Mobile Phone Study Released and Reported Very Very Badly
May 17th, 2010
|
| Share |
Tomorrow a major international study on cell phone use and cancer will be published in the International Journal of Epidemiology. The study, known as the “Interphone Study” was sponsored by the World Health Organization’s International Agency for Research on Cancer. Despite the study not being published in its entirely, some of the data has already been reported.
Based on the headlines, you might think that the study proved a link between cell phones and cancer. You might also think it didn’t. You might think that it’s inconclusive. Here are just a few of the headlines from various news sources:
The Scotsman: “Study links mobile phone use to brain tumours”
CP24 (Toronto News): “More than 30 mins of cell use/day increases cancer risk: study”
The Australian: “Industry study shows brain tumour link to heavy mobile phone usage”
Sydney Morning Herald: “Mobile phone-cancer link possible: study”
Wire Update: “INTERPHONE finds no increased risk of brain cancer from mobile phone use”
AFP: “Study finds no brain cancer link to mobile phone use”
BBC News : “WHO study on mobile phone cancer risk ‘inconclusive‘”
CTV News: “Link between cellphones, brain cancer ‘inconclusive’”
ABC News Australia: “Calls for more phone cancer research”
CBS News: “Cell Phone-Brain Cancer Link Deemed Inconclusive”
Independent Online: “WHO: No clear answer on cellphones and cancer”
CourierMail: “Brain cancer link to mobile phones”
Boston Herald: “No proof of cell phone, cancer tie”
Time Magazine: “Cell Phones and Cancer: a Study’s Muddled Findings”
Montreal Gazette: “Cellphone users shouldn’t relax yet: Cancer study”
BELIEVE IT OR NOT, ALL THESE HEADLINES REFER TO THE SAME STUDY!
So what did the study really find?
The study tracked more than 13,000 individuals over the course of ten years and compared phone usage to incidence of cancer in the brain and head. The results were that absolutely no increase in risk of any kind was found between phone usage and any kind of brain cancer. None. Zip. Nada. Zilch. Despite this, researchers (as usual) stopped short of calling their study conclusive and conceded that it was possible that there was an increase in risk that went undetected. Others pointed out that the study shouldn’t be considered the definitive word on cell phones and cancer because of the fact that it did not look at children and that there were some forms of cancer that were beyond the scope of the study.
The portion of the study data that has generated talk of a link between heavy usage of cell phones and cancer is some of the interview data collected as part of the study. This data is the most unreliable and subjective and thus the authors of the study used very tempered words when they described it as having “suggestions” of an association between heavy phone usage and glioma, a type of brain cancer.
The study found no increase in giloma among heavy users of cell phones. What it did find was that, when interviewed, those who were diagnosed with giloma were more likely to say that they were prone to holding the phone on the side of their head where the tumor was located. The trend was small but statistically significant. Of course, this is completely subjective and was even noted by the authors of the study noted “biases and error prevent a causal interpretation.”
That small piece of interview data is the entire basis for the study being stated as showing a risk between cell phone usage and cancer.
Now, I ask you. What side of the head do you usually hold your cell phone on? If you’re right handed, you might, at first, be prone to saying the right side. However, I’ve noticed that when I am making an outgoing call, I often hold the phone in my left hand and use my dominant right hand to dial the number, then put it up to my left ear. Also, I often am doing something else with my right hand. Then again, when I answer the phone, I do think I hold it in my right hand a lot.
The point being: off the top of my head, I can’t think of what side of my head I usually put my phone to and I bet many other people can’t either. Just throw in some selective memory and confirmation bias and you’d fully expect cancer sufferers to remember their phone usually being on the side of their head that has the tumor.
The study is officially “inconclusive” – that is to say, that it found no evidence of phone usage being linked to brain cancer – eventhough some otherwise reputable news sources have falsely reported that it did, apparently having some confusion over the side of the head interviews. The authors of the study, however, pointed out that the data is not compelling enough, entirely on its own, to put the issue to bed.
