Environmental Research 187 (2020) 109621

Contents lists available at ScienceDirect

Environmental Research
journal homepage: www.elsevier.com/locate/envres

Case-control study on occupational exposure to extremely low-frequency T

electromagnetic fields and the association with acoustic neuroma
Michael Carlberga,∗, Tarmo Koppelb, Mikko Ahonenc, Lennart Hardella,1
a
The Environment and Cancer Research Foundation, Studievägen 35, SE 702 17, Örebro, Sweden
b
Department of Labour Environment and Safety, Tallinn University of Technology, SCO351 Ehitajate Tee 5, 19086, Tallinn, Estonia
c
Institute of Environmental Health and Safety, Jaama 14-3, 11615, Tallinn, Estonia

A R T I C LE I N FO A B S T R A C T

Keywords: Exposure to extremely low-frequency electromagnetic fields (ELF-EMF) was in 2002 classified as a possible
Electromagnetic fields human carcinogen, Group 2B, by the International Agency for Research on Cancer at WHO based on an increased
ELF-EMF risk for childhood leukemia. In case-control studies on brain and head tumours during 1997–2003 and
Acoustic neuroma 2007–2009 we assessed life-time occupations in addition to exposure to different agents. The INTEROCC ELF-
Occupational exposure
EMF Job-Exposure Matrix was used for associating occupations with ELF-EMF exposure (μT) with acoustic
Job-exposure matrix
neuroma. Cumulative exposure (μT-years), average exposure (μT) and maximum exposed job (μT) were calcu-
lated. No increased risk for acoustic neuroma was found in any category. For cumulative exposure in the highest
exposure category 8.52+ μT years odds ratio (OR) = 1.2, 95% confidence interval (CI) = 0.8–2.0, p linear
trend = 0.37 was calculated. No statistically significant risks were found in the time windows 1–14 years, and
15+ years, respectively. In conclusion occupational ELF-EMF was not associated with an increased risk for
acoustic neuroma.

1. Introduction (CI) = +2.1–3.9%.

Acoustic neuroma or Vestibular Schwannoma is a benign tumour in
the eighth cranial nerve that leads from the inner ear to the brain. It is a
slow growing tumour in the auditory canal and expands gradually out
into the cerebellopontine angle with potential compression of vital
brain stem centres. It tends to be encapsulated and grows in relation to
the auditory and vestibular portions of the nerve. This tumour type does
not undergo malignant transformation. Tinnitus and hearing problems
are usual first symptoms of acoustic neuroma. Although it is a benign
tumour, it causes persistent disabling symptoms after treatment such as
loss of hearing and tinnitus that severely affect the daily life.
Acoustic neuroma is a rare tumour. The average age-standardised
incidence rates ranged during 1987–2007 from 6.1 per 1,000,000 in
Finnish men to 11.6 in Danish men. Women in Sweden had the lowest
average rate of 6.4 per 1,000,000 and the highest rate, 11.6, was found
in Denmark (Larjavaara et al., 2011). The incidence increased statisti-
cally significantly during the time period 1987–2007 when all Nordic
countries (Denmark, Finland, Norway and Sweden) and both genders
were combined, +3.0% per year, 95% confidence interval
The aetiology of acoustic neuroma is not well known. Ionising ra-
diation exposure at childhood has been indicated as a risk factor
(Schneider et al., 2008). Neurofibromatosis 2, a rare disease, is one
established risk factor for acoustic neuroma with 90–95% lifetime risk
(Asthagiri et al., 2009).
In laboratory studies extremely low-frequency electromagnetic
fields (ELF-EMF) was mutagenic, but also potentiated the mutagenicity
of ionizing radiation using microsatellite analysis for DNA damage in
human glioma cells (Mairs et al., 2007). A possible involvement of O6-
methylguanine DNA adducts in the development of glioma was dis-
cussed by Ohgaki (2009).
Animal studies on ELF-EMF exposure alone have been inconclusive.
Long-term ELF-EMF exposure was a risk factor for chronic myeloid
leukemia in female mice (Qi et al., 2015). Rat studies showed that
exposure to ELF-EMF enhanced the carcinogenic effect of γ radiation
(Soffritti et al., 2016a) and that life-span exposure to ELF-EMF and
formaldehyde induced statistically significant carcinogenic effect
(Soffritti et al., 2016b). In a study on occupational exposure to ELF-EMF
based on gender-specific JEM, no increased risk was found for acoustic

