What the WHO's Cellphone-Cancer Statement Really Means

A Q&A with bioengineer Kenneth Foster


On 31 May, the International Agency for Research on Cancer (IARC), a part of the World Health Organization, announced that after a concentrated week of combing the scientific literature on cellphone use and brain cancer, it decided to classify radio-frequency electromagnetic fields from cellphone use as "possibly carcinogenic."

Many press reports pointed out that this puts the hazards of cellphone use in the same category (Group 2B) as lead and chloroform. However, we fear lead because it causes brain development problems in children and chloroform because it’s acutely toxic. Neither of these effects has anything to do with cancer. On the other side of the fence, the wireless communications trade organization, CTIA, highlighted two innocuous-sounding members of Group 2B: coffee and pickled vegetables.

Seems like you can easily cherry-pick the list to inspire whatever emotion you want to. So what's the real deal? What does the designation actually mean, and how should we, as a society, respond to it?

To get some answers, I spoke with Kenneth Foster, a bioengineer at the University of Pennsylvania who has been studying the health effects of radio-frequency energy since 1971. Foster, an IEEE member who serves on the Institute's standards committee for setting radio-frequency exposure limits, gave me the scoop on how the IARC decision was made, what it means for exposure standards, and what it means for consumers.

Is there a known mechanism that links radiation from your cellphone to cancer?

That's the whole problem with the field. With ionizing radiation, such as X-rays, there's a very clear mechanism. The photons of X-rays will break chemical bonds producing free radicals, which are chemically very reactive. Scientists have been studying how electro-magnetic fields in the non-ionizing range, such as radio-frequency energy, interact with biological matter since probably around 1900 or even before, when the first technology was developed to produce this energy. And nothing has emerged that would indicate how such fields could produce biological effects other than heating.

We know that if you have extremely strong radio-frequency fields, they will exert forces on molecules themselves, but that requires field strengths that are vastly higher than anything you're going to get from a cellphone. This doesn't mean that it's absolutely physically impossible that cellphone radiation can do something, but it seems pretty unlikely.

Now, of course, the activists will say that there are non-thermal mechanisms, but after a century of study, none has emerged. I, myself, have spent the last 40 years thinking about mechanisms that might possibly cause biological effects other than thermal exposure levels, and I really haven't come up with anything. There's certainly a lot of speculation published in the scientific literature about possible mechanisms, but usually if you look at them, there's something wrong with the science, and the suggestions really aren't very plausible.

So in the absence of a mechanism, does it make sense that IARC gave radio-frequency radiation a “possibly carcinogenic” designation?

IARC operates under a very clearly defined and rigid set of rules, based mostly on epidemiology data. These are statistical studies of human health, as related to some kind of exposure. At this point, there are dozens of epidemiology studies. The most famous ones are those of the Interphone series, mostly conducted in Europe. And these studies generally are negative. In fact, they're quite solidly negative except that some exposure conditions — namely people who use cellphones for more than ten years — seem to suggest a barely detectable increase in the likelihood of a couple rare kinds of tumor: gliomas (a malignant tumor) and acoustic neuromas (a benign tumor). IARC made their decision mostly on this basis. They don't look at the basic biophysics, and they look at animal studies, but only in a secondary way.

And again, they didn't say that these fields do cause cancer, they say that the epidemiology is suggestive that something might be there, but the evidence falls short of actually demonstrating that fields do cause cancer. This is very different, say, from smoking and lung caner, where the epidemiology is extremely strong, and the animal studies are extremely strong.

With radio-frequency energy and these two brain cancers, the evidence is exactly the opposite: it's very weak, and the IARC committee is willing to say that it's possible there might be something there, but they don't view the evidence as strong enough to say that there actually is a health effect.

What further research is needed?

People who are concerned about this issue will quickly point out that these epidemiology studies are flawed and not really well designed to study or identify small effects from long-term use. Their basic problem is lack of exposure assessment. I don't know myself how much I used my cellphone two years ago, let alone 10 years ago. And so these studies really are quite flawed for the kinds of interpretations that people are trying to draw from them — mainly identifying small risks from long-term use.

