Ben Goldacre
Saturday May 13, 2006
The Guardian
“Electromagnetic fields stemming from gadgets such as kettles, computers and microwaves, contribute towards a cloud of unseen emissions – even when they are switched off.” It’s a sinister idea, and “Electromagnetic Hypersensitivity” is sweeping the nation, or at least the Independent and the Daily Mail last week. Symptoms include fatigue, tiredness, headaches, concentration difficulties, dizziness, nausea, heart palpitations, and digestive disturbances: and since these are real symptoms, causing genuine distress, the problem deserves to be considered seriously, and carefully.
There are two things the newspapers left out of their stories: the first is the scientific evidence, and the second is a very crucial nuance. We’ll deal with the evidence first, because science is “hard”. There have been 31 studies looking at whether people who report being hypersensitive to electromagnetic fields can detect their presence, or if their symptoms are worsened by them. A typical experiment would involve a mobile phone that was hidden in a bag, for example, with each subject reporting their symptoms without knowing if the phone was on or off.
31 is a good number of studies, and 24 found that electromagnetic fields have no effect. But 7 did find some measurable effect, and because I have a reputation for pedantry to uphold: in 2 of those studies with positive findings, even the original authors have been unable to replicate the results; for the next 3, the results seem to be statistical artefacts (details below); and for the final 2, the positive results are mutually inconsistent (one shows improved mood with provocation, and the other shows worsened mood). It’s a slightly obsessive form of value for money you get here, but it’s value nonetheless.
Meanwhile the “nuance” here is in whether recognising distress is the same as acknowledging causality. It’s easiest to see with examples. According to the Daily Mail, Health Protection Agency spokesman Dr Jill Meara “said household appliances could be health hazards to some. She advised those with a condition called electrical sensitivity to stay away from such devices such as kettles and microwaves.†As far as I can see, what she actually said was: “There is no proof that the symptoms are caused by exposure to electromagnetic fields, but people still have real symptoms. There is sufficient uncertainty that it is worth telling members of the public practical things they can do to reduce exposure in case they feel they are suffering from these fields.â€
The Mail go on. “The WHO, meanwhile, has described electrical sensitivity as ‘one of the most common and fastest growing environmental influences’. It ‘takes seriously’ concerns about the health effects, adding that ‘everyone in the world’ is exposed to the emissions, while ‘levels will continue to increase as technology advances’.†And yet, in the very same report from which they are quoting, WHO make it clear that there is no scientific evidence to link the symptoms to electrical fields. In fact they say this: “EHS has no clear diagnostic criteria and there is no scientific basis to link EHS symptoms to EMF exposure. Whatever its cause, EHS can be a disabling problem for the affected individual.”
This is a frightening area for any compassionate person to write in. Of those who play to the evidence, the the Today program on Radio 4 says this: “sceptics fear they are preying on the fears of people who are ill and desperate, another manifestation, they say, of the hypochondriac society.”
That’s harsh. I don’t think people who report being hypersensitive to electrical fields are hypochondriacs: I think they have real and distressing symptoms, but I also think, in the light of the evidence above, that electromagnetic fields probably aren’t causing those symptoms, and they may remain unexplained for the moment.
But explanations are what everyone wants, and not just for reassurance and meaning: in the age of science, it’s as if our culture demands that we have a biomedical explanation to justify the very existence our symptoms and distress, otherwise they’re simply not valid. If there is no real explanation, then perhaps a “placebo” explanation – like “electromagnetic hypersensitivity” – can have almost all the same properties as a real one.
Statistics:
The 31 studies looking at whether electromagnetic fields can have effects on sensitive individuals are examined in this excellent review paper from 2005:
iddd.de/umtsno/emfkrebs/rubin2005emf.pdf
Before we look at the apparent positive results in 3 of the 31 papers, it’s important to remember that none of this is to “trash” those papers, or ridicule their authors: they are not in the same category as a Gillian McKeith article, and these are technical criticisms of the kind you’d hear at any academic journal club taking a critical look at their work. These points are all also covered in the discussion section of the review paper above.
