"CCRG" is the Childhood Cancer Research Group at the University of Oxford. The original, 2005, CCRG paper on childhood cancer and power lines (often called the "Draper" study) looked at:
- 275 and 400 kV overhead power lines
- England and Wales
- Distances from the power lines up to 600 m
A "follow on" paper was published in 2014 which extends these to include:
- 132 kV lines (in certain areas of the country only, owing to availability of data)
- Distances up to 100 m
- more recent cases from 1998 - 2008
The results in summary
As with the original paper, there are no clear patterns of results for brain tumours or for "other cancers". All the following discussion relates to childhood leukaemia.
132 kV lines seemed to show similar results to the 275 and 400 kV lines, but generally a smaller effect and confined to smaller distances. See more on this further down this page.
Scotland produced results consistent with England and Wales, but because the number of cases in Scotland is much smaller, the results from Scotland on their own are not statistically significant.
The original elevated risks for childhood leukaemia extended to 600 m, the farthest distance tested, which raised the question of the whether there would have been an elevation beyond that had the study looked. This follow-on study finds no elevations between 600 m and 1000 m, suggesting that the original 600 m was in fact the outer limit.
However, the really interesting finding comes from including the later cases and looking at the results by decade. The risk for childhood leukaemia seems to have diminished over time, as summarised in this figure.
(These are the results for 275 and 400 kV lines and 132 kV lines in certain areas, in England, Wales and Scotland. "1962-1995" is the previous result so for 275 and 400 kV only.)
Results for 132 kV power lines in particular
The original paper just looked at 275 kV and 400 kV power lines. This paper looks at the next voltage down, 132 kV power lines as well.
Data on the locations of 132 kV power lines were provided by the Distribution Network operators (DNOs). Every DNO provided grid references, but only some were able to provide years of construction. If you included data from all DNOs, you would run the risk that a case would be recorded as being near a line when the line hadn't actually been built at the time. If you restricted the analysis to only areas with years of construction, you would have robust data, but you would reduce the size of the study. The analysis was done both ways and there wasn't a lot of difference. The paper presents the results for 132 kV lines just in areas where years of construction were available for 80% or more of lines.
The comparison of 275/400 kV and 132 kV power lines for the 0-199 m distance band is shown here:
And for the 200-600 m band (note the different vertical axis):
We observe from this:
- Within the closer distance band, 0-199 m, the elevation of leukaemia rates is there for 132 kV lines as well as for 275/400 kV lines - perhaps a slightly lower elevation, but given the uncertainty in the data, we can't be dogmatic about this
- The elevated leukaemia rates for 132 kV lines seem to fall over time in a similar way to for 275/400 kV lines
- In the farther distance band, 200-600 m, the elevation for 132 kV lines is fairly clearly less than for 275/400 kV, in fact it's not really there at all
So we conclude that 132 kV lines seem to produce a similar elevation to 275/400 kV lines, except perhaps slightly smaller, and not extending as far from the lines.
What does this mean?
We were already virtually sure that the CCRG finding could not be explained solely by magnetic fields, because it extended too far from the power lines - well beyond the range of magnetic fields. This finding makes it even more certain magnetic fields are not the explanation, as it is implausible that any biophysical effect of magnetic fields would change over time like this. (Magnetic fields could, in principle, still be the explanation for the bit of the elevation close enough to power lines, say within about 50 m.)
There is a theory that the power lines have health effects through the corona ions they produce (small ions created from air molecules on the surface of the conductors) which attach to existing airborne pollutants and cause increased deposition of these pollutants in the body. There are strong theoretical reasons to doubt this theory. But at least some pollutants have reduced over the same time period that the elevated risks near power lines also seem to have gone away, possibly putting this theory back into consideration. The test of this would be if the rates of leukaemia were greater downwind compared to upwind. The original 2005 result did not show any excess downwind, but using a rather simple test. A future paper will test this on the new results using a better model.
Rates of childhood leukaemia vary with a package of socioeconomic factors. We also know that some socioeconomic factors vary in places near power lines. So it is at least possible that the higher leukaemia risks near power lines are not a direct effect of the power lines but a consequence of the population characteristics of the areas power lines pass through. If the types of homes built near power lines have changed over the years, or if the effect were a transient one that only lasted a certain time after the power line was built, this could explain a risk that diminished over the years. We don't know exactly what the factors are which affect childhood leukaemia, nor exactly how the various factors vary near power lines, so this is rather speculative - but seems to the authors themselves to be possibly the most likely explanation.
Funding and independence
The original study was funded by the Department of Health. This follow-on was funded by the charity Children with Cancer (formerly Children with Leukaemia).
In all this work, the data and calculations on exposure were provided by National Grid (and, in this follow-on study, data on 132 kV lines was provided by the Distribution Network Operators). The sientist on National Grid's staff who performed this work is one of the authors of the paper. The paper states:
"JS worked on this paper with permission from National Grid, but no results were shared, nor was approval sought from anyone in National Grid and the views expressed are his alone."
This sample contract between industries and universities, designed to guarantee the independence of the study, is in fact the one first created to cover this study.
The abstract in full
Residential distance at birth from overhead high-voltage powerlines: childhood cancer risk in Britain 1962-2008
Bunch KJ, Keegan TJ, Swanson J, Vincent TJ, Murphy MFG
Background: We extend our previous study of childhood leukaemia and proximity to high-voltage powerlines by including more recent data and cases and controls from Scotland, by considering 132-kV powerlines as well as 275 and 400 kV, and by looking at greater distances from the powerlines.
Methods: Case–control study using 53 515 children from the National Registry of Childhood Tumours 1962–2008, matched
controls, and calculated distances of mother’s address at child’s birth to powerlines at 132, 275 and 400 kV in England, Wales and Scotland.
Results: Our previous finding of an excess risk for leukaemia at distances out to 600m declines over time. Relative risk and confidence interval for leukaemia, 0–199m compared with >1000 m, all voltages: 1960s 4.50 (0.97–20.83), 2000 s 0.71 (0.49–1.03), aggregate over whole period 1.12 (0.90–1.38). Increased risk does not extend beyond 600 m, but may be present, albeit less strongly, for 132 kV lines.
Conclusion: A risk declining over time is unlikely to arise from any physical effect of the powerlines and is more likely to be the
result of changing population characteristics among those living near powerlines.