Do high voltage power lines cause cancer?

This could be a very short article. I could just write “no, power lines don’t cause cancer"—but that wouldn't explain why so many people believe otherwise. And it won’t help people who are thinking about buying a home that has power lines nearby. So let’s look at this question a bit more closely.

For the past century or more, humans have been surrounding ourselves with an ever-growing array of electrical devices. All of these devices create electrical or magnetic fields, often called EMFs. There’s no doubt that our exposure to EMFs has increased dramatically in modern times. Not surprisingly, many people have worried that this is a bad thing. The belief is so pervasive that NIH has at least two websites devoted to this topic, one by NIEHS and one by NCI, as does the Medical College of Wisconsin. Realtors have created webpages to inform home buyers about how power lines might affect the value of their home. Not surprisingly, you can easily find companies on the Internet that will sell you devices (such as SafeSpace and EMFshield) to protect your body from the supposed perils of EMF.

People worry especially about high-voltage power lines, probably because they are carried by very large, highly visible structures that look vaguely threatening. This fear seems to have started with a 1979 study in which Nancy Wertheimer and Ed Leeper reported a correlation between high-voltage power lines and childhood leukemia in the area around Denver, Colorado. Wertheimer's results spurred numerous studies in the years since. A review of the evidence in 1995 pointed out that
“There is no known mechanism by which magnetic fields of the type generated by high voltage power lines can play a role in cancer development. Nevertheless, epidemiologic research has rather consistently found associations between residential magnetic field exposure and cancer.”
Scientifically, the question at the time was, were these associations real or coincidental?  If they were real, what’s the mechanism? Clearly, further studies were needed. Well, twenty years later, the data are in: power lines do not cause cancer.

In 2002, the WHO commissioned a huge (339 pages) and very thorough report on all the types of electrical and magnetic fields on the planet and how these EMFs might effect our health. Among its findings were:
“There is little experimental or theoretical evidence that mutations could be directly caused by ELF [extremely low frequency] magnetic fields…. There is little evidence that ELF electric or magnetic fields can cause malignant transformation of cells in culture.”
The final conclusion of the WHO commission was that
“Static electric and magnetic fields and extremely low-frequency electric fields are not classifiable as to their carcinogenicity to humans (Group 3).”
Group 3 means we don’t have any positive evidence that EMFs cause cancer. The only lower category, Group 4, would mean we have evidence that electromagnetic fields do NOT cause cancer, but such evidence is very difficult to produce. In other words, they concluded that the evidence didn't support a link, but more studies might yet find something.

After the 2002 report by the WHO, a study in 2005 raised the alarm again. In that study, Gerald Draper and colleagues claimed to find an association between the distance to the nearest high voltage power line and childhood leukemia. Draper found that living less than 200 meters from these power lines (in England and Wales) raised the risk of leukemia significantly compared to living at least 600 meters away.

The scientific reaction to the Draper study immediate and highly critical. Hepworth and colleagues pointed out that the results did not support a causal role for electromagnetic fields (which were not measured), but at best a geographic correlation. Kheifets and colleagues demonstrated out that the effect disappeared when the control groups were analyzed differently. Other critiques quickly emerged as well: a sign that science was working to self-correct, as it often does. But Draper’s study was widely reported, while the criticisms were not. The critiques, though, paint a compelling picture that Draper’s work was seriously flawed.

One of the most recent studies is from 2013 by Elliott et al. who looked at over 50,000 cases of cancer, including leukemia, brain cancer, breast cancer, skin cancer, and others. They found no increased risk for any of these cancer types and concluded
“Our results do not support an epidemiologic association of adult cancers with residential magnetic fields in proximity to high-voltage overhead power lines.”
This debate sounds very familiar. Many false hypotheses, such as the notion that vaccines cause autism, or that acupuncture can reduce pain, show the same pattern: a few small studies produce weak positive evidence, but then larger, better studies fail to back them up. Proponents always call for more studies, but if the effect is real, it doesn't disappear when you do a bigger study. If anything, the effect should appear stronger.

A major problem that the EMF alarmists have, which none of the proponents have ever answered, is one of mechanism: how is the very weak EMF from a power line supposed to cause cancer? Multiple theories have been suggested: maybe EMFs affect the movement of magnetic particles within cells, or alter the voltages across cell membranes—but as the editor of BMJ, Geoff Watts, put it in his response to the 2005 Draper study:
“Evidence to support these and other ideas is at best thin and at worst non-existent.”
So no, electrical power lines do not cause cancer. But they're still ugly. We should bury them all underground.


  1. Unfortunately many people have the habit to cherry pick those studies which proves their point, but [at best] ignore the others or [at worst] claim that those are biased somehow.

  2. Something we really need to get out to the public is that although correlation studies may be useful in some cases for finding trends (such as how the smoking/lung cancer association was originally found), the whole point of science is to show *causation*, because it is quite possible that two correlated trends do not have a causal relationship (as in the classic correlation of crime and ice-cream sales as both are higher in summer rather than one causing the other)


Markup Key:
- <b>bold</b> = bold
- <i>italic</i> = italic
- <a href="">FoS</a> = FoS