So, clearly, you'd need some kind of long-term exposure assessment to remedy these problems. Of course, that's a very hard thing to do.

The other thing that the health agencies are calling for are the studies of the effects of mobile phones on children based upon the general idea that kids may be more susceptible to toxic effects. Of course, they have longer lives in front of them than adults, so if there's something there, it might be a greater risk of developing something over the long term than adults. Those also are very problematic to do.

I think that some of the studies that activists and scientists are calling for are going to provide employment for scientists for many years in the future. It's an extremely hard problem to address.

So what should we do in the meantime? Specifically, any recommendations for children and cellphone use?

My own perspective is that, I've been in the field since 1971 and I've been hearing this same general concern expressed by some lay people for forty years, and science has been saying we need more research. After all that time we haven't identified any real, clear hazards. My general inclination is that it's pretty unlikely we will with more research. So I'm not really advocating a huge research program on this, although other people are.

If for any reason whatsoever someone is concerned, the most useful thing they can do is either not use a cellphone or use a hands-free kit. This is a very inexpensive and totally effective way for them to reduce their exposure.

And there is possibly a lot of work that should be done in risk communication. Looking over the press coverage of this IARC study, I see lots of really misleading coverage, even by well-intentioned journalists. For instance, journalists will point out that other 2B carcinogens are gasoline or lead and they will compare cellphone use with gasoline or lead. Gasoline and lead are both very hazardous, but what IARC is interested in is just one specific biological effect: connection with cancer. In both cases — gasoline and lead — the connection is unproven. And I think to explain to the public what these IARC classifications mean, to try to help the public put this issue in a broader context to reduce their overall risks is something that really needs to be done.

Could you say a little about the IEEE standards committee that you're on?

It's the IEEE International Committee on Electromagnetic Safety (SCC39), which has been in business in one name or another probably since the 1960s, and every five years or so it sets a new set of limits for human exposure to radio-frequency radiation. The committee consists of biologists and engineers and some physicians and it searches through the literature to find any evidence of health effects or safety hazards of radio-frequency energy. Every few years, there's this massive literature review to identify hazards and then the limits are set below the identified hazards by a safety factor of somewhere between 10 and 50. The hazards that have been identified generally involve the excessive heating of tissue, and in fact the only standards that are universally accepted involve too much heating. If you put a rat in a microwave oven, it's obviously very dangerous for the rat. And these are the kinds of hazards that are generally described in the standards.

There's another set of standards by a group whose acronym is ICNIRP, whose studied a similar process, mostly in Europe, and they've developed a similar idea what the hazards are of radio-frequency energy, and their standards are quite similar to those of the IEEE.

Are those standards or limits going to change in light of this new designation?

I think it's highly unlikely. The IARC designation is a long way from actually saying it is causing cancer. And I'm quite sure that the standards committees — both the IEEE and ICNIRP and those elsewhere around the world — are going to continue to base their limits on identified hazards. What will most likely happen is that various government agencies might issue precautionary recommendations in addition to these limits, saying that we know that your exposure to radio-frequency energy from cellphones is below these accepted limits, but as a precautionary matter, you might want to use a hands-free kit or limit your calls.

It would be very difficult to change the limits unless you had some clear idea of what the problem is, what you're trying to protect against. And the evidence that led IARC to call radio-frequency fields a class 2B carcinogen just really isn't specific enough to be able to use it to set any kind of exposure limits. It's not even clear that there actually is a hazard — there's just suspicion that there might be.

Could you talk about specific absorption rates (SARs) and whether or not you think cellphones should be labeled with them?

Specific absorption rates are basically the amount of power that's absorbed in the body per given mass. This is a measure of exposure that makes most sense if you're talking about thermal effects — the temperature rise is related to how much power you’re absorbing in your body. In the United States and virtually every place else, governments require manufacturers of cellphones to show that the exposure in terms of SAR is below the international limits. As a practical matter, the exposures are always very far below the international limits as the cellphones are actually used.

Consumers can find the SAR values from mobile phones by going to the FCC website or various other websites, and this is already publicly available information.