On this I have to strongly disagree. While scientists tend to be purists when it comes to levels of certainty (which are never really 100.00000000000000000000000%), this seems to be lost on most of the public and media when it comes to these kind of issues. When this study is added to the mountains of scientific data we’ve already accumulated over the past 70+ years, it only serves to confirm the well established fact that RF radiation does not cause cancer. As a matter of public health, policy and precautions, this issue is as certain as we can ever expect any issue to be.
I was going to wait until I could read the study to comment on it, but the amount of bad reporting on this is staggering!
This entry was posted on Monday, May 17th, 2010 at 8:18 pm and is filed under Bad Science, Enviornment, Quackery, inverse square. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.
View blog reactions
May 17th, 2010 at 9:28 pm
Yes I was livid about this when I read it this morning. Elsewhere one of the principles, who is from L’Université de Montréal, was saying that the researchers were very disappointed that ethical standards in force at the moment, and privacy laws, prevented them from developing full medical profiles of the subjects, rendering much of the study useless.
Quote Comment
May 17th, 2010 at 10:19 pm
In a related story, Dr. Steve Buzz0, an unrepentent skeptic who is not actually licensed to practice medicine, had this to say in his blog, which he often uses to attack positions that do not agree with the mainstream medical establishment,
It is unclear how he reconciles this with the increase in cases of childhood leukemia near nuclear power plants. Though never proven, Dr. Buzz0 is suspected of receiving payments from nuclear and medical lobbies to promote their agendas.
Look, Ma! I can be a reporter, too!
Quote Comment
May 17th, 2010 at 10:24 pm
Shafe said:
Do not feed the idiot trolls.
Quote Comment
May 17th, 2010 at 11:09 pm
Shafe said:
Damn straight.
Shafe said:
Once again… the reason I have that name is that when I started this blog I wasn’t totally sure how anonymous or public I wanted to be. Eventually I decided to just say who I was since I’d expect to be public if I did well enough anyway.
“Dr Buzz0″ is just my internet handle that I used since 1993 for various forums, messengers, services etc. I always used the same name for interruptibility – it made it easy for people to find me since it was the same name on all services.
It’s not meant to imply I’m a doctor. It’s one of those silly titles like “Major Pain” or “Captain Amazing” or “Reverend Irreverent”
I very clearly state that I’m not a doctor. VERY clear.
I just want to mention this because I don’t want to be accused of disrespecting the title, which obviously is a meaningful one for those with an MD or PhD or whatever other doctorate.
Shafe said:
You left out “RF” those two letters make a huge difference.
You do understand the difference between ionizing radiation and RF radiation, right? No, you don’t, do you…?
Shafe said:
Oh that one never gets old.
The only ones who suspect that are nitwits who can’t believe that maybe these statements are true.
tell you what, slick, you find any proof I’ve received a single payment of any kind from the nuclear, medical, pharmaceutical, chemical or any other industry for my posts and you can have the money.
Shafe said:
Yes you can. It’s called blogging. Lots of people do it. If you’re half-descent some people might actually read your blog. I would not count on that in your case.
Quote Comment
May 17th, 2010 at 11:15 pm
Reading this, I just realized something. I do not know which hand I usually use. Like buzz, I’m right handed. I just picked up my phone to see how I use it and I realize I hold it in my left hand to dial with my right. However, I also know I do often hold it in my right hand to my ear, when I don’t use a headset. I would think if I am doing something it would probably be the left hand to free up the right, but I don’t know which one is more common for me.
I will have to pay attention to this to figure it out!
I can easily see how a person might not be 100% sure and would be biased by a tumor.
Shafe said:
Not that there is even any proof of that in any systemic review, but whatever the case, I believe you left out the word “RF” in your quote. RF radiation is not the same as nuclear radiation as anyone who knows the first thing about physics knows. “Radiation” can mean the heat coming off a radiator or the light from a flashlight, the gamma rays from a radioactive object or the radio signals from a transmitter. These are totally different.
It’s amazing anyone could be so ignorant as to not even know the meaning of the word and yet still actually comment on this like they know something.
It’s not like the info is not out there.
Quote Comment
May 17th, 2010 at 11:31 pm
Whoa! Slow down guys. I didn’t realize that I had to actually write the words “tongue in cheek” in the post. That was meant to be completely ironic.