∗
Corresponding author.
E-mail addresses: michael.carlberg@environmentandcancer.com (M. Carlberg), tarmo.koppel@ttu.ee (T. Koppel), mikko.ahonen@tutanota.com (M. Ahonen),
lennart.hardell@environmentandcancer.com (L. Hardell).
1
Department of Oncology, Faculty of Medicine and Health, Fakultetsgatan 1, Örebro University, SE-701 82, Örebro, Sweden

https://doi.org/10.1016/j.envres.2020.109621
Received 14 January 2020; Received in revised form 8 April 2020; Accepted 29 April 2020
Available online 07 May 2020
0013-9351/ © 2020 Elsevier Inc. All rights reserved.
M. Carlberg, et al.
neuroma (Forséen et al., 2006). No increased risk for acoustic neuroma
was found for occupational and residential exposure to EMFs, based on
only 32 cases, however (Baldi et al., 2011). Extremely low-frequency
(ELF)-EMF was in 2002 classified by the International Agency for Re-
search on Cancer (IARC) as “possibly carcinogenic to humans”, Group
2B based on an increased risk for childhood leukemia (IARC, 2002).
There was no consistent pattern of an association between brain tu-
mours and residential exposure to ELF-EMF. Data for occupational
cancer was limited. In summary the IARC panel concluded that the
evidence for an increased risk associated with ELF-EMF exposure for
other diseases than childhood leukemia was inadequate.
During the last couple of decades an increasing number of studies
have associated brain tumours with use of wireless phones (Hardell
et al., 2006, 2013a; Interphone Study Group, 2010; Coureau et al.,
2014; Hardell and Carlberg, 2015). During use they emit radio-
frequency electromagnetic fields (RF-EMF). In May 2011 IARC eval-
uated the carcinogenic potential from RF-EMF. The expert group clas-
sified RF-EMF in the frequency range 30 kHz–300 GHz as “possibly
carcinogenic to humans”, Group 2B (Baan et al., 2011; IARC, 2013).
Regarding use of mobile phones and the risk for acoustic neuroma a
meta-analysis of Interphone (Interphone Study Group, 2011) and
Hardell et al., (2013b) yielded statistically significant increased risk for
cumulative ipsilateral use ≥1640 h; odds ratio (OR) = 2.71 95%
confidence interval (CI) = 1.72–4.28 (Belpomme et al., 2018). Gender-
specific analyses in our study yielded similar results (Hardell et al.,
2013b). Cumulative use of wireless phones gave OR = 2.9, 95%
CI = 1.5–5.6 for men in the fourth quartile and OR = 1.9, 95%
CI = 1.1–3.4 for women; thus the results for both genders were sta-
tistically significant with 95% CI overlapping ORs (data not shown in
article).
Similarly, as in Interphone, our results on use of wireless phones and
brain tumour risk were based on case-control studies. We used a
structured questionnaire but with certain differences compared to the
Interphone study, such as that we used postal questionnaires sent to
cases and controls, supplemented over the phone instead of personal
interviews. In contrast to our studies even bedside interviews of cases
were performed in Interphone. Furthermore, we assessed in addition to
mobile phones also use of cordless phones (DECT); the latter use not
assessed by Interphone. Detailed comparison of the studies may be
found elsewhere (Hardell et al., 2008).
Wherever alternating electric current is used, this generates ELF-
EMFs. Most often exposure to ELF-EMFs is from machines operating on
the mains power, which operates at 50 Hz frequency. 50 Hz mains
power is used in most of the world, except for Americas and some parts
of Asia where 60 Hz used. Next to the frequency, the main exposure
feature is the amplitude of the wave. Magnetic fields depend on the
current – the more electrical power consumed the higher the magnetic
field. Therefore, powerful electrical machinery provides high ELF-EMF
exposure to workers working nearby these machines. High exposure to
the ELF magnetic field may also take place next to generation and
distribution of electrical power, such as power lines, transformers etc.
Electrical motors, which are used in many industrial processes, are
another source of high ELF magnetic field exposure.
Electrical motors and electromagnets, which incorporate coils pro-
duce often higher magnetic fields than the cables powering these. The
number of turns of electrically conductive element i.e. wire by which
the electrical current flows, also determines the resultant magnetic field
– the more there are turns, the higher the magnetic field. Therefore,
where workers use or are next to powerful electrical motors they are
exposed to strong ELF-EMFs. Where strong magnetic fields occur, like in
