Content uploaded by Martin Röösli
Author content
All content in this area was uploaded by Martin Röösli on Jul 23, 2017
Content may be subject to copyright.
Review
Electromagnetic hypersensitivity (EHS) and subjective
health complaints associated with electromagnetic fields of
mobile phone communication—a literature review published
between 2000 and 2004
H. Seitz
a,
*, D. Stinner
a,1
, Th. Eikmann
a,1
, C. Herr
a,1
,M.Ro¨o¨sli
b,2
a
Institute of Hygiene and Environmental Medicine, Justus-Liebig-University Giessen, Friedrichstraße 16, 35392 Giessen, Germany
b
Department of Social and Preventive Medicine, University of Bern, Finkenhubelweg 11, 3012 Bern, Switzerland
Received 21 January 2005; accepted 9 May 2005
Available online 21 June 2005
Abstract
Literature published between 2000 to 2004 concerning electromagnetic fields (EMF) of mobile communication and
electromagnetic hypersensitivity (EHS) or unspecific symptoms of ill health, respectively, is reviewed. Basically, literature
from established databases was systematically searched for. For each study, the design and quality were evaluated by means of a
criteria list in order to judge evidence for causality of exposures on effects. Finally, 13 studies of sufficient quality were
considered for this review.
In only one provocation study, individuals with self-reported electromagnetic hypersensitivity were exposed to EMF.
Their perception of field status was no better than would have been expected by chance. Results of five randomised
cross-over studies on impaired well-being due to mobile phone exposure were contradictory. Even though these studies
would allow more reliable exposure assessment, they are limited due to short exposure period and the small study
size.
No firm conclusion could be drawn from a few observational epidemiological studies finding a positive association between
exposure and unspecific symptoms of ill health due to methodological limitations. Causality of exposure and effect was not
derivable from these cross-sectional studies as field status and health complaints were assessed at the same time. In addition,
exposure assessment has not been validated.
0048-9697/$ - see front matter D2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.scitotenv.2005.05.009
* Corresponding author. Tel.: +49 641 99 41451; fax: +49 641 99 41459.
E-mail address: heike.seitz@hygiene.med.uni-giessen.de (H. Seitz).
1
Tel.: +49 641 99 41451; fax: +49 641 99 41459.
2
Tel.: +41 31 631 3511; fax: +41 31 631 35 20.
Science of the Total Environment 349 (2005) 45 – 55
www.elsevier.com/locate/scitotenv
In conclusion, based on the limited studies available, there is no valid evidence for an association between impaired well-
being and exposure to mobile phone radiation presently. However, the limited quantity and quality of research in this area do
not allow to exclude long-term health effects definitely.
D2005 Elsevier B.V. All rights reserved.
Keywords: Electromagnetic hypersensitivity; Unspecific health complaints; Subjective symptoms; Mobile phone; Electromagnetic fields
Contents
1. Introduction .................................................... 46
2. Methods ...................................................... 47
2.1. Outcome selection ............................................. 47
2.2. Literature selection ............................................. 47
2.3. Literature search .............................................. 47
2.4. Quality evaluation ............................................. 48
3. Results....................................................... 48
3.1. Selection of Studies ............................................ 48
3.2. Perception of electromagnetic fields ................................... 49
3.3. Electromagnetic hypersensitivity ..................................... 49
3.4. Sleep quality ............................................... 51
3.5. Dizziness ................................................. 51
3.6. Headache ................................................. 51
3.7. Skin problems ............................................... 51
3.8. Problems in concentration and memory loss ............................... 52
3.9. Nervousness................................................ 52
3.10. Other unspecific symptoms of ill health ................................. 52
4. Discussion ..................................................... 52
5. Conclusion .................................................... 54
Acknowledgement ................................................... 54
References ....................................................... 54
1. Introduction
Public discussion on possible health risks from
electromagnetic fields even below the legal thresholds
is common (Sage, 2001). As mobile communication is
being increasingly established, this discussion is fo-
cussing on health impairment associated with mobile
phones and base stations. Still, ongoing rapid, nation-
wide implementation of this technology is accompa-
nied by its wide use in the population. The heath risk
discussion focuses on carcinogenic and bioregulatory
effects as well as impaired well-being, i.e., headache,
sleep disturbances or problems in concentrating,
sometimes referred as electromagnetic hypersensitiv-
ity (Silny et al., 2004).
