Over 40 epidemiological studies have now been performed looking just at a possible link between childhood leukaemia and EMFs. Scientific papers have an abstract and you can read the abstracts of the more important epidemiological studies here. Some of those studies found no association with magnetic fields, but most have found associations. Overall there is clearly a statistical association within the studies that have been performed between unusually high magnetic fields and childhood leukaemia, although it's fair to say it may be getting slightly weaker in recent years.
By contrast, the evidence from the laboratory is that low level EMFs of the type experienced by the public do not seem to have the harmful effects that have been claimed, and we have not identified a mechanism:
Most review groups agree the evidence for EMFs causing diseases other than childhood leukaemia is weaker - WHO described it as "much weaker". That's why give a lot of detail here on childhood leukaemia, then summarise the evidence on all the other suggested health effects.
Summarising the epidemiological evidence
There have been over 40 epidemiological studies of childhood leukaemia. See the abstracts of the dozen or so most important studies on childhood leukaemia.
We have more details on some of the important studies:
- the UKCCS, a UK study that measured fields in the home
- the CCRG or "Draper" study, a UK study that looked at proximity to power lines and calculated fields from those power lines
- Geocap, a French study
- CAPS, a study from California
These individual studies of childhood leukaemia are best summed up in the various pooled analyses, notably the Ahlbom study from 2000 and the Kheifets et al follow-up in 2010. Pooled analyses (as the name suggests) pool the results from all the individual studies to calculate a single overall result. This graph is a summary of the results from the 2000 Ahlbom pooled analysis:
This suggests little or no evidence of risk in the intermediate exposure categories, but a doubling of risk - a relative risk of 2 - for fields in the home above 0.4 µT.
Since then, the suggested increase in risk may have got as bit smaller, but it's still there and still statistically significant.
Another stand of epidemiological evidence is what's called an "ecological" study - looking at changes over time in exposures, and changes over the same time period in incidence rates, and seeing if the two correlate, as you'd expect them to if one caused the other. This has been done for magnetic fields and childhood leukaemia, and shows a poor correlation (they both went up in the 20th century but at different times), but it's not a particularly strong type of evidence.
See more detail on pooled analyses - alternative interpretations of this result, and the update to it in 2010.
Two important UK studies
The Childhood Cancer Research Group Study
This study (also known as the “Draper” study), which published its first results in 2005, is a collaboration between the CCRG at the University of Oxford and National Grid. It looks at whether children who get cancer were born near power lines or not. It found:
“While few children in England and Wales live close to high voltage power lines at birth, there is a slight tendency for the birth addresses of children with leukaemia to be closer to these lines than those of matched controls.”
The strange thing about this result is that it seems to extend much further from power lines – up to 600 m – than magnetic fields do. The authors say:
“…our results do not seem to be compatible with the existing data on the relation between magnetic fields and risk.”
“We have no satisfactory explanation for our results in terms of causation by magnetic fields or association with other factors.”
Later findings from the study made the results even more puzzling, because they seemed to change over time.
The United Kingdom Childhood Cancer Study
The biggest of all the epidemiological studies of EMFs and childhood cancer until the Draper study was conducted during the 1990s, called the United Kingdom Childhood Cancer Study (UKCCS). It looked at a number of suggested causes of childhood cancer including EMFs. Its particularly large study population – over 2000 cases of cancer in total, three times larger than the next largest study before it – makes it very powerful. It was conducted by an eminent group of scientists, led by Sir Richard Doll, who first identified the link between smoking and lung cancer.
The UKCCS was designed to look at every case of childhood leukaemia occurring in the UK over roughly a four-year period. It is hard to envisage a study better able to give comprehensive answers about childhood cancer in the UK, and if a link with EMFs exists, this study was expected to find it.
In December 1999, the UKCCS published its report on exposure to magnetic fields and concluded:“This study provides no evidence that exposure to magnetic fields associated with the electricity supply in the UK increases risk for childhood leukaemia, cancers of the nervous system, or any other childhood cancer.”
Nonetheless, even though the UKCCS on its own provides "no evidence" of a risk, the pooled analysis, which included the UKCCS, still found a statistical association.
more detail on the UKCCS including further publications on electric fields and proximity to power lines.
What could the other explanations for the evidence be?
Associations from epidemiological studies on their own do not establish causation (More on what epidemiology can or cannot show). The associations could be caused by two other things:
- Bias in the study. If the children chosen to take part in the study were not properly representative of the population as a whole, the association that was found in the study might not exist in the rest of the population. We know that this problem (often called “control selection bias”) exists in most of the studies; what we don’t yet know is exactly how significant it is. More on bias.
- Other factors which were not properly controlled for. Something which does actually cause childhood cancer might be associated with magnetic fields so that children are exposed to both at once. Then the association that the study found would be caused by the other factor and not by EMFs. This is known as a “confounding factor”. We list here some of the other factors which vary with magnetic fields.
