D.L. Henshaw, November 20th, 2004
Summary of a paper published in Medical Hypotheses:
Does our electricity distribution system pose a serious risk to public health?
D.L. Henshaw, 2002, 59 No.1, 39-51
Note: This summary published in 2002 has been largely superceded by the more
California Health Department Report
The research literature reveals that for some illnesses there is a degree of consistency in the evidence suggesting adverse health effects of living near high voltage powerlines. The evidence comes from two principal sources: (i) the body of epidemiological studies and (ii) a risk analysis based on increased exposure to air pollution near powerlines.
The following areas can be considered:
1. Electric field effects
In the case of illnesses associated with air pollution, risk analyses can be performed based on the increased probability of lung deposition of inhaled pollutant aerosols that have been electrically charged by powerline corona ions compared with uncharged aerosols.
1.1 Corona ion effects
1.1.1 Childhood leukaemia
The attached reference list cites a number of papers where childhood
leukaemia has been associated with traffic density and motor vehicle pollution.
ions are assumed to be effective at increasing exposure to air pollution up
to 300 metres downwind of powerlines, in the prevailing south westerly wind
direction (Fews 1999a). A 30% increase in exposure has been assumed. The proportion
of the population living within 300 metres of 132, 275 and 400 kV powerlines
is assumed to be 2.9%. The number of excess cases of childhood leukaemia is
therefore given by 600 × 2.9% × 0.3 × 0.5 = 2.6 cases or
approximately 3 cases. However, the non-downwind quadrants near powerlines
might also be affected by corona ion effects. To reflect this uncertainty the
table provides the range of the possible number of cases from both magnetic
field and corona ion effects.
Approximately 3 cases annually
1.1.2 Lung cancer
Erren (1996) reviewed five studies where lung cancer has been associated with EMF exposure. This included the UK study by McDowall (1996). Here, the author considered cancer incidence in East Anglia, in populations living up to 50 m from electrical installations, mainly substations, although he did not specify which were fed by overhead powerlines. Within 15 m of an installation, elevated SMRs were seen for lung cancer, all leukaemias, other lymphatic neoplasms and all respiratory disease. Only the result for lung cancer was statistically significant (odds ratio = 2.15, 95% CI = 1.18 – 3.61) and this was mainly driven by an effect in women. The odds ratios for lung cancer showed a consistent gradient of increasing excess mortality with proximity to the line, but at distances greater than 15 m these were not statistically significant.
Lung cancer is known to associated with air pollution with increased risks in the range 1.3 to 2.5 (Katsouyanni & Pershagen 1997). Corona ions emitted from high voltage powerlines increase the charge state of pollutant aerosol particles in the air. Aerosols in the size range 20 to 200 nm are of special interest, especially those containing PAHs such as benzo[a]pyrene. There is evidence that in this size range the effect of single charges on aerosols is sufficient to increase the deposition of inhaled aerosols in the tracheobronchial lung region by a factor of 2 to 3 (Cohen et al 1998).
The risk calculation takes the affected population as living within 400 metres of high voltage powerlines, downwind of the prevailing south-westerly wind. An average 15% aerosol charging by single charges is assumed to lead to a 30% increase in lung deposition of inhaled aerosols. The average male/female lung cancer rate in the UK is taken to be 74 per 100,000 per year. The number of people living within 400 m of 132, 275 and 400 kV powerlines is taken to be 4.6% × 6 × 107 people = 2.76 × 106 people. A 30% increase in risk downwind compared with upwind of powerlines is assumed. This yields 306 cases annually. The range quoted in the table of 250 – 400 cases annually takes into account two possibilities: (i) that on average corona ion effects may not extend to 400 m from powerlines or (ii) that a contribution to risk in those living upwind of the prevailing south-westerly wind should be included.
Approximately 250-400 cases annually
1.1.3 Other illnesses linked to air pollution
Seaton (1995) has discussed the range of illnesses associated with air pollution,
especially respiratory and cardiovascular disease. If these are increased near
powerlines as a result of increased lung deposition of inhaled aerosols charged
by corona ions, then the number of excess cases could reach several thousand.
Not all of these would be fatal.
Approximately 2,000-3,000 cases annually
1.2 Oscillation of polluted particles leading to increased
deposition on the skin
Fews et al 1999b made experimental measurements of the increased deposition of radon decay product aerosols on model heads under high voltage powerlines outdoors. Increased deposition in the range 1.4 to 2.9 was found. It was also observed that the deposition rate of radon decay product aerosols outdoors is about 20 times higher than indoors. This is consistent with the known deposition velocity of outdoor and indoor aerosols. The practical result is that in the UK the dose rate to the basal layer of the skin outdoors arising from the plateout of radon decay products is likely to be around ten times higher than that indoors, even though the radon decay product concentration in air outdoors is very low.
