Cellphone radiation: only the facts

Is cellphone radiation really bad for you? The debate rages.

This is a difficult topic. It's difficult because of the strong emotions on each side, and because science and popular opinion are rarely in sync. Every so often a new study hits the headlines, claiming to have found (or disproved) a connection between cellphones and cancer, and each time one side claims progress, the other side points out flaws in the argument. But some of the arguments really are nonsense, and some of the science is questionable. So let's try to separate the wheat from the chaff.

(There is a lot of research and related scientific content out there. I'm only going to touch on selected bits of it, or this article would go on forever. But, ITWeb will cover new research as it comes out, and update this story as time goes by. If there are specific issues you'd like to discuss, please feel free to use the comments section to raise points or highlight research you think we've missed.)

Fact 1: No research has proved any ill effects caused by the electromagnetic radiation used in cellular networks. To the best of our knowledge, there is no health risk.

Fact 2: Cellphone radiation can indeed have biological effects on the human body.

Fact 3: No authoritative body has issued health warnings about cellphone radiation.

Fact 4: It is possible there is a health risk we don't know about yet. Public bodies have acknowledged this possibility, and are frequently misunderstood when they do.

Fact 5: Fewer cell towers means more radiation, not less.

Let's dive right in to the research. There is no research currently available which shows any health risks from cellphones.

That, frankly, doesn't mean much. There is also plenty of research which shows there might be health issues, but nothing conclusive yet. One of the best research projects (Interphone) observed no definite health risk, but was, by its own admission, probably inconclusive.

To understand this, you need to understand how scientific studies work. Research, to be conclusive, has to follow very strict protocols to eliminate circumstantial factors, and it has to be consistent over a long period of time. This is especially true when investigating possible carcinogens (causes of cancer), because cancer can take a long time to manifest. Studies generally span at least a decade, and have to control for a huge number of possible variations in the study group, which could skew the results.

This immediately makes studying cellphones extremely difficult, simply because the field evolves so fast. Technology in use today is profoundly different from that of a decade ago. So is our usage of that technology, and so are the environments we live in. That makes conducting a study (or at least, drawing conclusions from one) very difficult indeed. Scientists accommodate the shifting playing field by setting up control or by casting the net as wide as possible, but it is vanishingly rare to see research which cannot be criticised or questioned.

The largest study into cellphone radiation was the Interphone initiative, which collected research data from 13 countries and spanned over a decade, starting in 2000. Interphone is worth highlighting because of its sheer scope, but there is an enormous amount of research that has generally found similar results.

Despite the scope of the Interphone project, nothing conclusive was found - specifically, no link between cellphones and cancer. The key findings from the Interphone study boil down to the following:

1) If you have ever used a cellphone, you may be at slightly higher risk from cancer compared to someone who has not, but that higher risk may well be due to other environmental factors reflecting modern lifestyles, and as such cannot be considered conclusive.

2) How much you use your phone doesn't appear to have any impact on the likelihood of cancer.

3) Because the field has changed so much, the findings may well be inconclusive. When the study started, the maximum amount of time they imagined anyone using a phone was 30 minutes a day. These days, that's likely a minimum. And newer mobile phones generate much less radiation than they did when the study began.

Most other studies offer similarly conclusions, and similar flaws. Those flaws aren't a reflection on the scientists, just on the difficulty of the subject. Studies which have found cellphone users with higher rates of health problems, for example, often fail to eliminate environmental factors, which means you have correlation, not causation, a red flag to a scientist.

A follow-up study to Interphone, Mobi-kids, is investigating more specifically whether young people are at risk from cellphone radiation.

Meta-studies, which aggregate data from multiple sources to identify patterns, have similar challenges. They try to smooth out circumstantial data by casting a broader net, but immediately raise questions about the selection criteria for the studies themselves.

To conclude the science part of this discussion, I'll remind you that it is impossible to prove a negative. Science cannot prove, for example, that cellphones are not bad for you (if there is no evidence of health risks, maybe we were looking in the wrong places).

This means scientists will never be able to state categorically that cellphones are not bad for you. Anyone who claims that is fibbing.

Conversely, critics who use that lack of assertion to prove their converse point of view - the “scientists can't say it's safe, therefore it's unsafe” argument - are equally mistaken.

Mobile networks use radio waves to communicate, in several frequencies in the microwave spectrum, but particularly around 800MHz, 1 800MHz and 2 600MHz.

Microwave radiation is “non-ionising”, meaning it lacks sufficient energy to separate electrons from atoms. Ionisation creates ions, and can stimulate chemical reactions. In living tissue, that process can cause cancer due to damaged DNA, or radiation poisoning in extreme doses. You're probably familiar with the triangular “radioactive” warning sign - that symbol more specifically means “ionising radiation hazard”. Ultraviolet light is at the very beginning of the ionising radiation spectrum, and we all know the health risks of too much exposure to the sun.