electric transport, industrial settings etc. the workers' exposure level
depends on the distance to the source. People working with hand-held
electrical power tools have strong magnetic fields exposure, whereas
others further away from even more strong sources, may be exposed at
much less levels. Therefore, a crucial role in determining workers' ex-
posure to the magnetic field is from the design of the instrumentation
2
Environmental Research 187 (2020) 109621
and the working areas with respect to the workers’ position.
Our case-control studies had detailed occupational history including
job titles, branch of different occupations and years for the specific jobs.
Thus, it was possible to calculate ELFEMF job exposure for cases and
controls using a job-exposure matrix (JEM).
2. Materials and methods
2.1. Data material
Similar methods were used in all our studies; detailed information
has been published previously (Hardell et al., 2006, 2013a). In brief, 6
administrative regions with oncology centres covering Sweden regis-
tered new cancer cases. Regarding 1997–2003, cases and controls
covered central Sweden (Hardell et al., 2006), whereas the 2007–2009
study included the whole country (Hardell et al., 2013a). The oncology
centres reported new cases with histopathologically verified brain tu-
mour, either benign or malignant, to us during these periods, although
the actual reporting interval varied from centre to centre. In total, 4038
brain tumour cases (2437 cases 1997–2003 and 1601 cases 2007–2009)
were included. Of these, 3563 (2158 cases 1997–2003 and 1405 cases
2007–2009) answered the questionnaire. Among the responding cases,
1498 were diagnosed with a malignant brain tumour and 2068 with a
benign brain tumour (3 were diagnosed with both malignant and be-
nign tumour). Regarding acoustic neuroma, 338 were included from the
brain tumour cases (257 from 1997 to 2003 and 81 from 2007 to 2009)
and of these, 316 answered the questionnaire (243 from 1997 to 2003
and 73 from 2007 to 2009).
Men and women, aged 20–80 years (1997–2003) and 18–75 years
(2007–2009) at the time of diagnosis, were included. Only living cases
were included after asking the responsible physician for permission
before inclusion in the study. Tumour localization in the brain was
based on reports to the cancer registries and medical records, which
were obtained after informed consent from the patients.
Controls were ascertained from the Swedish Population Registry.
The registry is continuously updated, so that each person could be
traced by a unique ID number. It also records the address of each
person. For each case, one control subject of the same gender and in the
same 5-year group was drawn at random from the Population Registry.
They were assigned the same year for cut-off of all exposure as the year
of diagnosis of the respective case. In total, 4038 controls (2437 con-
trols 1997–2003 and 1601 controls 2007–2009) were included; of
these, 3530 controls (2162 controls 1997–2003 and 1368 controls
2007–2009) answered the questionnaire. All these controls were used
in the analysis of risk of acoustic neuroma.
Exposure was assessed using a mailed questionnaire sent to each
person. The questionnaire contained a number of questions relating to
the overall working history, exposure to different chemicals and other
agents, smoking habits, X-ray investigations of the head and neck, and
heredity traits for cancer. Regarding use of a mobile phone and cordless
phone, time period, average daily use (min per day), use of hands-free
device, and external antenna in a car were asked for. The ear mostly
used during phone calls, or equally both, was also noted.
When questionnaire answers were unclear, they were resolved by
phone by trained interviewers using a written protocol for clarification
of each question. The interviewer supplemented the whole ques-
tionnaire during the phone call. Each questionnaire had received a
unique ID-number that did not disclose whether it was a case or a
control; i.e. the interviewer was unaware of the status during further
data processing. All information was coded and entered into a database;
case or control status was not disclosed until statistical analyses were
undertaken.
In this study we included acoustic neuroma cases. All controls were
used as comparison group. This was possible since we adjusted for
potential confounding factors such as year of diagnosis, age at diag-
nosis, gender and socioeconomic index (SEI). SEI was divided into three
M. Carlberg, et al. Environmental Research 187 (2020) 109621