The term electromagnetic hypersensitivity (EHS)
relates to subjects attributing health symptoms to an
exposure to electromagnetic fields. In population-
based surveys, prevalence of EHS was reported to
be 1.5% in Sweden (Hillert et al., 2002) and 3.2%
in California (Levallois et al., 2002). In this context,
EHS can be related to radio and microwave frequency
fields of mobile communication, fields in the kilohertz
range of display units as well as extremely low-fre-
quency fields of domestic power supply. It should be
noted that the term EHS exclusively depicts self-
reported hypersensitivity. So far, objective criteria
classifying these subjects as hypersensitive have not
been established. In analogy to environmental patients
reporting Multiple Chemical Sensitivities (MCS) or
H. Seitz et al. / Science of the Total Environment 349 (2005) 45–5546
Sick-Building–Syndrome (SBS), individuals with
EHS describe multiple non-specific health complaints
(David et al., 2002; Hietanen et al., 2002; Raczek et
al., 2000). Many of these complaints are subjective,
e.g., redness and burning sensation of the skin and
face, tingling, pain and dryness of the mouth and
throat, problems in concentrating, nervousness or
headache. The health complaints related to EHS result
in considerable psychological stress in these patients.
Due to a lack of knowledge of the pathophysiology of
this complex of symptoms, adequate medical treat-
ment for these patients is difficult.
An additional phenomenon in this context is the
proclaimed ability to perceive electromagnetic fields
at a much lower threshold than the general population
without necessarily developing health symptoms
(Leitgeb and Schro¨ ttner, 2003). Therefore, the pres-
ence of considerably decreased perception threshold,
on one hand, and the attribution of health symptoms
to EMF exposure, on the other hand, can be consid-
ered as two independent phenomena. Nevertheless, in
a survey among self-declared EHS individuals, 56%
declared their ability to perceive electromagnetic
fields (Ro¨o¨sli et al., 2004). The ability to perceive
EMF at a considerably decreased threshold can be
investigated by so-called provocation studies. In con-
trast, to investigate a possible association between
symptoms and EMF exposure is methodologically
more challenging, in particular effects demonstrable
after long-term exposure or with some latency period
only.
The aim of the presented review of the literature is
to elucidate whether EHS can be considered an entity
and whether subjects with higher electrosensitivity
toward frequencies from mobile communications ac-
tually exist.
2. Methods
2.1. Outcome selection
Electromagnetic hypersensitivity as an entity and a
considerably decreased perception threshold was con-
sidered to be outcomes of priority. In addition, the most
common unspecific symptoms of ill health reported in
relation to exposure to electromagnetic fields were
derived from the literature and viewed as further out-
comes. These were the following: fatigue/difficulties in
sleeping, dizziness/nausea, headache, disturbances in
concentrating and memory, pain other than headache,
nervousness, depressive mood/or state, skin-related
sensation, e.g., itching, tickling, redness, burning or
increased temperature of the skin including warm sen-
sation of the ear, tinnitus/ringing of the ear.
2.2. Literature selection
Literature search was based on a number of criteria.
Eligible studies had to be published between 2000 and
March 2004 and had to be related specifically to
electromagnetic fields from mobile communication
(mobile phones or base stations), i.e., analogue signals
used in the past as well as digital signals (GSM,
UMTS). Only original papers but not reviews were
considered. Besides peer-reviewed journal contribu-
tions so-called bgrey literatureQwas included whenev-
er possible. These publications had to be informative
enough to allow a critical quality evaluation, e.g.,
comprehensive study reports. Short conference ab-
stracts were not considered to supply sufficient infor-
mation. Applying this procedure was to insure that all
of the relevant studies relating to the selected outcomes
were assessed.
2.3. Literature search
Literature search was performed independently by
two teams in order to reach a maximum completeness.
Eligible literature was systematically searched for in
The National Library of Medicine (Pubmed: www.
ncbi.nlm.nih.gov/PubMed/). In doing so the MeSH
search term bradiation, nonionizing/adverse effectsQ
was combined with numerous single words describing
the outcome of interest, e.g.: Tbradiation, nonionizing/
adverse effectsQ[MeSH] and headacheror TbEMFQ
[MeSH] and nausear. First triage was based on the
abstract. The completeness of the search was checked
with three strategies: (i) reference lists of review ar-
ticles were examined with respect to papers and grey
literature which fulfilled inclusion criteria; (ii) topic-
specific online data bases were consulted: Research
Centre Ju¨lich (http://134.94.127.170:8087/), the Uni-
versity of Basle (www.elmar.unibas.ch/index.html)
and the Research Center for Bioelectromagnetic In-
teraction in Aachen (www.femu.de/); (iii) related ar-
H. Seitz et al. / Science of the Total Environment 349 (2005) 45–55 47
ticle search was used in Pubmed as well as in the
online database of the Institute for Scientific Informa-
tion (ISI: www.isinet.com/).
In order to identify studies that addressed subjec-
tive health complaints only marginally, as a secondary
outcome, experimental studies assessing, e.g., brain
physiology or cognitive functioning in relation to
exposure to mobile phones or base stations were
also systematically screened.
2.4. Quality evaluation
The value of studies on possible health effects and
exposure is predominantly determined by their design.