Non-ionising radiation is much less dangerous, but definitely can affect the human body, mostly by generating heat via a phenomenon known as dielectric heating. Thus, some of the energy from cellular signals will be converted into heat inside the body.

How much heat? Not much, but answering that question in detail is going to get technical, so bear with me.

The strength of a signal is measured in watts, as the unit of power. Mobile phones emit signals with a peak output of 2W. WiFi access points usually peak at around 1W. Many other common household devices generate electromagnetic fields - we are constantly surrounded by low-strength electromagnetic fields.

The power of any radio signal attenuates exponentially according to the inverse square law. In English, that means the power decreases proportionally to the square of the distance from the source. Double the distance, quarter the power. Ten times the distance, one hundredth the power, and so on. I'll look at this more closely when I talk about cellphone base stations, but for now, I'll focus on phones.

The reason cellphones are so closely studied is that the distance between the phone and sensitive tissue is so small. The antenna is effectively pressed against your head for long periods of time, so the question of how much energy passes from phone to user is the key.

How much energy is absorbed by living tissue is known as the Specific Absorption Rate, measured in watts per kilogram, and safety standards are set to limit the SAR of phones on the market. In most countries, the limit is set at or below 2W/kg: most modern phones test below 1.5W/kg.

Most of that heat will be absorbed by the parts of the body in close proximity with the phone - the head, the hand, and perhaps the thigh if you carry it in your pocket - but studies focus almost exclusively on the head because of the concern about brain tumours.

1.5W/kg is a miniscule amount of energy, barely enough to raise the temperature of living tissue a measurable percent of one degree. This is distinct from the heat generated by the phone in normal operation - smartphones, like any computing device, generate heat, but we are looking here at the effects of radiation, since that penetrates deeper into tissue. The surface of your skin does a good job of dissipating the ambient heat of the cellphone.

Thermal energy from EM radiation is normally dissipated by blood flow through the tissues, and has no effect at all. Not all tissues can regulate their temperature - scientists managed to cause cataracts in rabbits by exposing their eyes to effective energies over 100W/kg. Other animals weren't affected the same way, and that is vastly more energy than you'll get from heavy cellphone usage, but the effect must be acknowledged.

There may be other effects too. One study suggests that the brain, when exposed to radiation from a cellphone, could exhibit higher than normal levels of glucose metabolism (the rate at which cells turn the glucose into energy). Whether that result will hold up, and whether it has any medical significance, isn't yet certain; but again, it's on the table.

And a number of questions revolve around cell development: whether exposing existing tumours to increased levels of radiation could stimulate development or malignancy, whether radiation could have an effect on a developing foetus, and whether reproductive cells could be affected. As with other studies, nothing concrete has been uncovered, but under the right conditions (as with the rabbits' cataracts) some studies have claimed to show results.

One thing is certain: there are a lot of cellphones in the world. Market penetration is nearly 100% in many countries, and blanket network coverage of urban areas is now the norm.

In South Africa, we've gone from zero coverage to saturation in about 20 years, though real growth happened relatively late. The US and Europe have had extensive networks longer. Given a population so consistently exposed to cellphone radiation, some correlation with health issues could be expected. Air pollution, for example, correlates closely with incidents of cardiopulmonary disease. But there has not been a sudden increase in rates of brain tumours or other medical conditions which could be attributed to cellphone usage.

It must be noted that because cancer can take many years to develop, that unimpressive epidemiology doesn't mean there's no effect at all - it may simply be taking longer to manifest. But the statistics, like the medical research, are reassuringly negative for now.

If there is one area where the detractors have got the issues badly, dangerously wrong, it is the question of cellular base stations.

Where studies into the effects of cellphone radiation have shown biological effects (which may or may not prove harmful), studies into cellular base stations have shown exactly nothing. There is no data which suggests any risk whatsoever.

The physics involved explain why, even in simplified terms. A base station transmitter usually pumps out less than 100W. That energy attenuates exponentially, in accordance with the inverse square law, which means that by the time it reaches you, the radiation in question is considerably less than your cellphone is emitting.

In other words, worrying about the mast, when you are surrounded by people with cellphones, is focusing on entirely the wrong source of radiation.

But the real mistake lies in misunderstanding how cell networks operate. Cellphones can push out a signal up to 2W, and will do so if their connection to the base station is weak. In areas of good coverage, phones reduce their output, mostly to save battery life.

In other words, anyone campaigning to reduce the number of base stations is effectively campaigning to increase the radiation from phones, the only possible source of radiation which might plausibly be dangerous. Anyone who wants to reduce cellphone radiation emissions should be campaigning for more base stations, not less. The use of femtocells, placing low-power transmitters closer to the phone, would minimise the amount of radiation to which users will be exposed, not increase it.

In South Africa, the Department of Health has stated that it will follow the World Health Organisation and International Commission on Non-Ionising Radiation Protection (ICNIRP) guidelines regarding the lack of health risks inherent in base stations.