categories: blue-collar worker (codes 11–22), white-collar worker Table 1

(codes 33–57) and self-employed (codes 60–89). Main characteristics of the study population. Distribution of acoustic neuroma
The questions regarding occupations included job title, branch and cases (n = 310) and controls (n = 3485) for age, gender, time period and SEI-
first and last year for each job in the work history of each participant. code.
The INTEROCC ELF Magnetic Field Job-Exposure Matrix (ELFJEM) was Variable Cases (n = 310) Controls (n = 3485)
used for associating occupations with ELF exposure (μT) (Turner et al.,
2014). The JEM used International Standard Classification of Occupa- No. % No. %

tions 1988 (ISCO88) four-digit codes for most jobs included; ISCO68 Mean age (median; min-max) 52 (53; 24–80) 54 (56; 20–80)
five-digit codes were used for more specific electrical jobs. The online Gender
version of the JEM is available at: Men 141 45.5 1472 42.2
http://radiation.isglobal.org/index.php/ca/radiation-programme- Women 169 54.5 2013 57.8
Time period
projects/2-general/55-database-elfjem.
1997–2003 237 76.5 2132 61.2
Job codes were coded using the Nordisk Yrkesklassificering (NYK 2007–2009 73 23.5 1353 38.8
85; five-digit codes) system and their validity was checked before they SEI-code
were translated to the International Standard Classification of Blue-collar worker 142 45.8 1568 45.0
White-collar worker 148 47.7 1695 48.6
Occupations 1988 (ISCO88; four-digit codes) using a coding key pro-
Self-employed 20 6.5 222 6.4
vided by Dr Erik Bihagen at Stockholm University (Bihagen, 2007). For
translation to the 1968 ISCO version for specific jobs (ISCO68; five-digit
codes) we compared with the NYK 85 system manually and selected the Table 2
most proper codes to be translated. Job exposure the year before di- Odds ratio (OR) and 95% confidence interval (CI) for acoustic neuroma cases
agnosis was excluded. No job was reported for 6 cases and 45 controls, (n = 310) for occupational exposure to ELF-EMF. Population based controls
so they were excluded from the analyses. (n = 3485) were used. Subjects with no coded occupation were excluded, 6
acoustic neuroma cases, and 45 controls. Unconditional logistic regression
2.2. Statistical methods analysis was adjusted for age at diagnosis, gender, socio-economic index (SEI)
and year of diagnosis. Exposure the year before diagnosis was excluded (“1-year
StataSE 12.1 (Stata/SE 12.1 for Windows; StataCorp., College lag”).
Station TX) was used for the analyses. Odds ratios (OR) and 95% con- Exposure metric Acoustic neuroma (n = 310)
fidence intervals (CI) were calculated using unconditional logistic re-
Ca/Co OR 95% CI
gression including acoustic neuroma cases and the whole control
sample (i.e. matched to both malignant and benign cases) to increase Cumulative exposure (μT-years)
the power.
Cumulative exposure (μT-years), average exposure (μT) and max- < 2.33 90/870 1.0 –
2.33- < 3.79 81/872 1.0 0.7–1.4
imum exposed job (μT) were calculated for the cases and controls for
3.79- < 5.55 59/869 0.8 0.5–1.2
lifetime work history and in time windows. Two sets of time windows 5.55- < 8.52 42/525 0.9 0.6–1.4
were analyzed, 1–14 and 15+ years before diagnosis. Cut points at the 8.52+ 38/349 1.2 0.8–2.0