The problem with this, however, is that the actual exposure that somebody gets from using a cellphone depends as much on how close a phone is to a base station as it is on the actual design of the phone. Phone's now all use adaptive power control, which has them reduce their power output if they're in a region with a strong signal from a base station, probably to conserve power in the batteries, but also perhaps to reduce interference with other users. So having a value for the SAR in a head measured under worst-case conditions in a laboratory doesn't necessarily help you very much in determining the exposure that you might actually get.

If you have two phones with different wireless carriers that have the same SAR value measured in the laboratory, the actual exposure that someone might get at a given place and time might be very different from one to the other simply because there might be a stronger base station signal.

So the listed SAR for a given phone — is that the upper limit of what that phone can put out?

A company cannot sell a phone that produces a greater SAR than the exposure limits. These exposures are measured in a laboratory with a phone modified to transmit one hundred percent of the time. But in real use, due to adaptive power control, the phone transmits at lower levels than its theoretical maximum.

It's like saying that my car has a maximum velocity of 150 miles an hour. Well, that's a theoretical maximum that maybe you might achieve under unusual conditions, but in practice is doesn't go that fast at any given time…at least not the way I drive.

The problem is that governments want to give consumers information about how much exposure they get from cellphones because of a general ethical idea that you want to give consumers information about the products they use. But unfortunately, the SAR is not a very useful measure of exposure because of all the factors that I just mentioned.

But there's no better measure.

To a consumer who wants to reduce their exposure to radiation from cellphones, they would be much better off just not even thinking about the SAR and using a hands-free kit, which takes the phone away from their head, and certainly reduces their exposure. And hands-free kits are very cheap.

Roughly speaking, how does the amount of radiation your head absorbs while you're talking on the cellphone compare to the typical amount of ambient radiation coming from base stations?

The phone itself typically transmits 100 milliwatts of power when it's transmitting. In a typical usage situation, it may be transmitting only a small fraction of the time, so the average power is somewhat less than that. And, on average, maybe one-third of the energy from a hand-set will go into your head, the other two-thirds will be transmitting into space. This is a low level compared with anything that has been measured to be dangerous, but it's a much higher exposure than you would get by living in the vicinity of a wireless base station, simply because you're putting this transmitter right up against your head.

And I think that probably the people who are inclined to worry about the health effects of cellphones are very aware of the fact that never before in human history has a large fraction of the world's population been walking around with radio-frequency transmitters glued to their heads. Whether this is dangerous or not, of course, is a very different question.

Anything else that you'd like to add?

I don't see any cause for great alarm from these findings. Given the way that IARC works, this decision is completely predictable, and probably even required by their rules. They're saying basically that there's some level of suspicion, and scientists have agreed with this all along. This doesn't, in my mind, justify extreme measures. We're talking about a barely detectable increase in the risk of very rare diseases, which nobody is even sure indicate a real biological effect as opposed to some problem with the studies.

In the meanwhile, though, people are exposed to much bigger risks in everyday life. Bad diet, driving around carelessly, and so forth. For the average person, I think they would be much better off making sure their kids are buckled into their seat belts and that they have a reasonable diet and don't smoke and reduce their alcohol consumption. These of course, involve much bigger risks than the theoretical possibility that maybe there might be some connection between the use of cellphones and specific kinds of brain tumors. I think people need to put this in a much broader perspective.

Corrections: "In fact, they're quite solidly negative except that some exposure conditions — namely people who use cellphones for more than ten years — seem to suggest a barely detectable increase in the likelihood of a couple rare kinds of tumor: gliomas (a malignant tumor) and acoustic neuromas (a benign tumor)." originally appeared as "...a couple rare kinds of cancer: gliomas and acoustic neuromas."

"IEEE International Committee on Electromagnetic Safety (SCC39)" originally appeared as "IEEE Standards Coordinating Committee 28."

PHOTO: Diana Trent

The Tech Alert Newsletter

Receive latest technology science and technology news & analysis from IEEE Spectrum every Thursday.

About the Tech Talk blog

IEEE Spectrum’s general technology blog, featuring news, analysis, and opinions about engineering, consumer electronics, and technology and society, from the editorial staff and freelance contributors.