Evidence for causality of exposures on effects can
only be obtained from studies excluding possible
bias; e.g., randomised double blind trials using well-
defined exposure settings. Cross-sectional studies
(surveys), measuring health effects and exposure at
the same time, are not able to depict the causal
relationship between exposure and effects, especially
if these effects are subjective complaints reported by
the subjects.
For these reasons, selected literature was weighted
according to study design for this review. In a second
step, study quality was judged. This judgement was
based on criteria for the evaluation of epidemiological
studies on radiation published by the Commission for
Radiation Protection (SSK, 2002). Applying a ten-
page questionnaire, each study was assessed as far
as wording of the aims and problems, study concept,
quality assurance, evaluation as well as interpretation
and discussion of results are concerned. Most impor-
tant minimal quality criteria which experimental stud-
ies had to fulfil were (i) adequate sample size; (ii)
blinding of the study participants; (iii) randomisation;
(iv) exposure characterized. Minimal quality criteria
for observational studies were (i) comprehensible
study participants recruitment, (ii) allowing for possi-
ble confounding factors in the analysis (at least age
and sex), (iii) approved exposure assessment with
respect to a reproducible personal exposure.
This procedure allowed for deriving four categories
of evaluation (Table 1).
Studies of sufficient quality with designs appropri-
ate to examine causal relationships are able to give
evidence as to whether (or not) and possibly to what
extent exposure to electromagnetic fields influences
occurrence of complaints. Studies of sufficient quality
with a study design not appropriate for finding causal
relationships are able to give indication or can be used
to derive hypotheses without contributing to evidence.
Studies of poor quality are not able to contribute to the
evaluation (Table 1).
3. Results
3.1. Selection of Studies
A total of 20 papers and two reports from the grey
literature were considered relevant and preselected.
Thereof, 5 papers were excluded due to double pub-
lication of the same material. Additional four papers
were excluded as they did not meet quality criteria:
the studies of Navarro et al. (2003) and Santini et al.
(2002, 2003a,b) did not state the applied methods of
subject recruitment, and the exposure assessment
based on distance was judged to be inadequate; as
well possible confounding factors were not allowed
for in the analysis (e.g., age and sex). Hietanen et al.
(2002) did not allocate actual and sham exposure in a
random counterbalanced way. Therefore, effects
resulting from exposure versus from sequence cannot
be differentiated. Ozturan et al. (2002) reported non-
blinded experimental results.
Overall, 11 papers and two reports remained for
this review. Thereof, seven publications were obser-
vational studies (Table 2). These non-experimental
studies were not considered adequate for examining
causal relationship. They are only able to build the
basis for generating hypotheses and indications
concerning associations. Four papers and two reports
described experimental studies allowing to draw con-
clusions on causality of unspecific health complaints
and exposure to electromagnetic fields of mobile
Table 1
Scheme of assessment for studies
Suitable
a
study
design
Unsuitable
a
study
design
Sufficient quality Evidence Hints/Hypothesis
Quality inadequate
a
Useless Useless
The suitability of the study’s design concerning evidence of causal
effects and the study’s quality are taken into consideration.
a
Suitability and assessment of quality referred to the aim of this
analysis of literature.
H. Seitz et al. / Science of the Total Environment 349 (2005) 45–5548
communication (Table 3). From these, three studied
health complaints as primary outcomes, the remaining
as secondary outcome. From the latter, only sparse
data about EHS or well-being can be obtained.
3.2. Perception of electromagnetic fields
A first group of studies consisted of blinded prov-
ocation studies. In theses studies, volunteers had to
state whether a source was actually emitting elec-
tromagnetic radiation or not. Therefore, they were
appropriate to give evidence for the ability of elec-
tromagnetic field perception at a low threshold.
Within one of the provocation studies 16 subjects
with self-reported EHS were individually, and as a
group, not able to perceive the field status more often
than expected by chance (Raczek et al., 2000).
According to the methods of one further double-
blinded cross-over study, exposure status was
assessed but results not given (Tahvanainen et al.,
2004).
3.3. Electromagnetic hypersensitivity
A second group of studies assessed exposure effects
on health status in general, not on specific complaints.