25th, 50th, 75th and 90th percentile for controls were used to cate- p, linear trend 0.37
gorize the exposure variables with the lowest category (< 25th per- Average exposure (μT)
< 0.11 80/830 1.0 –
centile) as reference group (OR = 1.0). Tests for linear trends were
0.11- < 0.13 73/912 0.8 0.6–1.2
performed using the Wald χ2 test with the median of each category 0.13- < 0.18 64/871 0.7 0.5–1.04
included as an ordinal variable in the analyses. 0.18- < 0.27 57/523 1.0 0.7–1.5
In all analyses adjustment was made for the matching variables 0.27+ 36/349 1.0 0.6–1.5
p, linear trend 0.64
gender, age (as a continuous variable) and year of diagnosis and also for
Maximum exposed job (μT)
SEI divided into three categories (blue-collar worker, white-collar < 0.13 82/823 1.0 –
worker, self-employed). 0.13- < 0.16 70/812 0.8 0.6–1.2
Restricted cubic splines were used to illustrate the relationship be- 0.16- < 0.24 73/968 0.7 0.5–0.96
tween cumulative exposure to ELF-EMF (μT-years) in time windows and 0.24- < 0.60 45/532 0.7 0.5–1.1
0.60+ 40/350 1.0 0.6–1.5
acoustic neuroma. As suggested by Harrell Jr (Harrell, 2001), four knots
p, linear trend 0.54
were used at the 5th, 35th, 65th, and 95th percentiles.
Cut points at 25th, 50th, 75th and 90th percentile for controls.
3. Results
we found no statistically significant linear trend for increasing ex-
The results were based on analyses of 310 acoustic neuroma cases posure. Cumulative exposure in the highest exposure category, 8.52+
(92% of all included; 6.5% did not answer the questionnaire and 1.7% μT-years, gave OR = 1.2, 95% CI = 0.8–2.0, p, linear trend = 0.37.
were excluded since they had no job codes registered) and 3485 con- Cumulative exposure in different time windows before diagnosis is
trols (86% of included; 12.6% did not answer the questionnaire and shown in Table 3 and Figs. 1 and 2. No statistically significant risks or
1.1% were excluded since they had no job codes registered). linear trends were found. Cumulative exposure in the highest exposure
The mean age of the acoustic neuroma cases was 52 years (median group 2.75+ μT-years yielded OR = 1.3, 95% CI = 0.8–2.0 (p, linear
53, range 24–80) and of the controls 54 years (median 56, range trend = 0.26) in the latency group 1–14 years, see Table 2. For longer
20–80). Of the cases there were 141 men and 169 women, versus 1472 latency time, 15+ years, OR = 1.1, 95% CI = 0.6–1.8 was calculated
men and 2013 women among the controls. The mean number of jobs for in the highest exposure group 6.59+ μT-years (p, linear trend = 0.74),
cases were 2.9 (median = 3, min = 1, max = 10) and for controls 2.7 see Table 3.
(median = 2, min = 1, max = 12). Main characteristics of the study In a separate analysis we grouped latency in 1–4, 5–9 and 10+
population are displayed in Table 1. years. In the highest exposure category 0.69+ μT-years we calculated
Table 2 displays cumulative exposure in μT-years, average exposure OR = 1.2, 95% CI = 0.7–1.8 for tumour induction period 1–4 years,
in μT, and maximum exposure job (μT). No statistically significant in- exposure 0.92+ μT-years in the latency group 5–9 years yielded
creased or decreased risk was found for any of the studied variables and

