and five offspring from the control group were examined at the age of 4 weeks. The exposure level was described as a “peak” SAR of 2W/kg, but details of the dosimetry were not given. The number of granule cells was 20% lower (p<0.01) in the exposed group than in the control group, suggesting that prenatal exposure to RF fields might cause inhibition of granule cell neurogenesis. No conclusions can be drawn from this study because of the low number of animals and inadequate reporting of dosimetry. Batellier et al. (2008) studied embryo mortality in fertile chicken eggs exposed to 900 MHz mobile phone fields. The exposure was generated by a mobile phone that was programmed to call once at 3 min intervals during the entire period of incubation. A single mobile phone was placed over a group of 60 eggs, resulting in a very inhomogenous exposure level. Significantly higher mortality was observed in the exposed group compared to the sham-exposed group. However, this difference was obvious in only two of four replicate experiments, and there was no significant dependence on exposure level (which varied with distance to the mobile phone). Interpretation of the results of this study is difficult because of the poorly controlled exposure and lack of proper dosimetry.

Reproduction

Two cross-sectional studies have examined fertility among men exposed to RF fields in the Norwegian Navy. Møllerløkken and Moen (2008) collected data from 1,487 military men (response rate 63%) using a questionnaire covering exposure to electromagnetic fields, lifestyle, reproductive health, previous diseases, work and education. Exposure to RF fields was assessed by an expert group; the work categories “tele/communication”, “electronics” and “radar/sonar” were considered as being exposed. Self-reported infertility assessed by a single question (“Have you and your partner ever tried for more than 1 year to get pregnant without success?”) was associated with working in the tele/communication category (OR=1.72; 95% CI: 1.04-2.85) and in the radar/sonar category (OR=2.28; 95% CI: 1.72-4.09). These results were obtained by logistic regression analysis adjusted for age, smoking, military education and physical exercise at work. However, the work categories did not differ with respect to objective measures of fertility (number of biological children and paternal age at birth of first child). No differences were observed in congenital anomalies, chromosomal errors, preterm births, stillbirths or infant deaths within 1 year. Associations with self-reported RF field exposure were found also for several self-reported diseases such as food and drug allergy, testicular cancer, cardiac infarction, and skin cancer. Baste et al. (2008) collected questionnaire data from 10,497 current and former male military employees of the Norwegian Navy. Self-reported exposure to RF electromagnetic fields (working close to equipment emitting such fields) was shown to be associated with self reported infertility (assessed as by Møllerløkken and Moen (2008)) by logistic regression analysis adjusted for age, smoking, alcohol consumption, and exposure to organic solvents, welding and lead. For self-reported work closer than 10 m to high frequency aerials to a “very high”, “high”, “some” and “low” degree, the odds ratios were 1.86 (95%CI: 1.46-2.37), 1.93 (95%CI: 1.55-2.40), 1.53 (95%CI: 1.25-1.84) and 1.39 (95%CI: 1.15-1.68). In addition, self-reported work within less than 3 m distance from communication equipment and within less than 5 m distance from radar were associated with self reported infertility. However, exposure to RF fields was neither associated with the ability of having biological children nor with the number of biological children. Self-reported exposures to high frequency aerials and communication equipment were associated with a decreased ratio of boys to girls. The weaknesses of these two studies include self- reporting of the endpoints, lack of objective assessment of RF field exposure and possibility of confounding factors such as long stays away from home (exposure to the RF field emitting equipment is associated with being on board of ships). Because of these limitations, it is not possible to draw conclusions about the potential causal role of RF fields.

Although the exposure of male reproductive organs to RF fields from mobile phone use is extremely low, two studies have addressed effects of mobile phone use on sperm quality among men attending infertility clinics. The authors reported that reduced sperm quality was associated with duration of daily exposure to mobile phones assessed by interview (Agarwal et al. 2008) and with duration of use of mobile phones assessed by questionnaire (Wdowiak et al. 2007). However, possible confounding due to lifestyle differences (associated with differences in the use of mobile phones) may have biased the results of both studies.

Two animal studies have addressed the effects of RF fields on male fertility. Dasdag et al. (2008) found no effects on caspase-3 activity (used as a measure of apoptosis) in the testes of male rats exposed to 900 MHz GSM-type fields 2 h/day for 10 months. Fourteen animals were exposed, seven animals served as sham-exposed controls and ten as cage controls. The dosimetry of the testes is not sufficiently characterised. The negative finding of this study has limited value, because only one endpoint was measured. Yan et al. (2007) exposed young male rats (eight animals per group) to mobile phone emissions 6 h/day for 18 weeks. Exposures were performed by placing a mobile phone at a distance of 1 cm from the head of immobilised animals. No information is given on how the mobile phone emissions were controlled during exposure, and there is no description of dosimetry. However, it is obvious that exposure of the testes must have been extremely low. No effects were observed on total sperm cell count or structural abnormalities of sperm cells. Sperm cell death was significantly higher in the exposed group compared to the control group. However, this seems to be mainly related to an unusually low number of live cells in two animals of the exposed group. Abnormal clumping of sperm cells and increased mRNA levels of two cell surface adhesion proteins, CAD-1 and ICAM-1, was also reported in the exposed animals. No conclusions can be drawn from this study because of the poorly controlled exposure and the possibility of individual differences not related to exposure.

Discussion

The recent studies that addressed RF field effects on prenatal development in animals and the association of maternal mobile phone use with behavioural effects in children have not provided new information that would change the conclusions of the previous opinion that there are no adverse effects at nonthermal exposure levels.

