Nov 23 2007

Wireless Technology and Autism

A recent press release reporting on an alleged association between the electromagnetic radiation (EMR) from wireless devices and autism has sparked a round of credulous news reporting, mainly in computer and technology magazines and websites. This article from is typical. The article begins:

Autism is a disabling neuro-developmental disorder. Its cause is not completely understood, but it is linked to heavy-metal toxicity.

This is a very misleading statement, and reveals both the biased nature of the study authors and the lack of basic journalism on the part of John-Paul Kamath, who wrote the article for Computer Weekly. He would have done well to speak for a few minutes to an actual autism expert. Autism has, in fact, not been linked to heavy-metal toxicity. That is a fiction perpectuated by the anti-vaccination cranks and the mercury militia – groups still clinging to the thoroughly discredited notion that mercury in vaccines is causing autism. Also, while saying that the cause of autism is “not completely understood” is technically accurate (this statement would be accurate about most things in biology and medicine) it omits the fact that autism is most strongly linked to a variety of genetic causes.

The article also states:

The authors says that the rise in cases of autism is paralleled by the huge growth in mobile phone and Wi-Fi usage since the late 1990s with worldwide wireless usage now having reached nearly 4 billion people.

Kamath has committed here a journalistic sin that is increasingly dangerous in a world where pseudoscience, politically or ideologically motivated science, biased science, and plain crank science has seamlessly infiltrated mainstream legitimate science. Namely, he failed to perform basic fact checking. Again, a moment with an autism expert or two would have been very useful. He may have learned that the consensus opinion is that autism rates are in fact not increasing at all – rather, the diagnosis of autism has increased as methods for surveillance have increased and the diagnosis was broadened.

The statement itself is also an example of the confusing correlation with causation logical fallacy, assuming that A causes B because they occur together. The premise in this case is false (true autism rates are likely not increasing) and the logic is not valid. Essentially autism can be linked by this logic to anything that was increasing over the last 15-20 years. This is the fundamental claim of those who believe that thimerosal in vaccines causes autism, because the number of childhood vaccines was increasing during this time. But perhaps the internet is causing autism, or computer use, or SUV’s. Name any trend over the last decade and a half and you have a correlation with the apparent rise in autism.

Most articles reporting on this new study do contain the customary counterpoint from an expert saying that Wi-Fi levels are below safety levels and pose no threat. This generic dismissal, however, is very ineffective at countering the specific false claims made by the study authors. Including a generic counterpoint for “balance” is now what substitutes for actual journalism in most modern science reporting.

The study itself was done by Dr. George Carlo and published in the Australasian Journal of Clinical Environmental Medicine. Although the press release claims the journal is peer-reviewed, this journal is not listed in the National Library of Medicine nor is it represented in search engines of peer-reviewed journals, such as PubMed. The fact that it is an obscure journal without official recognition does not mean that anything it publishes is necessarily flawed or wrong, but it does matter. The better a study the more likely it is to be published in a respected journal. Obscure “throw-away” journals tend to be obscure for a reason – they may not be peer-reviewed (or adequately so), and often they represent the biases of the organization that publishes them.

Apparently, the Australasian Journal of Clinical Environmental Medicine is published by the Australasian College of Nutritional and Environmental Medicine. The president of this organization, “Dr” Gary Deed, is a practitioner of “integrative” medicine. According to his bio his only medical training is in the use of medicinal herbs. So the journal and the group seems to be both obscure and fringe.

Dr. George Carlo is a legitimate researcher, and has been involved with research into the health effects of cell phones for years. He is at one end of the spectrum of this debate, however, claiming that the evidence supports the conclusion that cell phone EMR is dangerous. He has formed an organization, The Science and Public Policy Institute, and wrote the book, Cell Phones: Invisible Hazards in the Wireless Age. I have no problem with the fact that Dr. Carlo has a minority opinion and that he is advocating actively for his point of view. But his level of activism and alarmism should be taken into consideration when evaluating his own research that apparently supports his position. (Here is a radio interview with Dr. Carlo that will give you a good idea of his approach to this controversy.)

What about the study itself? I could not find any online version of the study (as I said, it is not listed in the online databases), and it is not published on the journal website itself. So I can’t review the methods used. The press release states:

Dr Carlo said, “These findings tie in with other studies showing adverse cell-membrane responses and disruptions of normal cell physiology. The electromagnetic radiation apparently causes the metals to be trapped in cells, slowing clearance and accelerating the onset of symptoms.”

The mechanism that Dr. Carlo is citing is dependent upon the premise that heavy metal poisoning causes autism, which has not been established and is probably not true.

So what we have is a dubious study published in an obscure journal by an apparently ideological organization and supporting the fringe position of its lead researcher. The press release if full of misinformation and questionable premises, and was appropriately ignored by the scientific community. It was, however, widely and mostly credulously reported by tech news outlets, without proper skepticism or even basic journalism. Such reporting, unfortunately, creates the meme that EMR causes autism, and no amount of later refutation is likely to eradicate it.


Thanks to Mikael Pettersson for sending me the pdf of this study. The study is far worse than I even imagined – a mishmash of pseudoscientific nonsense. The study was an uncontrolled observation of 20 subjects with autism who were treated for assumed heavy metal toxicity in an EMR free environment. Hair, urine, and feces analysis was used to measure heavy metal excretion. According to the authors, the results show a significant trend toward increased excretion toward the end of the cycle of forty treatments. They claim this suggests that the EMR free environment allowed for the heavy metals to be secreted more efficiently, and that is why they excreted more toward the end of the treatments.