These studies, performed on subjects with and without
reported EHS, were able to test the hypotheses of
individually different health complaints in association
to exposure. In one Australian study were changes
reported in the neural response during exposure to
GSM 900 mobile phones, which were correlated with
the self-reported health status. The health status was
assessed by a symptom check list (activation–deacti-
Table 2
Observational studies
Reference Study design Collectives Sample size Country/
City
Outcome Exposure Exposure
assessment
method
Chia et al.,
2000
Cross-sectional Random
population
sample
808 Singapore Headache, tingling,
burning, sense of warmth,
tiredness, loss of memory,
difficulty in concentration
dizziness, visual disturbance
Mobile
phone
Questionnaire
Hillert et al.,
2001
Cohort Self-declared
EHS
14 Norway Fatigue None –
Sandstro¨m
et al., 2001
a
Cross-sectional Employee of
companies with
mobile telephone
subscription
Norway: 2,828
Sweden: 7,803
Norway
and Sweden
Dizziness, discomfort,
concentration, memory loss,
fatigue, headaches, warmth
behind/on ear, burning skin,
tingling
Mobile
phone
Questionnaire
Santini et al.,
2001
Cross-sectional Students and
employee of an
engineer school
161 France Headaches, concentration,
memory loss, fatigue, sleep
disturbance, discomfort,
sense of warmth or burning,
tingle
Mobile
phone
Questionnaire
Frick et al.,
2002
Cross-sectional Random
population
sample
340 Germany/
Regensburg
Numerous health
complaints
None –
Hutter et al.,
2002
Cross-sectional Residents of a
mobile phone
base station
365 Austria Sleep quality, cognitive
performance, complaints
related to exhaustion,
digestive tract and
cardiovascular system
Base
station
Measurements
Ro¨o¨ sli et al.,
2004
Cross-sectional Self-declared
EHS
394 Switzerland Various health complaints Mobile
phone and
base station
Questionnaire
a
Several publications of the same study: Oftedal et al., 2000; Sandstro¨m et al., 2001; Wilen et al., 2003.
H. Seitz et al. / Science of the Total Environment 349 (2005) 45–55 49
vation check list by Thaya: individual health status is
judged with a four-point scale using 20 opposite pairs
of words). The mode of correlation is not given in the
paper (Croft et al., 2002). In the Dutch TNO study weak
exposure to UMTS was statistically significantly cor-
related with changes in health status in 36 subjects with
and 36 without electromagnetic hypersensitivity. In this
study, health status was assessed by a sum index of a
symptom questionnaire. On the contrary, exposure to
GSM did not influence well-being (Zwamborn et al.,
2003). In an experimental double-blinded cross-over
study giving no methodological details about health
complaint measurement, no differences in complaints
for actual and sham exposure were reported (Tahvanai-
nen et al., 2004).
One cross-sectional study assessed health status in
association with measured field levels at home from
base stations of mobile communication in 365 indivi-
duals (Hutter et al., 2002). Reported health was
grouped in complaints related to exhaustion, digestive
tract and cardiovascular system according to the com-
plaint list of Zerssen. After controlling for possible
confounding factors (age, sex, use of mobile phones
and health concerns), complaints related to the car-
diovascular system, but not the other groups of com-
plaints were significantly associated with measured
field levels from base stations.
Further studies addressed the question whether
physiological and laboratory parameters were differ-
ent in patients with EHS and reference groups without
complaints. To determine this, 14 patients with EHS
who reported excessive fatigue were examined. Ex-
cessive fatigue was not associated with changes in
cholinesterase (Hillert et al., 2001). Thus, it was not
possible to objectify this subjective complaint by a
laboratory parameter.
One further physiological parameter applied to
objectify heightened sensitivity to metal compounds
Table 3
Experimental studies
Reference Study
design
Collective nper
experiment
Outcome Exposure Exposure
duration
Intensity Blinding
Huber et al.,
2000
Cross-over 16 male, right
handed
(20–25 years)
16 Subjective sleep
quality
a
Mobile
phone
30 min SAR = 1 W/kg Double
Raczek
et al.,
2000
Cross-over Self-declared
EHS: 9 female,
7 male
(31–76 years)
16 Field perception Mobile
phone
21 tests with
3 min exposure
Incident field:
16 V/m
Double
Koivisto
et al.,
2001
Cross-over 48 male and 48
female (18–49
years)
48 Headaches,
dizziness, fatigue,
tingling, redness
on skin, sense
of warmth
Mobile
phone
Experiment 1:
60 min experiment
2: 30 min
Power: 0.25 W Single
Croft et al.,
2002
Cross-over 16 male and
8 women
(19–48 years)
24 Mood
a
Mobile
phone
20 min SAR = 0.0034W/kg Single
Zwamborn
et al.,
2003
Cross-over Self-declared
EHS: 11 male
25 female
(31–74 years)
reference group:
22 male 14 female
(18–72 years)
36 Well-being
(sum score)
Base
station
30 min Incident field:
1 V/m
Double
Tahvanainen
et al.,
2004
Cross-over 16 male and 16
female (average
age: 38.8)
16 Field perception
subjective
symptoms (not
specified)
a
Mobile
phone
35 min SAR = 1.6 W/kg
(GSM 900)
SAR = 0.7 W/kg
(GSM1800)
Double
a
Secondary outcome (main outcome of the study was not well-being related).