In one case there is the question of whether it was appropriate to use a one-tailed significance test. Using a one-tailed significance test makes the assumption that electromagnetic fields can only have an effect in one direction (for example, if you were measuring “mood”, then it would assume that EM fields could only possibly worsen your mood, and not improve it). This, I would argue, is not a valid assumption, with such an illusive and inconsistently measured effect as sensitivity to electromagnetic fields. A two-tailed significance test allows for the fact that the exposure (electromagnetic fields) could have an effect in either direction, and is more appropriate. One-tailed tests are often more likely to give a positive result, which is why it is wise to be cautious and only use them when you are sure the effect you are measuring could only go in one direction, to avoid false positives.
In the other case there is the issue of multiple significance tests being performed. This is a similar problem to the one we considered a few weeks ago here in the discussion on cocaine use doubling (according to the front page of The Times): essentially, the more significance tests you do, the more likely you are to turn up a positive one merely by chance.
And lastly, a personal note: the symptoms that people are currently attributing to electromagnetic hypersensitivity are unpleasant and often disabling, even if the electromagnetic fields are not the cause of them. Calling people who believe they experience symptoms because of electromagnetic fields “hypochondriacs”, weak-minded, or oversensitive: is offensive, unhelpful, and most damnably of all, untrue. Denying the reality of people’s symptoms is similarly offensive and unhelpful. I will give no quarter on this.
EDIT March 2007:
Another 5 studies have come out since that review was published, all negative. They are:
Rubin GJ, Hahn G, Everitt B, Cleare AJ, Wessely S. Are some people sensitive to mobile phone signals? A within-participants, double-blind, randomised provocation study. BMJ 2006; 332:886-889.
Frick U, Kharraz A, Hauser S, Wiegand R, Rehm J, von Kovatsits U et al. Comparison perception of singular transcranial magnetic stimuli by subjectively electrosensitive subjects and general population controls. Bioelectromagnetics 2005; 26:287-298.
Wilen J, Johansson A, Kalezic N, Lyskov E, Sandstrom M. Psychophysiological tests and provocation of subjects with mobile phone related symptoms. Bioelectromagnetics 2006; 23:204-214.
Regel SJ, Negovetic S, Roosli M, Berdinas V, Schuderer J, Huss A et al. UMTS base station-like exposure, well being and cognitive performance. Environ Health Perspect 2006; 114(8):1270-1275.
Wenzel F, Reibenweber J, David E. Cutaneous microcirculation is not altered by a weak 50Hz magnetic field. Biomed Tech (Berl) 2005; 50:14-18.
EDIT July 2007:
A 37th provocation study came out last month:
Oftedal G, Straume A, Johnsson A, Stovner LJ. Mobile phone headache: a double-blind, sham-controlled provocation study. Cephalalgia 2007; 27:447-455.
Negative. Same results as all the others.
Ben Goldacre said,
May 20, 2006 at 1:10 am
thanks coracle some of those look bang on.
igb said,
May 20, 2006 at 8:57 am
There’s a meta-argument against all the campaigners: power lines, mobile phones, pesticides, whatever. And that is that in the urban, developed west, life expectancies continue to rise. Mobile phones have been fixtures for the middle classes for fifteen years, and 900MHz transmitters have been present in cities for more than twenty. Where’s the excess deaths? Powerlines ditto, but for generations. Where’s the excess deaths?
Instead we end up with vague, non-specific, non-reproducible `symptoms’ found by dredging around in the statistical noise floor. It’s People With Problems, externalising their fears. Their non-illnesses are caused by non-mechanisms and can be cured by non-cures (I think there’s a market for homeopathic remedies for electrical smog: royalties to me, please. It contains
liverpoolmiss said,
May 20, 2006 at 9:30 am
Yes, and as I understand it childhood leukaemia rates have fallen steadily over the decades, while the number of electrical appliances etc has soared.
JQH said,
May 20, 2006 at 2:38 pm
igb
I don’t think it’s fair to refer to sufferers “non-illnesses”. I don’t think it’s disputable that they’re ill. What IS disputable is the cause.
Ben Goldacre said,
November 24, 2006 at 7:46 pm
i have never had so much hate mail, it’s still coming, i got bored of posting it a long time ago. this is the single subject that has generated the most vitriol out of anything i have ever written on. i think the new wave is because i got in touch with somebody from one of the electromagnetic hypersensitivity pressure groups, very politely, to ask a quick question. he must have mentioned it to someone, word got round, and another wave of hate comes in. amazing.
