For a number of years, there has been concern about cancer risks among persons living near to power lines and other sources of electromagnetic fields. Epidemiological studies have been conducted in the UK and in other parts of the world on the risks of childhood cancer, including childhood leukaemia. A pooled analysis of several studies has been performed recently1 which shows that the possibility exists of a doubling of the risk of leukaemia in children in homes at high levels of exposure to extremely low frequency (50-60 Hz) magnetic fields. Such levels are rarely encountered in the UK.
The analysis was a collaborative effort, involving researchers who had conducted studies in nine countries in Europe, North America and New Zealand. In total, records were available for 3,247 children with leukaemia and 10,400 control children. For each child, either a 24 or 48-hour measurement of power frequency magnetic fields had been made in their home, or – for the Nordic countries – magnetic fields were calculated retrospectively using electrical utility records. The UK Childhood Cancer Study, which reported its findings on magnetic fields in 19992, contributed the largest number of leukaemia cases from any single study, namely 1,073.
The large numbers of children allowed more precise estimation of risks at the higher exposure levels. In addition, the format of the data analyses was decided in advance of starting work, and the availability of individual records enabled the same definitions of exposure to be used throughout.
For the 3,203 children with leukaemia and 10,338 control children with estimated residential magnetic field exposures less than 0.4 µT, the data from the pooled analysis were consistent with there being no increased risk. For the 44 children with leukaemia and 62 control children with magnetic field exposures of 0.4 µT or more, the leukaemia risk, which is small nationally (less than 1 in 20,000 per year), was estimated to be doubled: the relative risk was 2.00 (95% confidence interval 1.27-3.13). Adjustment for factors such as socio-economic status and type of dwelling did not change the results appreciably.
The estimated relative risk above 0.4 µT was similar for measurement-based studies combined and for calculated field studies combined. However, some of the measurement studies were affected by selection bias, because controls with a low socio-economic status were less likely be participants than cases with a low socio-economic status; there is some suggestion that this might account for part of the elevated risk estimate. In contrast, there was little potential for selection bias in the calculated field studies, because of the use of population registries in the Nordic countries, and because in-home measurements were not performed. However, a disadvantage was that the calculated field studies were not able to take account of individual characteristics of the homes.
This large and well-conducted analysis provides more details of childhood leukaemia and magnetic fields than was available from individual studies. For the overwhelming majority of children living in homes with magnetic field levels below 0.4 µT – estimated to be 99.6% of children in the UK Childhood Cancer Study – the data were consistent with no increased risk. For magnetic fields levels of 0.4 µT or more, the leukaemia risk was estimated to be doubled; this finding is unlikely to be due to random variability, but the authors suggest this may be accounted for in part by selection bias.
Professor N Day, a co-author of the paper, is a member of the Board of NRPB.
The NRPB independent Advisory Group on Non-ionising Radiation is presently completing work on a detailed review of experimental and epidemiological studies of possible effects of extremly low frequency electromagnetic fields on the risk of cancer. The report should be published early in 2001.
NRPB Information Sheet
21 November 2000
Last updated 21st November 2000