H. Seitz et al. / Science of the Total Environment 349 (2005) 45–5550
in EHS patients is the lymphocyte transformation test
(LTT). In a study aiming to show electrosensitivity
towards reactions to the German D-net, various metal
compounds were found and thereby a possible elevat-
ed reaction towards electromagnetic fields was de-
rived (Raczek et al., 2000). It should be stressed,
however, that a specific LTT is able to demonstrate
sensitising but not the effector reaction, i.e., previous
contact to an antigen but not clinical manifestation of
sensitivity is shown. Clinical relevance of the LTT in
terms of diagnoses of environmental illnesses has not
been established and is still an issue for future re-
search (RKI 2002).
3.4. Sleep quality
In German-speaking countries, fatigue and distur-
bances in sleep quality belong to the most frequent
complaints related to EMF (Ro¨o¨ sli et al., 2004). While
a number of experimental studies have investigated
electromagnetic field effects on EEG during sleep,
effects on subjective rating of the sleep quality have
not been studied thoroughly. Huber et al. (2000) did
not find that exposure to EMF from mobile phones
changed the reported quality of sleep in 16 young
male subjects. Likewise exposure to a GSM mobile
phone for 60 min did not affect daytime fatigue in two
further experimental studies each with 24 individuals
(Koivisto et al., 2001). Contradictory results have
been reported from observational studies on associa-
tions of daytime fatigue and quantity of mobile phone
use (Chia et al., 2000; Sandstro¨ m et al., 2001; Santini
et al., 2001). Potential confounding factors taken into
account in the Singapore study (Chia et al., 2000)
were age, sex, occupation and extent of onscreen
work; in the Scandinavian study (Sandstro¨ m et al.,
2001), extent of onscreen work was considered, and in
the French study (Santini et al., 2001), no confounder
was taken into account. A cross-sectional study from
Austria (Hutter et al., 2002) did not find an association
between reported quality of sleep and exposure to
base stations of mobile communication.
3.5. Dizziness
Concerning dizziness, one randomised cross-over
study in 48 subjects did not find an association to
mobile phone handset exposure in two experiments
(Koivisto et al., 2001). Results of observational stud-
ies on mobile phone users were contradictory: while
one Norwegian study in 2800 individuals described an
association, studies from Sweden (n= 7800), Singa-
pore (n= 808) and a French University (n= 161) were
not able to confirm this (Chia et al., 2000; Sandstro¨m
et al., 2001; Santini et al., 2001).
3.6. Headache
One randomised cross-over study could not con-
firm an association between exposure to mobile
phones and headache (Koivisto et al., 2001). The
same applied also for headache resulting from an
experimental study on different electromagnetic fields
(GSM 900, GSM 1800, UMTS) from a base station
for mobile communication (Zwamborn et al., 2003).
On the other hand, two experimental studies found
significant associations between reports of headache
and use of mobile phones (Chia et al., 2000; Sand-
stro¨m et al., 2001). These results were consistent for
number of phone calls, time of use and type of tele-
phone (more symptoms when using analog phones
with higher output power than GSM). In a French
cross-sectional study, however, headache was not
reported more often by subjects using mobile phones
than those not using mobile phones (Santini et al.,
2001).
3.7. Skin problems
bItchiness, tingling and redness of skinQhave only
been studied once using a randomised cross-over
study (Koivisto et al., 2001). No increase in skin
problems was reported when using a GSM 900 mobile
phone for 60 min at a distance of 4 cm from the ear
compared to sham exposure. However, associations
were found in cross-sectional studies investigating,
e.g., btinglingQwith use of mobile phones. In a Scan-
dinavian study (Sandstro¨ m et al., 2001), these associa-
tions reached significance, while in Singapore, only a
tendency was reported (Chia et al., 2000) and no
association was found in a study from France (Santini
et al., 2001).
For increased temperature of the skin (burning and
warm ear) controlled experimental studies found no
associations (Koivisto et al., 2001). However, again,
in observational studies, associations were found.
H. Seitz et al. / Science of the Total Environment 349 (2005) 45–55 51
Even if mobile phone users did not report more
frequent feeling of warmness than non-users, these
complaints were reported significantly more often in
the subgroup of mobile phone users that phoned for
more than 2 min/day (Santini et al., 2001). Again in
the Scandinavian study, a consistent increase in warm-
ness was found with increasing exposure to mobile
phones (Sandstro¨ m et al., 2001), while in Singapore,
no difference was found for mobile phone users com-
pared to non-users (Chia et al., 2000).
3.8. Problems in concentration and memory loss
A number of experimental studies examined the
effect of mobile phone exposure on cognitive func-
tioning leading to contradictory results. However,
several times, an improvement (or acceleration) of
cognitive functioning was reported (Hamblin and
Wood, 2002). Tests of cognitive functioning depend
among others on power of concentration and memory.