Health organisations, governments and other public bodies have been reticent to come out on either side of the debate. Governments, of course, are responsible for looking after their citizens' telecom needs, industrial competitiveness as well as their health, so there is an obvious dilemma when these areas may be at conflict.

The lack of a concrete position leads to further conflict: there are minimum standards for safety parameters of electronic equipment, but critics argue these are insufficient to protect them from harmful radiation, while manufacturers can simply fall back on a blanket statement that they are complying with the standards.

Safety standards are based on science, though, and science has yet to find any evidence of harm, as I've discussed.

One public body whose stance is frequently misunderstood is the World Health Organisation (WHO), and IARC (the International Agency for Research on Cancer). In 2011, before the Interphone results were published, IARC announced “IARC classifies radiofrequency electromagnetic fields as possibly carcinogenic to humans”.

The critics seized this statement as proof of a link, but that's an exaggeration, stemming from misunderstandings of IARC's classifications. IARC classifies “radiofrequency electromagnetic fields, such as those associated with wireless phones” as on a list known as Group 2B (“possibly carcinogenic agents”). That list is for substances which are not provably carcinogenic. Lead is on that list, and so are titanium dioxide (the pigment in sunscreen), car exhausts, and coffee.

Higher than Group 2B are Group 1 (“definite carcinogens”) and Group 2A (“probably carcinogenic”). But more telling is a brief look at the levels below 2B: Group 3 (“not classifiable as carcinogenic”) and Group 4 (“probably not carcinogenic to humans”).

The categorisation doesn't include the amount of exposure. Group 1, for instance, includes oral contraceptives and UV radiation (such as that in sunlight). 2A includes glass manufacture and frying food at high temperature.

Only one substance in the whole world (caprolactam) has ever been placed in Group 4. And there is no Group 5, “definitely not carcinogens” (because you can't prove a negative, remember?)

Two lingering issues remain. One question revolves around the stories of ill effects from masts (or cellphones for that matter), and the other concerns the nocebo effect.

There is a saying in scientific circles: “the plural of anecdote is not data”. In other words, stories in isolation prove nothing, because they could be attributed to anything, and usually cannot be used to support conclusions. One scorched tree near a cellphone tower could have been burned by any number of things. If many trees near cellphone towers in different places (ie, subject to different environmental conditions) start to show similar symptoms, scientists would be more excited.

Some mysteries are just the results of poor communications. I'm told cellular operators often hear from frightened people who have noticed a circle of dead plants around a base station. And plants under cellphone towers do indeed frequently die back, but not because of radiation (which is weakest directly under the directional antennae used in masts anyway), but because the constructors use pesticide to clear the undergrowth.

Scientific studies are laborious and complicated precisely because they have to eliminate unrelated factors and identify root causes. And the fact is that despite multiple studies, no one has ever managed to prove sensitivity to EM radiation from a cell transmitter.

The WHO is blunt about “electromagnetic hypersensitivity”, noting that other factors are more likely to be causing distress, such as air quality or flickering fluorescent tubes: “EHS is not a medical diagnosis, nor is it clear that it represents a single medical problem. Treatment of affected individuals should focus on the health symptoms and the clinical picture, and not on the person's perceived need for reducing or eliminating EMF in the workplace or home.”

The nocebo effect is more interesting. Just as fake medicine can make someone feel better (the placebo effect), a non-functioning agent can make someone feel bad - both effects are deeply fascinating to medical scientists. Put a tower with a blinking light on it in a suburb, and some people genuinely might feel headachy or ill, even if the tower has no transmitter on it at all. (This has, I'm told, really happened in South Africa.)

This is a particular challenge for all concerned, because while the sufferers might be mistaken about the cause, they really are feeling ill and their quality of life may have suffered, and their complaints cannot be dismissed out of hand. Medicine is unlikely to help, since the cause is primarily psychosomatic.

Educating people about the real issues is probably the only way to help counter the nocebo effect, but scientists find competing with a vocal community of critics on the side of the nocebo to be a real challenge. Cellphones are not unique in this regard - witness how long the campaigns against MMR vaccination took to die down, even after the erroneous claims of health risks were publicly retracted and studies continued to demonstrate safety, even in the face of dramatically increased infections, and even deaths due to lack of vaccination.

For completeness, I must acknowledge the possibility that the cellular telecoms industry is one giant conspiracy. The vendors and network operators may indeed be in the possession of research, which proves the health risk of their products, and they may have bribed officials around the world to keep their findings secret, to perpetuate their profits.

Don't laugh that off too quickly - that would have accurately described the tobacco industry in the 1950s.

But the volume of research conducted outside of the industry, by universities, public sector bodies, the WHO and others, suggests this theory is highly unlikely.

The overwhelming lack of evidence suggests the same. Despite saturation penetration of cellphones, the rate of cancer just hasn't spiked accordingly.

One thing is for sure: if any conclusive evidence was found proving that cellphones are a health risk, you'd know. The uproar in the telecoms and medical communities would dominate headlines for months. You just wouldn't be reading them on your phone.