  |
|
|
||||||||||||||
|
||||||||||||||||
| Journal of the New Zealand Medical Association,
06-June-2003, Vol 116 No 1175
|
||||||||||||||||
|
Cellular telephone use and time trends for brain, head and
neck tumours
Angus Cook, Alistair Woodward, Neil Pearce and Cara
Marshall
The widespread use of cellular telephones represents a
relatively new form of electromagnetic radiation exposure for the human
population. The growth in users in New Zealand has been exponential since
cellular telephones first appeared in 1987, and regular users now exceed one
third of the population.
It has been conjectured that exposure to energy emitted by
cellular telephones, particularly radio frequency radiation (RFR), may lead to
late-stage promotion of tumours, including brain tumours. The mode of action is
uncertain, but laboratory studies indicate that the energy produced by RFR may
produce vibration and polarisation of cellular molecules and may have a
catalytic action for some cellular
enzymes.1
A small number of case-control studies have suggested a
possible elevation in brain tumour risk amongst users, particularly for users of
early analogue phones.2 Eye tumours have also
come under investigation and Stang’s recent study suggests an association
between uveal melanomas and cellular telephone
use.3 However, these findings have not been
supported by other recent papers; incidence and cohort linkage studies in
Denmark4,5 and a case-control study of brain
cancers in the US6 did not find an increase in
tumour risk associated with the use of cellular telephones.
Cellular telephones generally operate at no more than 0.25
watts,7 but an appreciable proportion of the
energy is absorbed by local structures in the head and neck (up to 1.1 W/kg). To
measure the effect of using cellular telephones, human equivalent models or
‘phantoms’ are used to provide a measure of absorption in biological
tissue.
Parts of the face, particularly the cheek adjacent to the
earlobe, are relatively highly exposed to RFR. The sites most exposed on the
side of the head against which the telephone is placed include the adjacent skin
and muscle, the parotid gland, the acoustic nerve, and meninges and brain
tissues at the outermost surface of the temporal and parietal lobes. In
contrast, other head and neck sites, such as the eyes, cerebellum, thyroid and
midbrain, receive substantially lower levels of radiation because of the rapid
attenuation of the specific absorption rate over a distance of 5
cm.8
We investigated whether trends in tumour incidence rates in
New Zealand have varied since the introduction of cellular telephones in 1987.
Incidence rates for the years before the start of cellular telephone services
provide an indication of the ‘pre-cellular telephone’ baseline in
New Zealand. If there were an effect of cellular telephones on incidence rates,
one would expect to observe an increase in rates above this baseline after a
certain latency period. In addition, it was conjectured that an increase in
tumours related to cellular telephone use was most likely to occur in anatomical
sites with the highest absorption rates.
For the purposes of this analysis, exposure will be
described in terms of the acronym REACT, or Radiation and Extremely low
frequency electromagnetic fields from Activation of Cellular Telephones, to
distinguish electromagnetic energy associated with cellular telephones from that
generated by other sources, such as microwave ovens, base stations, radio and TV
transmitters, and other devices (whose energy outputs fall well below
international environmental standards for the majority of the
population)9. The term REACT is also used to
encompass the full range of relevant exposures occurring during cellular
telephone activation, and is inclusive of both RFR and lower frequency forms of
electromagnetic energy although their individual or synergistic roles in tumour
development remains highly
speculative.9,10
MethodsIncidence
data Incidence data were collected for males and females aged 20 to 69
years because this encompasses the age group that has used cellular telephones
for the longest duration. Data were obtained from the New Zealand Cancer
Registry for the period from 1986 to 1998 inclusive.
The tumours and their corresponding ICD-9 codes are summarised in Table 1. The proportion of brain tumours classified as occurring in an ‘unspecified’ site (code 191.9; ie, no lobe indicated) on average fell below 10% of total brain malignancies. The following tumours were excluded from the study:
These exclusion criteria were
implemented because the incidence rates for these benign, unclassified and
metastatic tumours, and for lymphomas and leukaemias, are either not recorded,
or are inconsistently recorded, by the New Zealand Cancer Registry.
In addition to these tumour groups, cranial nerve malignancies were excluded because of their rarity. Table 1. Malignancies classified by REACT exposure and
ICD-9 coding
Use of dosimetry measures We used dosimetry data indicating that areas of the head proximal to the cellular telephone received more radiation than other, distal, sites. Cranial regions were classified as having high, medium and low REACT exposure (Table 1). Prevalence of cellular telephone use in the New Zealand population The exposure measure used was the proportion of cellular telephone subscribers within the national population. This was calculated using the number of subscribers provided by the cellular telephone service providers in New Zealand over the study period: Telecom and BellSouth/Vodafone. It was not possible to stratify prevalence of use by age or gender; thus an overall estimate of prevalence is provided across all age groups and both males and females. However, past consumer and community survey data confirm that a high proportion of users into the mid-1990s were in the 20–59 age group, although patterns have changed more recently (particularly with growing use amongst children and adolescents). Analysis Age-standardized rates for 20 to 59 years of age were aggregated and analysed separately by gender. Trend and regression analyses were conducted. MINITAB (version 13) was used to generate the graphical displays. ResultsFigure 1. Incidence of brain malignancies (ICD-9: 191)
at all sites 1986–1998, males and females (click
here to view Figure 1)
The graphs for the tumours emerging in the areas of high
REACT exposure are illustrated in Figure 2. Graphs for males only are displayed;
the pattern for females was comparable across all tumour types and thus these
are not illustrated separately.