The evaluation of the presented study only refers to
bproblems in concentrationQand bdisturbances in
memory,Qas they affect everyday life and can be
reported by the subjects. These complaints have not
been the object of controlled exposure studies even
though EHS patients report these complaints frequent-
ly (Ro¨o¨sli et al., 2004). In the cross-sectional study
from Singapore, no significant differences were found
in reported difficulties in concentration when mobile
phone users were compared to non-users (Chia et al.,
2000). In the Scandinavian study, an indication for
disturbances in concentration was observed for those
reporting more than four calls per day. In contrast,
memory loss was not associated with use of mobile
phones. In the French cross-sectional study, users of
mobile phones were not more likely to report pro-
blems in concentration and memory than non-users.
Among those using mobile phones, 1800 MHz phone
users were more likely to state disturbances in con-
centration than GSM 900 users. Likewise, more com-
plaints were reported by subjects using mobile as well
as stationary phones compared to those only using
mobile phones (Santini et al., 2001).
3.9. Nervousness
Incidence of nervousness during controlled expo-
sure was only studied for the TNO report and was
reported more frequently during UMTS exposure than
GSM exposure (Zwamborn et al., 2003).
3.10. Other unspecific symptoms of ill health
Other outcomes, e.g., depressive state and mood as
well as pain other than headaches and tinnitus, have
not been the focus of controlled exposure studies.
4. Discussion
The presented review of the literature comprising
the years 2000 through 2004 yielded a total of 13
studies with sufficient quality focussing on electro-
magnetic fields of mobile communication systems
and impaired well-being, i.e., health complaints or
electromagnetic hypersensitivity (EHS). The evalua-
tion depicted six randomised cross-over studies ap-
propriate to address causal associations of exposure
to EMF and well-being. Three of them focussed on
impaired well-being as primary outcome, the others
as secondary outcome. A further seven observational
studies fulfilled the quality standards but were not
designed to evaluate causality but only statistical
associations. Altogether, there was little evidence
for a particular population subgroup who is able to
perceive electromagnetic fields in an experimental
double blind setting at a much lower threshold
than the general population. No specific symptom
or symptom cluster was found to be caused by
exposure to mobile phone radiation. On the other
hand, based on the contradictory results of three
randomised cross-over studies, it cannot completely
be ruled out that exposure to EMF can lead to an
impaired well-being. If so, the studies to date would
indicate an individually different response and not a
typical bEMF-symptom clusterQ.
This review is based on a systematic compilation
and evaluation of possible effects of radiation from
mobile communication on well-being or unspecific
symptoms of ill health. Detailed evaluation of up-to-
date literature, including grey literature, became pos-
sible due to the additional focus on specific health
complaints and a limited number of years of publica-
tion. At the same time, this implies a clear restriction.
Older studies and other health effects such as carci-
nogenic or bioregulatory effects were not considered.
H. Seitz et al. / Science of the Total Environment 349 (2005) 45–5552
It should be noted that only a sparse amount of
literature focussed on health effects of mobile com-
munication before the year 2000. Likewise, papers on
EMF not related to exposure from mobile communi-
cation were not considered.
In the following, it will be discussed why there are
only relatively few and contradictory studies on this
topic although public interest and concern are obvi-
ous. Furthermore, the relevance of public health con-
cern is justified as the number of those exposed is
considerable.
It should be stressed that restrictions in well-
being, i.e., health complaints, are frequent in the
general population and individual reports of com-
plaints depend on a number of individual variables
(e.g., age, gender, social status, anxiety, accompa-
nying disease and personality traits) and individual
factors of exposure in everyday life. This clearly
shows that only well-designed studies allow to attri-
bute health complaints to one specific cause, i.e.,
exposure to radiation from electromagnetic field
from mobile communication.
In particular, observational studies on the phe-
nomenon EHS have two main restrictions: valid
exposure assessment and lack of objective criteria
for assessment of outcome, i.e., reported health com-
plaints. Reports of symptoms, complaints and well-
being are of subjective quality and are not or not
necessarily correlated with the clinical status
(Michel, 2004). Judgement of personal well-being
depends on individual characteristics (e.g., hypo-
chondria), present mood (Michel, 2004) as well as
context of questioning (Frick et al., 2002). Subjec-
tivity poses a problem for between subject compar-
isons and in particular, if long-term effects after a
latency period are addressed; both of which is stated
by some of those presuming to suffer of EHS. It
would be of utmost relevance to assess long-term
health effects. Cross-sectionals studies are often un-
derstood to detect long-term effects. However, due to
severe methodological limitations, no firm conclu-
sion can be drawn from the few published studies
on mobile phone radiation exposure. In cross-sec-
tional studies, exposure and effect are measured at
the same time. Thus cause and effect cannot be
differentiated. Furthermore, in published studies, ex-
posure assessment was based on a subjective state-
ment which has not been validated (use of mobile
phone) or on one single spot measurement (base
station). Co-exposures to other possible risk factors
were not taken into account, and only a few, if any at
all, potential confounding factors had been consid-
ered in the analysis. It is conceivable, e.g., that the
number of phone calls made is related to the indi-
vidual stress level which is in turn associated with
health complaints (Herr et al., in press). None of the
published studies made an attempt to take this aspect
into account. It is not surprising that so far only
experimental, randomised, cross-over designed stud-
ies with short exposure duration are able to give
evidence for or against causality of health status in
relation to exposure to mobile communication sys-
tems. However, experimental studies leave open a
number of unresolved issues even if the results had
been consistent across studies. Due to their design,
these experimental studies are only able to give
evidence for acute immediate effects. Long-term
effects cannot be studied or evaluated. A drawback
of experimental studies compared to epidemiological
studies is their small sample size. Thus, their power
to detect rare or subtle effects is small. Unfortunate-
ly, published studies in which no effect was found
did not state the effect size that could have been
detected with the respective study. Considering this,
the interpretation of results is difficult.