The ICRP quotes an excess relative risk of non-melanoma skin cancer of around 60% per Sv (NRPB 1997). On this basis a risk analysis can be made for radiation induced skin cancer as a result of living close to high voltage overhead powerlines. A preliminary study by Preece et al (1996) found a 1.6-fold increase in non-melanoma skin cancer in people living within 20 m of high voltage powerlines in south-west England (RR = 1.62, 95% CI = 1.06 – 2.47).
For the risk analysis, we take the average male/female non-melanoma skin cancer rate to be 41.6 per 100,000 per year. The exposed population consists of those living very close to the line, say within 25 m. This corresponds to 0.14% or 0.14% × 6 × 107 = 82,500 people. Applying the skin cancer rate and assuming a 40% increase in risk yields 14 cases annually.
Approximately 14 cases annually
2. Magnetic field effects
(i) In the case of childhood leukaemia, the pooled analyses by Ahlbom
et al (2000) and Greenland et al (2000) suggest an approximate doubling of
risk for magnetic field exposures above 0.3/0.4 µT.
(ii) In the case of depression and suicide, there is a body of evidence in the scientific literature showing a general consistency of increased risk in relation to magnetic field exposures. It is of interest that an increase in risk appears at a low threshold of ~0.1 µT. It should be noted that to date this literature has not been well reviewed by bodies such as the US National Institute of Environmental Health Sciences (NIEHS) nor by the UK National Radiological Protection Board (NRPB).
This document makes an initial attempt to quantify the likely number of cases of ill health annually that might occur in populations living near high voltage powerlines in the UK if the level of risk indicated by the epidemiological studies and the risk analyses was to be realised. If as is implied by these estimates, several thousand cases of illness annually are associated with living near high voltage powerlines in the UK then this could be of significant public health relevance.
2.1 Childhood leukaemia
Increased risk has been assumed above 0.4 µT, effective up to 50 metres either side of 132, 275 and 400 kV powerlines. The proportion of the population living within 50 metres is estimated as 0.275%. If there are 600 cases of childhood ALL annually in the UK, this corresponds to 1.7 or approximately 2 cases only.
Approximately 2 cases annually
2.2 Suicide and Depression
The literature contains a number of papers associating both suicide and depression
with exposure to magnetic fields, including near powerlines. Increased risk
of both suicide and depression are both considered biologically plausible either
by reduced production of melatonin by magnetic fields or by the magnetic field
induction of electric fields in the body. A discussion may be found in Wijngaarden
et al (2000).
The literature reveals a number of features:
1. A general consistency that both suicide and depression are associated with power frequency magnetic field exposure. Some studies also hint at an association with power frequency electric fields.
2. A threshold effect occurring at low magnetic field exposures, ~0.1 µT. Such a low threshold would embrace exposures near all types of powerlines not merely those at 132 kV and above.
3. Occupational studies appear to show lower effects than for residential studies. This would be consistent with a mechanistic effect associated with reduced melatonin production, which occurs mainly at night and therefore has a larger effect on chronically exposed populations.
The average suicide rate for males and females is taken to be 9.6 per 100,000 per year. An exposure threshold of 0.1 µT is assumed which is effective up to 150 m either side of 132, 275 and 400 kV powerlines. This embraces 1.05% of the population. The exposed population is therefore 1.05% × 6 × 107 = 630,000 people. Assume the risk to be doubled.
Approximately 60 cases annually
(ii) Mild depression
Again take an exposed population of 630,000 people. Some estimates suggest that 15% of the population experience an episode of mild depression each year. If there is a 40% increase in risk above 0.1 µT, this would lead to a large number of cases of mild depression associated with magnetic field exposure. The value quoted in the table of 9,000 cases annually is a conservative estimate.
Approximately 9,000 cases annually
Notes on the Table of Risks
These notes explain how the number of excess cases in each category was estimated. The proportion of the population living near powerlines has been estimated using the data for 275 kV and 400 kV given in figure 1 of UKCCS (2000). It has then been assumed that the proportion of the population living near 132 kV powerlines is a factor 1.5 greater. In each case a conservative estimate has been made of the range of effective magnetic fields.