Figure 2. Incidence of tumours in areas of high REACT
exposure (click here to view Figure 2)
The graphs for medium and low
REACT exposure sites – including
malignancies arising in the frontal and occipital lobes of the brain, midline
cranial structures and the eye – did not display any significant changes
in trend patterns for either gender over the years 1986 to 1998. The incidence
rates for some of the malignancies analysed were extremely low or zero (such as
for pineal gland malignancies) in some or all age strata.
ConclusionsIncidence rates for malignancies
arising in the head and neck, including those sites that hypothetically receive
the highest levels of radio frequency radiation during cellular telephone use,
have not changed materially since the introduction of cellular telephones to New
Zealand.
The latency period of the effect of cellular telephones,
assuming that such an effect actually exists, is unknown. In this study, it is
possible that the interval of 12 years after service commencement may be too
short to observe an effect on incidence rates. However, the laboratory evidence
suggests that if RFR does have an effect on carcinogenesis, it is as a
late-stage promoter. If promotion is indeed the mechanism of tumourigenic action
from cellular telephone use, then one would expect the period between exposure
and tumour presentation to be relatively short. Indeed, for many of the
case-control and cohort studies either conducted or underway, the possibility of
a relatively short interval (eg, 5–10 years) between cellular telephone
use and tumour emergence has been implied or
conjectured.2,11,12 Although these studies
benefit from the use of individual estimates of exposure not possible in this
study, our ecological analysis does not support a dramatic increase in cancer
rates within this time frame.
Major interpretational difficulties in this study relate to
the ‘ecological fallacy’, which limits our ability to attribute
disease rates to exposure based on information at a population level. For this
study, it cannot be determined whether individuals diagnosed with malignancies
were in fact cellular telephone users (or the degree to which they were users),
because we have information on cellular telephone use only for the overall
population. In addition, some degree of misclassification of exposure is
probable given that the data used represented cellular telephone subscription,
rather than use per se. Individual
subscription or possession of a phone does not always imply use, and individuals
may in turn use other phones that they do not actually own. Therefore, there is
likely to be some disparity between usage estimates, provided by such sources as
telecommunication companies, and true use. It is expected, however, that changes
in recorded ownership or subscription over time will provide a general measure
of increasing use at a population level.
Use of incidence rate analysis for some tumour groups was
constrained by the very low incidence for some malignancies, such as cancers of
the pineal gland. Using a New Zealand population alone, a trend may be difficult
to determine because of the small base population in which such rare tumours
might emerge.
The use of registry data in the interpretation of
site-specific tumours may be limited for a number of reasons. An obvious problem
is erroneous coding of hospitalisation and registry data, in which the
anatomical locations of tumours may be misrecorded. With regard to specific risk
of overclassification of brain tumours in the ‘unspecified’ category
(ICD-9 code 191.9; ie, no lobe indicated), this occurred for under 10% of total
brain malignancies indicating that the subsite information was relatively
complete. In other cases, such as for malignancies of the meninges, it is not
possible to determine where exactly the tumour has arisen based on the existing
codes. Only some malignant meningiomas will arise in the area of high RFR
exposure from cellular telephones; based on dosimetry data, meninges of the
frontal, occipital and midline regions are unlikely to have received substantive
levels of RFR. This illustrates the difficulty in addressing the issue of tumour
laterality (ie, emergence on the left or right side of the brain or head) in
such a study. In general, cellular telephone users who favour placement of the
device against one side of the head receive very little radiation on the
contralateral side, and determining tumour emergence on this basis necessitates
more detailed information on individual usage patterns (eg,
interviews).
In conclusion, we found no evidence of an increase in brain
malignancies in New Zealand in the years following the introduction of cellular
telephones to New Zealand. This suggests that if there is an increase in tumour
rates with cellular telephone use it is relatively weak, or is manifest after a
longer latency period. However, ecological studies of this nature are limited in
many ways, and a stronger study design is clearly needed to establish more
exactly any elevation in risk. This is the motivation of the case-control and
historical cohort research underway
internationally,11 including the multicentre
IARC Interphone Study in which New Zealand is a
participant.12 At a national level, it is
critical to use New Zealand’s high-quality incidence data to continue
monitoring tumour rates in the future, taking into account longer latency
periods. Future research will also need to incorporate wider age groups as ever
increasing numbers of children and adolescents take up regular use of cellular
telephones.
Author information:
Angus Cook, Research Fellow; Alistair Woodward, Professor and Head of
Department; Department of Public Health, Wellington School of Medicine; Neil
Pearce, Professor and Director, Centre for Public Health Research, Massey
University, Wellington; Cara Marshall, Research Fellow, Department of Public
Health, Wellington School of Medicine, Wellington
Acknowledgements: We
acknowledge the assistance and advice of the New Zealand Cancer Registry (New
Zealand Health Information Service) and Dr Elisabeth Cardis, Chief of the Unit
of Radiation and Cancer, International Agency for Research on Cancer, France.
Funding for the study was provided by the Health Research Council.
Correspondence: Dr
Angus Cook, Department of Public Health, Wellington School of Medicine, P O Box
7343, Wellington. Fax: (04) 3895 319; email: acook@wnmeds.ac.nz
References:
|
|
||||||||||||||
| | Current
issue | Search journal |
Archived issues | Classifieds
| Hotline (free ads) Subscribe | Contribute | Advertise | Contact Us | Copyright | Other Journals |
|