In addition, the investigation of EHS individuals
by means of laboratory studies is hampered by the
fact that many patients reporting EHS have psycho-
logical impairments and in part paranoid fears about
EMF (Ha¨rma¨, 2000). For this reason, many of the
patients do not agree to participate in studies of
unknown laboratory environment. The reasons
given for non-participation are fear of effects of the
experimental exposure or lack of objectivity within
the study.
An open issue remains the question whether po-
tential effects occur above a high, localised threshold
or whether cumulative total body exposure possibly
including a latency period is most relevant. Latter is
often suggested from EHS persons and would suggest
that beside exposure due to mobile phones them-
selves, mobile phone base station exposure may be
relevant. Following the concept of a highly localised
threshold would mean that only exposure to mobile
phones themselves could be considered relevant for
triggering health complaints.
H. Seitz et al. / Science of the Total Environment 349 (2005) 45–55 53
5. Conclusion
Only a restricted number of studies has been pub-
lished on well-being (i.e., health complaints) and
exposure to electromagnetic fields from mobile com-
munication between 2000 and March of 2004. The
results are contradictory and the greater part of these
studies is not able to address the issue of causality
between exposure and outcome. Therefore, an effect
of exposure to electromagnetic fields from mobile
communication on well-being cannot be derived
based on these limited studies. In order to obtain
more insights in the phenomenon EHS an interdisci-
plinary research effort is needed, including psycho-
logical, pathophysical, laboratory and epidemiological
disciplines as well as the improvement of personal
dosimetry.
Acknowledgement
This study was funded by the Forschungszentrum
Ju¨lich, Germany. Many thanks to Professor Robert
Snipes, University of Giessen, for the careful proof-
reading of this manuscript.
References
Chia SE, Chia HP, Tan JS. Prevalence of headache among handheld
cellular telephone users in Singapore: a community study. En-
viron Health Perspect 2000;108(11):1059.
Croft RJ, Chandler JS, Burgess AP, Barry RJ, Williams JD, Clarke
AR. Acute mobile phone operation affects neural function in
humans. Clin Neurophysiol 2002;113(10):1623.
David E, Reissenweber J, Wotjysiak A, Pfotenhauer M. On the
phenomenon of electromagnetic hypersensitivity. Umweltmed
Forsch Prax 2002;7(1):7.
Frick U, Rehm J, Eichhammer P. Risk perception, somatization, and
self report of complaints related to electromagnetic fields—a
randomized survey study. Int J Hyg Environ Health
2002;205:353.
Hamblin DL, Wood AW. Effects of mobile phone emissions on
human brain activity and sleep variables. Int J Radiat Biol
2002;78(8):659 – 69.
Ha¨rma¨ M. Electric hypersensitivity and neurophysiological effects
of cellular phones—facts or needless anxiety. Scand J Work
Environ Health 2000;26(2):85 – 6.
Herr CEW, zur Nieden A, Lindenstruth M, Stilianakis NI. Relat-
ing use of mobile phones to reported sleep quality. Somno-
logie, in press.
Hietanen M, Hamalainen AM, Husman T. Hypersensitivity symp-
toms associated with exposure to cellular telephones: no causal
link. Bioelectromagnetics 2002;23(4):264.
Hillert L, Flato S, Georgellis A, Arnetz BB, Kolmodin-Hedman B.
Environmental illness: fatigue and cholinesterase activity in
patients reporting hypersensitivity to electricity 5. Environ Res
2001;85(3):200.
Hillert L, Berglind N, Arnetz BB, Bellander T. Prevalence of self-
reported hypersensitivity to electric or magnetic fields in a
population-based questionnaire survey. Scand J Work Environ
Health 2002;28(1):33.
Huber R, Graf T, Cote KA, Wittmann L, Gallmann E, Matter D, et
al. Exposure to pulsed high-frequency electromagnetic field
during waking affects human sleep EEG. Neuroreport
2000;11(15):3321.