Replacing the letters “RF” with an ellipsis was supposed to signify selectivity in reading. An obvious theme in the original story was that some reporters write the story that sounds good to them, not necessarily the truth. It occured to me that the same reporter who wrote “Study links mobile phone use to brain tumours” might also miss the distinction between RF and ionizing radiation. My blurb was satire expressing what I thought that reporter would think of this blog.
The rest is a compilation of multiple inane criticisms I’ve witnessed on this blog. I really thought that my post would be so obvious in its satire that it would not need any disclaimer, yet I included the “Look Ma” line just in case. As much as I hate to admit it, I suppose I’m just not good enough of a writer to imply “Just kidding.” I need to write it explicitly next time.
Sorry for the confusion.
Quote Comment
May 17th, 2010 at 11:53 pm
Shafe, I’d like to congratulate you on a very successful, if unintentional, troll. It gave me a good laugh.
Quote Comment
May 17th, 2010 at 11:59 pm
Shafe: When I saw the comment I thought to myself “Shafe? I thought that was one of the rational commenters. Ah nuts! Is someone name spoofing? Now I might have to require registration!”
One of the problems with satire on these kind of topics – no matter how far out in left field you try to be, there are probably really people who would believe it!
Quote Comment
May 18th, 2010 at 1:26 am
In an effort to balance out my radiation exposure and limit my risk of glioma I will now be attempting to hold my phone directly in front of my face, probably flat against my forehead.
This will serve to make me appear as intelligent as most newspaper science reporters.
Quote Comment
May 18th, 2010 at 10:18 am
I suppose that’s one problem with trying to be satirical on the internet. The internet’s background “Crazy level” is so high, it’s almost impossible to write anything crazy enough that people will realize you can’t possibly be serious.
I think that if Jonathan Swift were alive today and had posted “A Modest Proposal” as a blog entry, 85% of his readers would have thought he was serious and flamed him accordingly. 14% would have thought he was serious and agreed with him.
Quote Comment
May 18th, 2010 at 11:58 am
I just heard a piece on this study on the CBC and was very disappointed at how it was handled. Not surprised though. Reporters tend to have a limited understanding of statistics, favour “balanced” reporting that is inappropriate in a scientific context and share the common misconception that anything to do with “radiation” is dangerous.
Quote Comment
May 18th, 2010 at 2:37 pm
In a sense, this study was doomed to ‘failure’. If there was an actual increase in cancer then the study should show if it is related to mobile phone use. If there was a suspected cancer-causing mechanism, then the study should show if there is an actual increase in cancer. Without a definite objective, the study is just a fishing expedition.
It is very easy to produce false correlations in epidemiological studies – if the sample size is small enough and there are enough variables under consideration. Thus the extremely large sample size – which is difficult to keep free from ‘bias’. As soon as you divide the sample into groups according to mobile usage you are also dividing them according to lifestyle – there is a considerable difference in lifestyle between someone who uses a mobile occasionally and someone who never uses one. The lifestyle of someone who is on the phone for 12 hours a day is another thing entirely. As the cause of these types of cancer is unknown, just the process of selecting those that have the disease may be introducing a hidden bias.
If, as seems likely, there is no direct connection between cancer and mobile phone usage, then the study is going to throw up indirect connections – often silly and contradictory ones. When it comes to extreme use, the numbers are so small (50 over 5 hours a day) the sample size alone will cause random errors. In a sense, it is comforting to have these bizarre results included – nobody could have made them up.
Quote Comment
May 19th, 2010 at 1:08 am
I believe Shafes post is a perfect example of Poes law, which states:
“Without a winking smiley or other blatant display of humor, it is impossible to create a parody of Fundamentalism that SOMEONE won’t mistake for the real thing.”
Also, DrBuzz0 you missed one of the better headlines for a story about the same study:
“Mobile Phones actually protect you from brain cancer”
http://www.theregister.co.uk/2010/05/18/un_mobile_emf_study_wrong_answer/
Quote Comment
May 19th, 2010 at 1:20 am
Alan Barnard said:
This reminds me of a study published in the UK a week or so ago claiming that working overtime caused heart disease – I’ll freely admit I’ve not done all the leg-work but I’d imply it’d be hard to normalise for things like “stress”, “exercise” and “diet” which would almost certainly be interfered with by someone’s decision to work a few more hours each day.