The worst case of bad science i have ever come across is you. I worked in child psychiatry when children with asthma were sent to a psychiatric wing. One nearly died so we stopped having them to treat. I know electrosensativity exists, and Professor Olle Johansson proved it and why. You wont print anything by him though, or my letter to you i bet. You are a disgrace, and that is choosing kind words for how despicable you are. Either that or your ignorance or the other connections you mention might be true. Bet you don’t want to get a link to Olle or Professor Magda Haves to find out the truth either.
Catherine Gamba
Flipsidereality » Blog Archive » EM Radiation WIFI Sickenss, Factors that risk being left out of the equation said,
November 29, 2006 at 12:31 am
[…] badscience » Factors that risk being left out of the equation “Electromagnetic fields stemming from gadgets such as kettles, computers and microwaves, contribute towards a cloud of unseen emissions – even when they are switched off.†It’s a sinister idea, and “Electromagnetic Hypersensitivity†is sweeping the nation, or at least the Independent and the Daily Mail last week. Symptoms include fatigue, tiredness, headaches, concentration difficulties, dizziness, nausea, heart palpitations, and digestive disturbances: and since these are real symptoms, causing genuine distress, the problem deserves to be considered seriously, and carefully. […]
Neuromesh » Blog Archive » Wireless Networking said,
November 29, 2006 at 8:59 am
[…] For those who may be feeling these symptoms and are now concerned that they may be more than just November burnout, rest assured that there is no reliable study linking anything to Electro-Magnetic radiation. Longtime readers of Neuromesh may also recall however that your electric blanket will kill you before any of your gadgets do. […]
KCC Children, Families and Education - Digital Curriculum : Is wi-fi in schools a risk to health? said,
December 2, 2006 at 9:43 am
[…] Is wi-fi in schools a risk to health? An article appeared in the Times newspaper that questioned the safety of wi-fi in schools. In it, two schools are mentioned in particular. Online news agencies have repeated and created froth and fervour, giving greater creedence, but adding nothing new.When I first read it, I was of course aghast! Could it be that one of the central planks of anytime anywhere learning is causing health problems? I instinctively know it isn’t, and have worked with high power fields in my telecomms career, so know the health and safety issues (and my friends aren’t dropping with cancer!). I did what my background taught me to do, … some private research to prove one way or the other! In the cold light of scrutiny, the stories appear to owe rather more to the small ‘p’ political considerations of good relations within a school community than a risk assessment of health issues. There is clearly a fundamental problem with understanding the basic science, so i’ll summarise the facts:Don’t get confused between electric fields and radio frequency fields. Try these flash based tutorials from the Health Protection Agency to understand why.If you are considering banning wi-fi on safety grounds, then out must also go public broadcast radio, terrestrial TV, police fire and ambulance systems, mobile telephones, and wireless microphones, as these sources are higher powered.Those who do not understand basic science should not make decisions that affect the rest of us based on their lack of knowledge. Radio frequency waves as cancer-causing magic moonbeams is more to do with fear and ignorance than fact.Radio frequency transmission is a well understood technology.Guidelines already exist, and are robust. The only concern they have is for heat generated by close proximity – not a concern at the very low power levels of wi-fi. The scientific research by the EU, and the guidance by the world health organisation is clear that the perception of a condition called electromagnetic hypersensitivity exists is more to do with perception than reality, as in blind tests, no afflicted individual could tell whether the test transmitter was on or off, and reported the symptoms in both cases, mmm!The world is full of radio waves from BBC stations to mobile phones, and at much higher levels than school wi-fi, which is restricted to 10 milliwatts, which is 5% of the power of a single mobile telephone, and this restriction is only to stop it interfering with other peoples wi-fi networks, not on the grounds of safety!To bow to irrational pressure may be expedient, (and may save some money!), but let’s not confuse this with science. I’m not ready yet to give up all of the benefits on the strength of zero evidence, and my whole family bathes in the warm glow of a home wi-fi network, on visits to McDonalds and Starbucks. Indeed the whole of Westminster is bathed in wi-fi, and there are no signs of poor health amongst our politicians (tee hee).There is more evidence that poor nutrition has a detrimental and long-lasting effect on our children, yet parents still pass bags of chips through railings. All a matter of perception really. To yield to the palpitations and twitches of a single classics teacher when confronted by teenagers in a classroom (doesn’t have the symptoms at the weekend or during school breaks!) is absurd. The symptoms are probably real, the attributed cause isn’t! Published 02 December 2006 08:32 by AlanDay […]