Hutter HP, Moshammer H, Kundi M. Mobile phone base stations:
effects on health and wellbeing. In: Kostarakis P. Rhodes7
Workshop; 2002. p. 344 – 52.
Koivisto M, Haarala C, Krause CM, Revonsuo A, Laine M, Hama-
lainen H. GSM phone signal does not produce subjective
symptoms. Bioelectromagnetics 2001;22(3):212.
Leitgeb N, Schro¨ttner J. Electrosensibility and electromagnetic
hypersensitivity. Bioelectromagnetics 2003;24(6):387.
Levallois P, Neutra R, Lee G, Hristova L. Study of self-reported
hypersensitivity to electromagnetic fields in California. Environ
Health Perspect 2002;110(Suppl. 4):619.
Michel G. Symptom reporting: exploration of situational and indi-
vidual predictors. PhD Thesis, University of Fribourg, Switzer-
land; 2004.
Navarro EA, Segura J, Portoles M, Gomez-Perretta de Mateo C.
The microwave syndrome: a preliminary study in Spain. Elec-
tromagn Biol Med 2003;22(2):161.
Oftedal G, Wilen J, Sandstrom M, Mild KH. Symptoms experienced
in connection with mobile phone use. Occup Med Lond
2000;50(4):237.
Ozturan O, Erdem T, Miman MC, Kalcioglu MT, Oncel S. Effects
of the electromagnetic field of mobile telephones on hearing.
Acta Otolaryngol 2002;122(3):289.
Raczek J, Runow K, Oetzel H, Gailus T, Herget I. Investigations of
electrosensitivity to a GSM signal at 900 MHz for a self-reported
electrosensitive target group. In: BEMS I, editor. Abstract book.
Mu
¨nchen7Bioelektromagnetische Society; 2000. p. 269.
Ro¨o¨ sli M, Moser M, Baldinini Y, Meier M, Braun-Fahrla¨ nder C.
Symptoms of ill health ascribed to electromagnetic field expo-
sure—a questionnaire survey. Int J Hyg Environ Health
2004;207(2):141.
Sage C. U
¨bersicht u¨ber Studien zur Wirkung hochfrequenter Felder
(mit Relevanz fu¨ r die Mobilkommunikation und Daten). Umw
Med Ges 2001;14(1):23 – 34.
Sandstro¨m M, Wilen J, Oftedal G, Mild KH. Mobile phone use and
subjective symptoms Comparison of symptoms experienced by
users of analogue and digital mobile phones. Occup Med (Oxf)
2001;51(1):25.
Santini R, Seigne M, Bonhomme-Faivre L, Bouffet S, Defrasne E,
Sage M. Symptoms experienced by users of digital cellular
phones. Pathol Biol 2001;49(3):222.
H. Seitz et al. / Science of the Total Environment 349 (2005) 45–5554
Santini R, Santini P, Danze JM, Le Ruz P, Seigne M. Investigation
on the health of people living near mobile telephone relay
stations: I Incidence according to distance and sex. Pathol
Biol 2002;50(6):369.
Santini R, Santini P, Danze JM, Le Ruz P, Seigne M. Survey study
of people living in the vicinity of cellular phone base stations.
Electromagn Biol Med 2003a;22(1):41.
Santini R, Santini P, Danze JM, LeRuz P, Seigne M. Symptoms
experienced by people in vicinity of base stations: II. Incidences
of age, duration of exposure, location of subjects in relation to
the antennas and other electromagnetic factors. Pathol Biol
2003b;51(7):412.
Silny J, Meyer M, Wiesmu¨ller GA, Dott W. Gesundheitsrelevante
Wirkungen elektromagnetischer Felder des Mobilfunks und
anderer neuer Kommunikationssysteme. Umwelt Med Forsch
Prax 2004;9(3):127 – 36.
SSK (Strahlenschutzkommission). Kriterien zur Bewertung strahle-
nepidemiologischer Studien Checkliste als Orientierungs-und
Bewertungshilfe. Band 2002;50.
Tahvanainen K, Nino J, Halonen P, Kuusela T, Laitinen T, Lansimies
E, et al. Cellular phone use does not acutely affect blood pressure
or heart rate of humans. Bioelectromagnetics 2004;2(2):73.
Wilen J, Sandstrom M, Hansson MK. Subjective symptoms among
mobile phone users—a consequence of absorption of radiofre-
quency fields? Bioelectromagnetics 2003;24(3):152.
Zwamborn APM, Vossen SHJA, van Leersum BJAM, Ouwens MA,
Ma¨kel WN. Effects of global communication system radio-fre-
quency fields on well-being and cognitive functions of human
subjects with and without subjective complaints. FEL-03-C148,
TNO Physics and Electronics Laboratory, 2003, Niederlande.
H. Seitz et al. / Science of the Total Environment 349 (2005) 45–55 55