Quote Comment
May 19th, 2010 at 3:26 am
by all means. you could probably link working overtime a lot to heavy phone usage as well. it seems to come with the overtime lifestyle.
Quote Comment
May 19th, 2010 at 10:37 am
the new scientist got some interesting articles about “deniers”:
http://www.newscientist.com/issue/2760
Quote Comment
May 19th, 2010 at 8:57 pm
Kingbob said:
This is not the first study to find this, actually. My own opinion is that the radiation from mobile phone usage is not likely to lessen the risk of brain cancer, because it doesn’t interact with tissue on that level. It’s hard for me to imagine a mechanism by which the rf emissions could alter chance risk one way or the other.
Most of the studies that have found lower rates found a fairly modest reduction. It may sometimes qualify as “statistically significant’ but only barely. There are some fairly simple explanations for this:
1. It’s just random variation. You don’t really expect the two groups to have zero difference, there will always be a little variation just due to random noise. Some of them will show an ever so slightly higher rate and others an ever so slightly lower rate. Do enough studies and some of them will show modest reduction and others a modest increase.
2. There’s some minor statistical bias. When a study like this is done they try to make sure that the group of those with cancer are compared to those of similar age, socioeconomic background, gender composition and so on. However it’s damn near impossible to insure that you have groups which are absolutely identical in all lifestyle aspects. People who talk on the phone more may tend to be more prone toward certain jobs, more tech-savvy, more social or something like that. It’s very hard to account for every single possible bias.
If the demographic controls are descent, these differences should be minor, but if the study is big enough and precise enough, even these tiny biases can creep in and slant the results.
What is happening here is that these repeated studies are struggling to attempt to detect any increase in cancer, even a vanishingly small one. In doing so, they’re squeezing statistical analysis and demographic controls to their absolute limits. It’s like trying to zoom in too far into a photograph – no matter how good the film stock, eventually you’re overwhelmed by grain noise.
Still, it’s an annoying thing to see this considered the “Wrong answer” because as far as I can tell, the hypothesis that radiation from mobile phones protects from cancer is at least as credible and evidence-based as the hypothesis that it causes cancer.
Quote Comment
May 19th, 2010 at 9:48 pm
DrBuzz0,
Study Sees ADHD-Pesticide Link
http://online.wsj.com/article/SB10001424052748703315404575250701118632926.html
Here’s a news story that I think is right up your alley. Covers the organic food industry fear mongering fairly well and is almost a copy cat example of your the mobile phone cancer study reaction. It’s clear cut correlation is not causation but I’d thought you’d enjoy ripping it apart.
Quote Comment
May 19th, 2010 at 10:44 pm
Carletes said:
Yeah, I just downloaded a PDF of the full study. I’m still reading through it.
Quote Comment
May 20th, 2010 at 8:15 am
Carletes said:
I presume that this isn’t trying to claim that ADHD causes the use of pesticides?
Is there any way we can reasonably link anything to ADHD? Reasonably we’re talking about a medical/psychological condition which has only fairly recently been widely known and diagnosed – a hundred years ago someone with ADHD would just have been described as a misbehaving little nutjob (and a thousand years ago as posessed by evil spirits).
Okay, the word “hyperactive” might have been suggested a hundred years ago (a cursory scan of the ADHD wiki page indicates that being first put forward around 1800) with ADHD only appearing in 1987.
Much like dyslexia, the moving “goalposts” on diagnosis have markedly changed the frequency of diagnosis – as has increasing education of teachers, doctors and parents.
[For info: Dyslexia is most commonly associated with major issues with literacy but can also manifest as marked numeracy problems, lack of organisational skills and messy, unco-ordinated handwriting. One of my old school friends missed out on the education he needed to follow his dream career in cybernetics and prosthetics because of a failure to pick up a dyslexia problem until just before his final exams.]
Quote Comment