Wifi health risk paranoia (at wongaBlog) said,
December 14, 2006 at 12:19 pm
[…] Others claim they immediately started having headaches after a wireless network was enabled. Every time this is mentioned it’s difficult not to notice that knowledge of the wireless network came before the headaches, never the other way around. It’s very much like people who claim to suffer from ‘electrical sensitivity’, but can’t replicate it under controlled conditions. It all has an air of woo that has apparently turned up with with the introduction of many technologies, such as the original radio transmissions, or ozone from laser printers, or radiation from the first computers. It seems that once these things become commonplace the symptoms disappear. […]
Freedom for Information » Blog Archive » Wi-Fi - A risk to health? said,
January 28, 2007 at 8:29 pm
[…] To bow to irrational pressure may be expedient, (and may save some money!), but let’s not confuse this with science. I’m not ready yet to give up all of the benefits on the strength of zero evidence, and my whole family bathes in the warm glow of a home wi-fi network. My children are far more likely to experience health problems through consuming menu items on visits to McDonalds and Starbucks than through their wi-fi networks. Indeed the whole of Westminster is bathed in wi-fi, and there are no signs of poor health amongst our politicians (Umm, tee hee). […]
How To Spot A Psychopath :: How about “electronic Viagra”? :: April :: 2007 said,
April 24, 2007 at 3:34 am
[…] (Ben Goldacre’s been catching abuse for a year or so now as a result of his clearly inhuman and evil view that people who believe electromagnetic “pollution” is making them ill have real symptoms, but are incorrect about the cause. For some reason, it’s hard to find similar pressure groups who believe that people who hear voices should be provided with earplugs.) […]
Harnessing Technology in Kent - Digital Curriculum : Is wi-fi in schools a risk to health? said,
April 24, 2007 at 1:25 pm
[…] Is wi-fi in schools a risk to health? An article appeared in the Times newspaper that questioned the safety of wi-fi in schools. In it, two schools are mentioned in particular. Online news agencies with their cut and past culture have repeated and created froth and fervour, giving greater creedence, but adding nothing new.When I first read it, I was of course aghast! Could it be that one of the central planks of anytime anywhere learning is causing health problems? I instinctively know it isn’t, and have worked with high power fields in my telecomms career, so know the health and safety issues (and my friends aren’t dropping with cancer!). I did what my background taught me to do, … some private research to prove one way or the other! In the cold light of scrutiny, the stories appear to owe rather more to the small ‘p’ political considerations of good relations within a school community than a risk assessment of health issues. There is clearly a fundamental problem with understanding the science, so i’ll summarise the facts: Don’t get confused between electric fields and radio frequency fields. Try these flash based tutorials from the Health Protection Agency to understand why. If you are considering banning wi-fi on safety grounds, then out must also go public broadcast radio, terrestrial TV, police fire and ambulance systems, mobile telephones, and wireless microphones, as these sources are higher powered. Probably better to put tin foil on the inside of the buildings and shield it completely! Radio frequency waves as cancer-causing magic moonbeams is more to do with fear and ignorance than fact. Radio frequency transmission is a well understood technology. Guidelines already exist, and are robust. The only concern they have is for heat generated by close proximity – not a concern at the very low power levels of wi-fi.Scientific research by the EU, and guidance by the world health organisation is clear that electromagnetic hypersensitivity is more perception than reality, as in blind tests, no afflicted individual could tell whether the test transmitter was on or off, and reported the symptoms in both cases, mmm! Giving it a name doesn’t make it real!The