Some specific alternative explanations that have been looked at:
- the corona ion hypothesis - that effects near high-voltage power lines are caused not by the magnetic field but by the corona ions emitted by those lines
- the contact current hypothesis - that effects might be caused by contact voltages in homes producing effects in bone marrow
Some of the evidence relates specifically to the magnetic fields from power lines. That opens up the possibility that the mere presence of the power line might be relevant - maybe power lines are routed through areas with certain characteristics, or maybe the presence of the power line affects the area, and maybe this in turn links to leukaemia rates. This is a bit speculative but is an active line of research.
What the various expert review groups say about childhood leukaemia
The view of PHE
In a major review of the evidence for a possible association between exposure to power-frequency electric and magnetic fields and the incidence of cancer published in March 2001, the NRPB Advisory Group on Non-Ionising Radiation concluded:
“Laboratory experiments have provided no good evidence that extremely low frequency electromagnetic fields are capable of producing cancer, nor do human epidemiological studies suggest that they cause cancer in general. There is, however, some epidemiological evidence that prolonged exposure to higher levels of power frequency magnetic fields is associated with a small risk of leukaemia in children. In practice, such levels of exposure are seldom encountered by the general public in the UK. In the absence of clear evidence of a carcinogenic effect in adults, or of a plausible explanation from experiments on animals or isolated cells, the epidemiological evidence is currently not strong enough to justify a firm conclusion that such fields cause leukaemia in children. Unless, however, further research indicates that the finding is due to chance or some currently unrecognised artefact, the possibility remains that intense and prolonged exposures to magnetic fields can increase the risk of leukaemia in children.”
In its new advice on exposure limits in 2004, the NRPB stated:
“In the view of NRPB, the epidemiological evidence that time-weighted average exposure to power frequency magnetic fields above 0.4 µT is associated with a small absolute raised risk of leukaemia in children is, at present, an observation for which there is no sound scientific explanation. There is no clear evidence of a carcinogenic effect of ELF EMFs in adults and no plausible biological explanation of the association that can be obtained from experiments with animals or from cellular and molecular studies. Alternative explanations for this epidemiological association are possible: for example, potential bias in the selection of control children with whom leukaemia cases were in some studies and chance variations resulting from small numbers of individuals affected. Thus any judgements developed on the assumption that the association is causal would be subject to a very high level of uncertainty.”
The view of IARC
The International Agency for Research on Cancer (IARC) is an agency of the World Health Organisation. Its Unit of Carcinogen Identification and Evaluation has, since 1972, periodically published Monographs which assess the evidence that various agents are carcinogenic and classify the agents accordingly. In June 2001, a Working Group met to consider static and extremely-low-frequency electric and magnetic fields. The complete results have been published as Monograph number 80. Power-frequency magnetic fields were classified as “possibly carcinogenic”, on the basis of “limited” evidence from humans concerning childhood leukaemia, “inadequate” evidence from humans concerning all other cancer types, and “inadequate” evidence from animals. Power-frequency electric fields were judged “not classifiable” on the basis of “inadequate” evidence from both humans and animals.
The view of ICNIRP
A major review on epidemiology published by ICNIRP in 2001 concluded:
“Following the original report by Wertheimer and Leeper linking the three most common forms of childhood cancer with a proxy measure of residential EMF (wire codes), more than 18 studies in nine countries have shown no convincing evidence of a relationship between childhood leukaemia and residential EMF exposures among children with estimated exposure levels under 0.2 µT. A 2-fold increase in relative risk of childhood leukaemia, confined to a very tiny fraction of children (estimated as 0.8% in one large pooled analysis) with residential EMF exposures ≥0.4 µT, is difficult to interpret in the absence of a known biological mechanism or reproducible experimental support of carcinogenesis. There is also some evidence to suggest that selection bias may account for some of the increase in risk among the proportion of children with high residential EMF exposure.”
The view of WHO
The WHO Environmental Health Criteria Monograph published in 2007 concluded:
"The IARC classification was heavily influenced by the associations observed in epidemiological studies on childhood leukaemia. The classification of this evidence as limited does not change with the addition of two childhood leukaemia studies published after 2002. Since the publication of the IARC monograph the evidence for other childhood cancers remains inadequate.
New human, animal and in vitro studies, published since the 2002 IARC monograph, do not change the overall classification of ELF magnetic fields as a possible human carcinogen."
Studies into particular aspects of childhood leukaemia
- Most studies have looked at whether EMFs cause childhood leukaemia but one recent study has looked at whether magnetic fields affect survival from leukaemia.
- One study has looked at children with Down's syndrome rather than children in general.
- There has been some interest in whether night-time exposure in particular might be relevant
- Two epidemiological studies have looked at possible alternative explanations: corona ions and contact currents
How many children may be affected?
The HPA calculate that the raised risk for childhood leukaemia suggested by epidemiological studies, if it were real, would correspond roughly to an extra two cases of childhood leukaemia per year in the UK, compared to an annual total of around 500 cases. More on numbers of children and cases.