3
M. Carlberg, et al. Environmental Research 187 (2020) 109621

Table 3
Odds ratio (OR) and 95% confidence interval (CI) for acoustic neuroma
(n = 310) for occupational exposure to ELF-EMF in time windows; 1–14 and
15+ years before diagnosis. Unconditional logistic regression, adjusted for age
at diagnosis, gender, socioeconomic index (SEI) and year of diagnosis was used.
Exposure the year before diagnosis was excluded (“1-year lag”).
Cumulative exposure (μT-years) Acoustic neuroma (n = 310)

Ca/Co OR 95% CI

1–14 years

< 0.91 63/770 1.0 –

0.91- < 1.42 78/872 1.1 0.8–1.6
1.42- < 1.82 71/778 1.2 0.8–1.7
1.82- < 2.75 51/537 1.1 0.8–1.7
2.75+ 38/329 1.3 0.8–2.0
p, linear trend 0.26
15 + years
< 1.44 94/782 1.0 –
1.44- < 2.55 70/777 0.9 0.6–1.3 Fig. 2. Restricted cubic spline plot of the relationship between cumulative ex-
2.55- < 4.17 54/787 0.7 0.5–1.1 posure to ELF-EMF in μT-years and acoustic neuroma in the 15+ years' latency
4.17- < 6.59 29/471 0.7 0.4–1.1 group. The solid line shows the odds ratio (OR) estimate and the broken lines
6.59+ 32/313 1.1 0.6–1.8 represent the 95% confidence interval (CI). Adjustment for age at diagnosis,
p, linear trend 0.74
gender, SEI-code, and year of diagnosis was made.
Cut points at 25th, 50th, 75th and 90th percentile for controls in each time
window. CI = 0.7–1.2, and p linear trend = 0.45 were calculated. No statisti-
cally significant risks were found in different time windows. Thus, oc-
cupational ELF-EMF was not associated with an increased risk for me-
ningioma.
The current study was based on the same set as above of case-
control studies during the time periods 1997–2003 and 2007–2009 on
brain and head tumours. This time acoustic neuroma was analyzed in
relation to occupational exposure to ELF-EMF. The whole group of
controls was used also this time.
In total 316 cases (93%) and 3530 controls (87%) participated in
the whole study (Hardell et al., 2013b). The analyses in this study were
based on 310 cases and 3485 controls that had reported an occupa-
tional history. Thus, the response rates were high and it is unlikely that
the impact of non-responders would not be so substantial that it would
affect the result in a statistically significant way.
The validity of the present results is further strengthened since the
results for occupational ELF-EMF exposure differed for glioma and
meningioma (Carlberg et al., 2017, 2018). It is unlikely that the results
would differ according to tumour type in the same study due to bias.
Fig. 1. Restricted cubic spline plot of the relationship between cumulative ex- The mean age for cases (52 years) and for controls (54 years) was
posure to ELF-EMF in μT-years and acoustic neuroma in the 1–14-year latency similar as well as range of ages. The percentage of female controls
group. The solid line shows the odds ratio (OR) estimate and the broken lines (58%) was somewhat higher than of female cases (55%) since in the
represent the 95% confidence interval (CI). Adjustment for age at diagnosis, whole case-controls study controls were matched on gender and me-
gender, SEI-code, and year of diagnosis was made. ningioma is more common among women than among men (Cea-
Soriano et al., 2012). The mean number of occupations was similar for
OR = 1.3, 95% CI = 0.8–1.9, and exposure 7.28+ μT-years gave acoustic neuroma cases (n = 2.9) and for controls (n = 2.7).
OR = 1.1, 95% CI = 0.6–1.8 for latency 10+ years. There was no No statistically significant increased risk for acoustic neuroma was
statistically significant trend (data not in table). found for cumulative exposure in μT-years in any exposure category or
in different time windows. To our knowledge no other studies exist on
occupational exposure to ELF-EMF and the risk for acoustic neuroma.
4. Discussion No such analyses have been published from the Interphone study
groups (Interphone Study Group, 2010, 2011; Turner et al., 2014).
We have previously reported the results on occupational exposure to Recently, the US National Toxicology Program (NTP)'s review con-
ELF-EMF and the risk for glioma (Carlberg et al., 2017). Cumulative firmed increased incidence of malignant schwannoma in the heart of
exposure gave for astrocytoma grade IV (glioblastoma multiforme) in radiofrequency radiation exposed test animals (National Toxicology
the time window 1–14 years OR = 1.9, 95% CI = 1.4–2.6, p linear Program, 2018a, 2018b; Hardell and Carlberg, 2019). Similar finding
trend < 0.001, and in the time window 15+ years OR = 0.9, 95% was observed in the Ramazzini Institute's rat study (Falcioni et al.,
CI = 0.6–1.3, p linear trend = 0.44 in the highest exposure categories 2018).
2.75+ and 6.59+ μT years, respectively.
These results related to glioma, are in contrast to our findings for
meningioma risk (Carlberg et al., 2018). No increased risk for me- 5. Conclusion
ningioma was found in any category. For cumulative exposure in the
highest exposure category 8.52+ μT years OR = 0.9, 95% In conclusion no association was found between occupational ELF-