world is full of radio waves from BBC stations to mobile phones, and at much higher levels than school wi-fi, which is restricted to 10 milliwatts, which is 5% of the power of a single mobile telephone, and this restriction is only to stop it interfering with other peoples wi-fi networks, not on the grounds of safety!To bow to irrational pressure may be expedient, (and may save some money!), but let’s not confuse this with science. I’m not ready yet to give up all of the benefits on the strength of zero evidence, and my whole family bathes in the warm glow of a home wi-fi network. Indeed the whole of Westminster is bathed in wi-fi, and there are no signs of poor health amongst our politicians (Umm, tee hee).There is more evidence that poor nutrition has a detrimental and long-lasting effect on our children, yet parents still pass bags of chips through railings. All a matter of perception really. To yield to the palpitations and twitches of a single classics teacher when confronted by teenagers in a classroom (doesn’t have the symptoms at the weekend or during school breaks!) is absurd. The symptoms are probably real, the attributed cause isn’t! Published Saturday, December 02, 2006 8:32 AM by AlanDay […]
B0ll0cks - FreddysHouse said,
May 21, 2007 at 10:52 pm
[…] Take a look at this Jup: Bad Science » Factors that risk being left out of the equation. Personally I think this wireless headache mojo is generally a crock of shit. We’ve just deployed a wireless network throughout the college I work at so it’d be interesting to see how that affects people if at all though __________________ <Insert witty comment here> […]
Beware of radiators « A little Jack with that? said,
May 22, 2007 at 11:59 am
[…] Tuesday, May 22nd, 2007 in Technology After last night’s Panorama propaganda about the evils of WiFi, I was going to write something expressing my outrage as a former BBC employee over the appalling lack of editorial standards that allowed such a piece of unbalanced, biased, unscientific nonsense to be presented as an exercise in rigorous journalism. However, it doesn’t really need me to pile on. I will leave you to read one of the many posts on the subject taking the BBC and the report to task. Or just read some of the comments on the BBC’s own Have Your Say site, which are by and large much more sensible and informed than the report. […]
Wifi e preti pedofili « Dema FON blog said,
May 23, 2007 at 3:01 pm
[…] Ora , chiunque abbia seguito almeno in parte i dibattimenti sulle conseguenze nell’organismo delle microonde wireless , si rende conto che sono cazzate , e se non se ne rende conto e vuole documentarsi , puo’ farlo qui , qui oppure qui (english) […]
WiFi is eating your children at starshun.com said,
May 29, 2007 at 1:52 pm
[…] One of my favourite programs Panorama has broadcast an ‘expose’ on the dangers of wifi.Unfortunately the show has no science whatsoever behind and in fact 31 studies have shown no link between symptoms and electrical sensitivity, none of these are referred to in the Panorama program. It’s incredibly disappointing when a program that you otherwise respects does such incredibly bad science. […]
blog.tjomlid.com | Bli kvitt tinnitus på 10 timer said,
October 26, 2007 at 9:33 pm
[…] om han merker så mye forskjell når han ikke vet at det er et WLAN i nærheten: There have been 31 studies looking at whether people who report being hypersensitive to […]
How To Spot A Psychopath :: My (irradiated) balls are always bouncing :: March :: 2009 said,
March 11, 2009 at 6:12 am
[…] and over, alleged “electrosensitives” have failed to demonstrate that they can even perceive […]
“Dirty Electricity” « The Winnipeg Skeptics said,
September 14, 2010 at 10:07 pm
[…] Award-winning Journalist and Medical Doctor Ben Goldacre: Cherry Picking and the Professional Association of Teachers Why don’t journalists mention the data? Factors that risk being left out of the equation […]
Wi-Fi fear mongering enters the Lower Mainland said,
March 7, 2011 at 11:08 pm
[…] 37 times. […]
How To Put On Lean Muscle (Smart Training Solutions) said,
April 18, 2011 at 10:20 am
[…] it up and many studies to prove that it is, in fact, horseshit. Just to prove I can use Google too. Factors that risk being left out of the equation – Bad Science Reply With Quote + Reply to […]
Mer el-allergi i Dagbladet « Skepsisbloggen said,
August 26, 2012 at 11:20 am
[…] om at den typen provokasjonsstudier som definitivt er relevante for saker som i oppslaget nok så tydelig antyder at det ikke er stråling som utløser […]
Harmoniser bølgene dine for kr 4700 « Skepsisbloggen said,
August 28, 2012 at 6:21 pm
[…] forskningen så langt bekrefter bekymringene, og i dobbeltblindede forsøk er de el-overfølsomme ute av stand til å si om det er strøm eller ikke på apparatene de mener de blir dårlige […]