Intended for healthcare professionals


Childhood cancer in relation to distance from high voltage power lines in England and Wales: a case-control study

BMJ 2005; 330 doi: (Published 02 June 2005) Cite this as: BMJ 2005;330:1290

Childhood cancer in relation to distance from high voltage power lines: Some important considerations

The paper from Draper and colleagues [1] on the relation between
childhood cancer and the distance of birth residence close to high voltage
power lines presents some notable findings. The study has distinct
advantages of size, in terms of the number of case children, and the
unbiased selection of the control sample. However, the findings are
inconsistent with another large UK study where estimates of dose to
extremely low frequency magnetic fields from power lines were used [2].
If this investigation was established primarily to examine risk in
relation to exposure to magnetic fields it is not clear why the categories
used as a measure of exposure were extended beyond 200m distance to high
voltage power lines, a point at which their contribution to exposure can
be considered equivalent to 'background' levels compared to the
contribution from other sources [3]. The strength of the findings are
based on trend statistics with the reference group comprising birth
residences over 600m distance from power lines, an analysis that can have
no basis for inferring associations with extremely low frequency magnetic
fields. No plausible biological evidence yet exists linking magnetic
field exposure to cancer per se or to childhood leukaemia. Despite this,
the paper quantifies the possible number of cases of childhood leukaemia
'associated' with high voltage lines for which the main exposure will be
to magnetic fields.

The statistically significant associations revealed in this
geographical analysis lack any adjustment for population characteristics
other than social class, estimated by the Carstairs index (it is unclear
how this measure was calculated for the period prior to the 1981 census).
Crucially, it is known that the areal distribution of childhood leukaemia
varies with other factors, also measurable using census data, such as
population density and population mixing [4], neither of which have been
adjusted for in the analysis as potential confounders for the excess risk.
The authors indicate that the mobility of cases did not differ with
respect to power line proximity as assessed by different postcodes
recorded between birth and diagnosis. Apart from mobility of the
individual, characteristics of the area in which they live may also
influence the risk of disease and these need to be considered in the

It is of interest that the level of risk was diluted by the use of
all controls as the comparison group, although the authors fail to clarify
this. Matched analyses may be preferred in a matched study design but
findings can be considered to be less robust if the estimates are
noticeably different when the matching is broken. All controls,
irrespective of the case diagnosis, were selected to represent the entire
population and the reasons why differences were observed merit more
detailed investigation.

The findings of this study are of interest in that they point towards
geographical correlates of risk for childhood leukaemia but do not support
the hypothesis that electromagnetic fields have a causal role.

1. Draper G, Vincent T, Kroll ME, Swanson J. Childhood cancer in
to distance from high voltage power lines in England and Wales: a case-
study. BMJ 2005;330:1290-2.

2. UKCCS Study Investigators (Writing committee: J Skinner, MP
Maslanyj, TJ Mee, SG Allen, J Simpson, E Roman, NE Day). Childhood cancer
and residential proximity to power lines. Br J Cancer 2000; 83(11): 1573-

3. Vistnes AI, Ramberg GB, Bjørnevik LR, Tynes T, Haldorsen T.
Exposure of children to residential magnetic fields in Norway: is
proximity to power lines an adequate predictor of exposure?
Bioelectromagnetics 1997; 18: 47-57.

4. Parslow RC, Law GR, Feltbower R, Kinsey SE, McKinney PA.
Population mixing, childhood leukaemia, CNS tumours and other childhood
cancers in Yorkshire. Eur J Cancer 2002; 38: 2033-2040

Competing interests:
None declared

Competing interests: No competing interests

03 June 2005
Sarah J. Hepworth
Medical statistician
Richard G. Feltbower, Roger C. Parslow, Patricia A. McKinney
Paediatric Epidemiology Group, University of Leeds, Leeds, LS2 9LN