4
M. Carlberg, et al.
EMF exposure and acoustic neuroma.
Funding
This study was supported by Kone Foundation, Helsinki, Finland,
Cancerhjälpen, Sweden, and Pandora-Foundation for Independent
Research, Berlin, Germany. Support by a grant from Mr Brian Stein is
acknowledged. Funding sources had no role in study design, data col-
lection, analysis and interpretation of the data, in the writing of the
manuscript, or in the decision to publish the results.
Ethics approval and consent to participate
All study subjects participated after informed consent; the studies
were approved by the Ethical Committee (Örebro County Hospital DNR
351/96, Uppsala University DNR, 2005:367).
CRediT authorship contribution statement
Michael Carlberg: Conceptualization, Formal analysis, Data cura-
tion. Tarmo Koppel: Validation, Writing - review & editing. Mikko
Ahonen: Validation, Writing - review & editing. Lennart Hardell:
Conceptualization, Supervision, Writing - original draft.
Declaration of competing interest
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influ-
ence the work reported in this paper.
References
Asthagiri, A.R., Parry, D.M., Butman, J.A., et al., 2009. Neurofibromatosis type 2. Lancet
373, 1974–1986.
Baan, R., Grosse, Y., Lauby-Secretan, B., et al., 2011. Carcinogenicity of radiofrequency
electromagnetic fields. Lancet Oncol. 12, 624–626.
Baldi, I., Coureau, G., Jaffe, A., et al., 2011. Occupational and residential exposure to
electromagnetic fields and risk of brain tumors in adults: a case-control study in
Gironde, France. Int. J. Canc. 129, 1477–1484.
Belpomme, D., Hardell, L., Belyaev, I., et al., 2018. Thermal and non-thermal health ef-
fects of low intensity non-ionizing radiation: an international perspective. Environ.
Pollut. 242, 643–658.
Bihagen, E., 2007. Nya möjligheter för stratifieringsforskning i Sverige: internationella
yrkesklassificeringar och stratifieringsmått över tid [In Swedish]. Sociol. Forsk. 44,
52–67.
Carlberg, M., Koppel, T., Ahonen, M., Hardell, L., 2017. Case-control study on occupa-
tional exposure to extremely low-frequency electromagnetic fields and glioma risk.
Am. J. Ind. Med. 60, 494–503.
Carlberg, M., Koppel, T., Ahonen, M., Hardell, L., 2018. Case-control study on occupa-
tional exposure to extremely low-frequency electromagnetic fields and the associa-
tion with meningioma. BioMed Res. Int. 5912394 2018.
Cea-Soriano, L., Wallander, M.A., García Rodríguez, L.A., 2012. Epidemiology of me-
ningioma in the United Kingdom. Neuroepidemiology 39, 27–34.
Coureau, G., Bouvier, G., Lebailly, P., et al., 2014. Mobile phone use and brain tumours in
the CERENAT case-control study. Occup. Environ. Med. 71, 514–522.
Falcioni, L., Bua, L., Tibaldi, E., et al., 2018. Report of final results regarding brain and
heart tumors in Sprague-Dawley rats exposed from prenatal life until natural death to
mobile phone radiofrequency field representative of a 1.8 GHz GSM base station
environmental emission. Environ. Res. 165, 496–503.
Environmental Research 187 (2020) 109621
Forséen, U.M., Lönn, S., Ahlbom, A., et al., 2006. Occupational magnetic field exposure
and the risk of acoustic neuroma. Am. J. Ind. Med. 49, 112–118.
Hardell, L., Carlberg, M., 2015. Mobile phone and cordless phone use and the risk for
glioma - analysis of pooled case-control studies in Sweden, 1997-2003 and 2007-
2009. Pathophysiology 22, 1–13.
Hardell, L., Carlberg, M., 2019. Comments on the US National Toxicology Program
technical reports on toxicology and carcinogenesis study in rats exposed to whole-
body radiofrequency radiation at 900 MHz and in mice exposed to whole-body
radiofrequency radiation at 1,900 MHz. Int. J. Oncol. 54, 111–127.
Hardell, L., Carlberg, M., Söderqvist, F., Hansson Mild, K., 2013a. Case-control study of
the association between malignant brain tumours diagnosed between 2007 and 2009
and mobile and cordless phone use. Int. J. Oncol. 43, 1833–1845.