Studies on male fertility are inadequate due to low statistical power and/or methodological problems.

3.3.6. Miscellaneous human

The previous report concluded that there are no substantiated indications for other (miscellaneous) health effects and no studies that change this have been published since the previous report.

Source & ©: SCENIHR,  Health Effects of Exposure to EMF (2009),
3.3.5 Reproduction and Development and 3.3.6 Miscellaneous human, p.31-33

4.4 Are children more vulnerable to possible effects of mobile phones?

The SCENIHR opinion states:

3.3.7. Dosimetric aspects of children's exposure

Due to the increasing discussion of the potential vulnerability of children to RF fields it was decided to include this section on dosimetric aspects of children’s RF exposure in this opinion. Relevant citations published earlier than 2007 have also been included since this subject was not covered in the previous opinion. Certain other aspects of children’s exposure are already included in previous chapters, e.g. on childhood leukaemia in chapter 3.3.2.1 or on life time exposure of test animals (chapter 3.3.4)

Concerns about the potential vulnerability of children to RF fields have been raised as their nervous system is developing and therefore potentially more susceptible than the nervous system of adults. Another aspect is that starting to use mobile phones in childhood will result in a longer cumulative lifetime exposure compared to starting to use mobile phones as an adult. While the anatomical development of the nervous system is completed at around two years of age, functional development continues up to adulthood and could possibly be disturbed by RF fields. Although children do not usually use mobile phones before two years of age, they can nevertheless be exposed from sources such as the recently introduced DECT baby phones.

Few relevant epidemiological or laboratory studies have addressed the possible effects of RF field exposure on children. Owing to widespread use of mobile phones among children and adolescents and relatively high exposures to the brain, investigation of the potential effect of RF fields in the development of childhood brain tumours is important. However, the characteristics of mobile phone use among children, their potential biological vulnerability and longer lifetime exposure render extrapolation from adult studies difficult.

Many scientific questions such as possible differences of the dielectric tissue parameters remain open. Several studies demonstrated a decrease of the dielectric properties permittivity and conductivity of animal tissue with age (Gabriel 2005, Martens 2005, Schmid and Überbacher 2005, Peyman et al. 2007). Possible reasons for this are the decrease of water content in the tissue with increasing age, increased myelination of the neurons and changes of the thickness of the dura matter. However, the extrapolation from animal data to children remains questionable. In addition there are still considerable uncertainties regarding the extrapolation from dead tissue to living conditions. Although there are studies on post-mortal human tissue (Schmid et al. 2003b) and living porcine tissue (Schmid et al. 2003a), additional studies are recommended.

There are conflicting opinions regarding possible differences in RF absorption between children and adults during mobile phone usage (Wiart et al. 2005, Christ and Kuster 2005b). The outcomes of existing studies are not consistent. The way the human is modelled is one important factor which explains the different views. The investigation of the exposure of humans in electromagnetic fields and analysis of the arising electromagnetic field distributions inside the human body requires the use of sophisticated numerical software tools. For this purpose numerical phantoms of humans were developed which allow very accurate calculation of the field distributions inside different parts of the body. However, such numerical phantoms may be representing the general population to a limited extent due to the great variability in morphology among humans. Thus, results from one or a few phantom models may be insufficient. Consequently, it is necessary to use a “family” of phantoms that have different anatomical and morphological characteristics, and also to use appropriate statistical approaches when analysing the obtained data (Conil et al. 2008). Additional factors that have to be considered are the impact of the hand on the exposure, the impact of external objects like glasses, the pinna thickness and elasticity, and the design of the phone and the antenna matching (Christ et al. 2005a, Fernández et al. 2005, Hadjem et al. 2005a, Hadjem et al. 2005b, Beard et al. 2006, Lee et al. 2007, Wiart et al. 2007).

Protection limits given in international guidelines, standards and other documents (e.g. ICNIRP 1998) are intended to protect the population against adverse effects arising due to the exposure to electromagnetic fields from 0 Hz to 300 GHz. To ensure reaching the protection goal of these guidelines, compliance with the basic restriction specific absorption rate is warranted. Because the measurement of the SAR is very challenging, reference levels of the electric and magnetic field strength were defined. These magnitudes are rather easy to assess: compliance with the reference levels should guarantee compliance with the basic restrictions.

However, recent studies on whole body plane wave exposure of both adult and children phantoms demonstrated that when children and small persons are exposed to levels which are in compliance with reference levels, exceeding the basic restrictions cannot be excluded (Dimbylow and Bolch 2007, Wang et al. 2006, Kühn et al. 2007, Hadjem et al. 2007). While the whole frequency range has been investigated, such effects were found in the frequency bands around 100 MHz and also around 2 GHz. For a model of a five year old child it has been shown that when the phantom is exposed to electromagnetic fields at the reference levels, the basic restrictions were exceeded by 40% (Conil et al. 2008). Several factors are relevant for the specific exposure conditions, e.g. size, anatomy, BMI (Body Mass Index). Moreover, a few studies demonstrated that multipath exposure can lead to higher exposure levels compared to plane wave exposure (Neubauer et al. 2006, Vermeeren et al. 2007).

It is important to realise that this issue refers to far-field exposure only, for which the actual exposure levels are orders of magnitude below existing guidelines.

The exposure of possibly sensitive groups of the population such as children should be investigated using adequate numerical phantoms taking multi-source and multi-path exposure conditions into account. Finally, such investigations should not be restricted to the radio frequency range only.

Source & ©: SCENIHR,  Health Effects of Exposure to EMF (2009),
3.3.7, Dosimetric aspects of children’s exposure, p.33-35.