The fact that the study was not blinded or controlled in any way completely invalidates the results, especially since the study authors used multiple dubious and pseudoscientific methods in the study. Also, they first looked for a trend in their “sentinel case” and then assumed that whatever trend they observed indicated the outcome they desired. So in essence they peaked at the results first, and then declared those results a positive outcome. They did not compare the outcome to non-autistic subjects, or to subjects not in an EMR free environment. We have no idea that the trend (assuming it is real, which is not an assumption earned by the study authors) means anything. Yet the authors conclude from this trend that: heavy metal poisoning causes autism, treating heavy metal poisoning helps autism, EMR interferes with the clearance of heavy metals, and unicorns fart rainbows (OK, I made up that last one).

Let’s look at their methods.

Subjects were given intervention in a sequential protocol that included a series of non-chelation provocations and nutritional formularies focused on mitochondrial resuscitation depending on the clinical profile of the client. Two general categories of subjects were defined for clinical purposes: those with liver clearance as an indicated vulnerability and those with kidney function weakness. These determinations are critical for precision in intervention for each subject and were based on a priori laboratory analyses, acupuncture meridian tests, medical history, consultations with subject’s parents and clinician observations.

“Mitochondrial resuscitation.” That sounds really scientific. Unfortunately, they did not describe in detail how the acupuncture meridian tests determined which subjects had problems with liver clearance and which ones had problems with kidney clearance.

In order to make an EMR free environment we are informed that:

Applications of body worn sympathetic resonance technology, energy resonance technology and molecular resonance effect technology were introduced as appropriate.

They really spared no pseudoscientific expense in this study. Sympathetic resonance technology – I confess I had no idea what that was, but I found in a study published in the Journal of Alternative and Complementary Medicine that:

By interacting with key component frequencies in the biofield, SRT may stabilize the organism homeodynamically, thereby protecting it from the effects of stressful stimuli.

Well that explains it – SRT is homeopathic energy medicine magic.

Should we trust the unblinded clinical assessments used in this study? Of the sentinal case, the authors write:

Prior to presentation at the clinic, he had been chelated, virally provocated, detoxed with far-infrared sauna therapy, been given Secretin and IVIG, but still had made only modest progress with his symptoms.

This poor kid has been put through the full spectrum of pseudoscientific treatments for autism, and yet had only made “modest progress” – i.e. he was developing along his own curve, unaltered by the ministrations of charlatans.

There’s more, but you get the idea. What we are witnessing is a high level of pseudoscience – all the trapping and jargon of real science; they are going through the motions and reporting their results with a straight face. But it is all utter nonsense – a house of cards with speculation piled on top of fantasy built on top of pseudoscience.

11 responses so far

11 thoughts on “Wireless Technology and Autism”

  1. gaz says:

    Oh no! “Adverse cell-membrane responses and disruption of normal cell physiology”?! Something that general must be really serious!

  2. Stephen_Bain says:

    Ah, it is clear here that the author got his logical fallacy wrong.

    What he meant to say was that the rise in autism has caused us to use more wireless devices.

  3. Niobe says:

    Or that the rise of the Internet lead to more autism self-diagnosis.
    From google:
    Results 1 – 10 of about 4,620,000 for asperger

  4. DLC says:

    Thanks for writing about this. In my email I used the words “craptacular” and that it “Redlined my BS Detector”.
    I spent about an hour looking for the study online and came up with nothing. I wish more science writers in the popular media had better BS detectors.

  5. hagus says:

    If I recall correctly, aren’t the autism rates highest in places like Seattle and the Bay Area?

    The theory is that you have already selected for autism in these communities by having them filled with engineers and scientists who already exhibit autistic- or aspergers-like traits (ability to focus on tasks, tendency to be introverted and less socially apt, high intelligence in general).

    These people then interbreed and the result is apparently a skyrocketing austism and aspergers rate.

    Or is this another case of confusing cause and effect?

  6. Scepticon says:

    Reading your excerpts from this study made me laugh, thanks.

    I had a kind of “jinx” moment when I read the mitochondrial resuscitation part and then your comment,

  7. EmilyB says:

    Dr. Novella,
    Have you addressed the theory of low glutathione and autism?
    I hadn’t heard anything about this but Jenny McCarthy has been mentioning it in recent interviews, so I’m sure the theory will become more widespread. Is there any possible grain of truth to this, or just more bad science?

  8. JoH says:

    It’s SO frustrating to witness how quickly these things are accepted uncritically and roam the world in the blink of an eye… A couple of days ago this report appeared in Belgium’s biggest newspaper(not really known for its in-depth reporting…), and after reading just a couple of lines I had a strong suspicion already that we would hear more about it on this Blog. My BS detector has improved considerably over the last couple of months with Neurologica, lol…

  9. mattdick says:

    The statement itself is also an example of the confusing correlation with causation logical fallacy, assuming that A causes B because they occur together.

    The author quoted the right stats, but definitely got causation wrong. The obvious conclusion is that autism causes increased cell phone use.

  10. daedalus2u says:

    In that it was likely people with Asperger’s who designed the chips, computers, and electronic networks that made cell phones possible, that may well be true.

  11. belair says:

    I’m concerned like most of us with cell phone radiation, naturally the telecommunications industry has been claiming that cell phone usage is safe, but with the WHO and International Agency for Research last year 2011 that there were risks for glioma, a malignant type of brain cancer..and also as parents concern for our children’s since there heads absorbs up to double the energy that of a large adult..upon many researches for emf, I found the Matrix, a new product using military stealth technology successfully blocking more than 85% of emf… if this could help out go for it..

Leave a Reply