Hardell, L., Carlberg, M., Hansson Mild, K., 2006. Pooled analysis of two case-control
studies on use of cellular and cordless telephones and the risk for malignant brain
tumours diagnosed in 1997–2003. Int. Arch. Occup. Environ. Health 79, 630–639.
Hardell, L., Carlberg, M., Hansson Mild, K., 2008. Methodological aspects of epidemio-
logical studies on the use of mobile phones and their association with brain tumors.
Open Environ. Sci. 2, 54–61.
Hardell, L., Carlberg, M., Söderqvist, F., Hansson Mild, K., 2013b. Pooled analysis of case-
control studies on acoustic neuroma diagnosed 1997-2003 and 2007-2009 and use of
mobile and cordless phones. Int. J. Oncol. 43, 1036–1044.
Harrell Jr., F.E., 2001. Regression Modeling Strategies. With Application to Linear
Models, Logistic Regression and Survival Analysis. Springer, New York.
IARC, 2002. Monograph - Non-ionizing Radiation, Part 1: Static and Extremely Low-
Frequency (ELF) Electric and Magnetic Fields, vol. 80 World Health Organization,
International Agency for Research on Cancer, Lyon. http://monographs.iarc.fr/ENG/
Monographs/vol80/mono80.pdf.
IARC, 2013. Monograph - Non-ionizing Radiation, Part 2: Radiofrequency
Electromagnetic Fields, vol. 102 World Health Organization, International Agency for
Research on Cancer, Lyon. http://monographs.iarc.fr/ENG/Monographs/vol102/
mono102.pdf.
Interphone Study Group, 2010. Brain tumour risk in relation to mobile telephone use:
results of the INTERPHONE international case-control study. Int. J. Epidemiol. 39,
675–694.
Interphone Study Group, 2011. Acoustic neuroma risk in relation to mobile telephone use:
results of the INTERPHONE international case-control study. Canc. Epidemiol. 35,
453–464.
Larjavaara, S., Feychting, M., Sankila, R., et al., 2011. Incidence trends of vestibular
schwannomas in Denmark, Finland, Norway and Sweden in 1987-2007. Br. J. Canc.
105, 1069–1075.
Mairs, R.J., Hughes, K., Fitzsimmons, S., et al., 2007. Microsatellite analysis for de-
termination of the mutagenicity of extremely low-frequency electromagnetic fields
and ionising radiation in vitro. Mutat. Res. 626, 34–41.
National Toxicology Program, 2018. NTP Technical Report on the Toxicology and
Carcinogenesis Studies in Hsd:Sprague Dawley Sd Rats Exposed to Whole-Body Radio
Frequency Radiation at a Frequency (900 MHz) and Modulations (GSM and CDMA)
Used by Cell Phones, vol. 595 NTP TR March 26-28, 2018. https://ntp.niehs.nih.
gov/ntp/about_ntp/trpanel/2018/march/tr595peerdraft.pdf.
National Toxicology Program, 2018. NTP Technical Report on the Toxicology and
Carcinogenesis Studies in B6C3F1/N Mice Exposed to Whole-Body Radio Frequency
Radiation at a Frequency (1,900 MHz) and Modulations (GSM and CDMA) Used by
Cell Phones, vol. 596 NTP TR March 26-28, 2018. https://ntp.niehs.nih.gov/ntp/
about_ntp/trpanel/2018/march/tr596peerdraft.pdf.
Ohgaki, H., 2009. Epidemiology of brain tumors. Methods Mol. Biol. 472, 323–342.
Qi, G., Zuo, X., Zhou, L., et al., 2015. Effects of extremely low-frequency electromagnetic
fields (ELF-EMF) exposure on B6C3F1 mice. Environ. Health Prev. Med. 20, 287–293.
Schneider, A.B., Ron, E., Lubin, J., et al., 2008. Acoustic neuromas following childhood
radiation treatment for benign conditions of the head and neck. Neuro Oncol. 10,
73–78.
Soffritti, M., Tibaldi, E., Padovani, M., et al., 2016a. Life-span exposure to sinusoidal-50
Hz magnetic field and acute low-dose γ radiation induce carcinogenic effects in
Sprague–Dawley rats. Int. J. Radiat. Biol. 92, 202–214.
Soffritti, M., Tibaldi, E., Padovani, M., et al., 2016b. Synergism between sinusoidal-50 Hz
magnetic field and formaldehyde in triggering carcinogenic effects in male Sprague-
Dawley rats. Am. J. Ind. Med. 59, 509–521.
Turner, M.C., Benke, G., Bowman, J.D., et al., 2014. Occupational exposure to extremely
low-frequency magnetic fields and brain tumor risks in the INTEROCC study. Canc.
Epidemiol. Biomarkers Prev. 23, 1863–1872.