Skip to content
 
If you can read this, either the style sheet didn't load or you have an older browser that doesn't support style sheets. Try clearing your browser cache and refreshing the page.

(telegraph.co.uk)   Newly released study says mobile phones cause brain cancer, but other studies conclude they don't cause brain cancer. I hope that clears things up for you   (telegraph.co.uk) divider line
    More: Obvious  
•       •       •

1643 clicks; posted to Main » and Fandom » on 25 Jan 2007 at 9:59 AM (14 years ago)   |   Favorite    |   share:  Share on Twitter share via Email Share on Facebook



40 Comments     (+0 »)
 
2007-01-25 10:01:31 AM  
Nick Naylor unavailable for comment.
 
2007-01-25 10:01:50 AM  
What about ball cancer? I carry my phone in my front pocket because I refuse to wear one of those leather holsters.
 
2007-01-25 10:04:52 AM  
Haven't you ever heard of radiation therapy? Cell phones cure cancer.

/At very minor risk of brain cancer this morning.
 
2007-01-25 10:05:43 AM  
any of you watch that internet video of the egg being cooked between two phones? Is that legit? Does our brains look like ground beef in a microwave?
 
2007-01-25 10:06:10 AM  
A new study confirms that 72% of the statistics in studies are completely made up.
 
2007-01-25 10:07:07 AM  
Key question, how many of those other studies were industry funded studies published on the web vs. peer reviewed?
 
2007-01-25 10:08:07 AM  
Feh, everyone on both sides of my family dies of cancer eventually. I figure it's only a matter of time for me, cell phones or not.
 
2007-01-25 10:09:34 AM  
I'll be ok, i'm not a smoker.
 
2007-01-25 10:09:58 AM  
Call me a pessimist, but if one study says something is bad for you, but another says it is not, I tend to believe the one that says it is bad.
I also tend to believe that the study which says it is not bad, is propaganda created by the industry that makes the product.
It is hard for me to believe that having something emitting microwaves close to your brain can possibly be healthy. I am also afraid of keeping it in my pocket all day.
I have started using the hands free kit, and keeoing the cell phone away from me whenever possible.
 
2007-01-25 10:11:29 AM  
Oh, people can come up with statistics to prove anything. 14% of people know that.

/not obscure at all
 
2007-01-25 10:13:05 AM  
So cell phones that didn't cause cancer, but then did cause cancer, but then didn't cause cancer, now cause cancer again?

I sooooooooo confuse.
 
2007-01-25 10:22:56 AM  
[image from i2.photobucket.com too old to be available]

"The battle for the mind of North America will be fought in the cellphone arena: the Cellodrome. The cellphone is the eardrum of the mind's ear. Therefore, the cellphone is part of the physical structure of the brain. Therefore, whatever we hear on the cellphone emerges as raw experience for those who listen to it. Therefore, the cellphone is reality, and reality is less than the cellphone."

[image from i2.photobucket.com too old to be available]
 
2007-01-25 10:23:02 AM  
While ordinarily, betting on the safer of two competing theories may be worthwhile, in this case it's an extreme claim warranting extremely compelling evidence. What makes me feel safe using a cellphone so close to my brain? Not using my brain helps, but for everyone else it's because we aren't talking about especially high-intensity or high-frequency radiation. It takes some serious juice to induce DNA damage to the degree that cancer is the result. If the claim were true to any significant degree there should be a cataclysmic epidemic of cancer, highest location frequency at the ear and radiating outward. Is there? There is not. I'm as worried over getting cancer from answering my phone as I am from aspartame.
 
MrT
2007-01-25 10:26:04 AM  
Walker
Essentially what happened was the original study that made this link showed that people with tumors tended to report that they hold their mobile phone on the same side of their head as the tumor. The some perceptive scientist pointed out that the overall tumor rate was no higher in phone users than non-users, so as a result, if there is an increased chance of getting cancer on that side of the head, there must be a corresponding decreased chance of getting it on the other side, such that the overall rate is the same.

So while the main conclusion of the study was
mobile phone use increases the risk of cancer close to where you hold the phone
an equally valid conclusion would be
mobile phone use decreases the risk of cancer in the parts of your body away from where you hold the phone

So really, the only sensible conclusion you can draw is that the study is likely to be flawed, and people with tumors are more likely to report that they held their phone on the same side as the tumor, even if the phone has done nothing at all.

This study also seems to have found no overall increase in tumor rate, so it suffers from exactly the same problem. In order for there to be a genuine correlation, the mobile phone would have to be causing cancer in one place, while preventing it elsewhere. I.e. same old shiat. It doesn't matter how you polish it, this one's still a turd.
 
2007-01-25 10:28:13 AM  
I remain skeptical.

When a study finds a correlation between A and B, what they typically really mean is "If there weren't a correlation between A and B, then there'd be less than a 5% chance that we'd see the results we're seeing in our sample." For example, if I study 5000 people to see if they have a certain form of tumor, I can look at the statistics for how many people get that form of tumor and say "Well, normally about 12 people out of 5000 would get this tumor. And there's only a 5% chance that more than 20 people out of a random group of 5000 would have this tumor. So if I find 20 or more tumors in this group, there's something funny going on."

The problem with this is that, of course, every so often you will just happen to pick a really unlucky group of people purely by chance. So in a field that's studied to death (poor choice of words) like cancer, and especially where there are so many different tumor varieties that can be studied (so studies often include lots of different sample groups), you can expect studies to come along every so often that show a "statistically significant correlation" between cancer and something or other, purely by chance.

Now, if somebody else does another study, with the same sort of sample group, and finds the same correlation with the same type of tumor, things get interesting. But until the results of this study can be confirmed in that way, I'm inclined to think it's just random.

/but I still don't hold my phone next to my head
//or next to my weener
 
2007-01-25 10:31:35 AM  
Prepared:
[image from img.timeinc.net too old to be available]
 
2007-01-25 10:38:21 AM  
KickahaOta Ah, yes. Type I errors vs Type II errors. You're right --unless a confirming study is done, the results are best taken with a grain of salt.
 
2007-01-25 10:39:52 AM  
Sorry, didn't mean to get so bold here.

/closing tags are for anal-retentives.
 
2007-01-25 10:42:04 AM  
So is that why there's blood oozing out of my ears?
 
2007-01-25 10:47:06 AM  
Proof that correlation is not causation.
 
2007-01-25 10:47:16 AM  
Cell phones use radio waves. You know what else produces radio waves? The Sun. Let's get the scientists working on turning the Sun off.
 
2007-01-25 10:55:20 AM  
Today lesson on odd ratios is brought to you by a junior level biostatician at an unnamed cancer research hospital.

An odds ratio is most often obtained from one of two types of test, Logistic Regression or Cox Regression (sometimes called Cox Proportional Hazards).

In these tests, a dependant variable (or outcome) is numerical coded into a dichotomous code. That basically means that there are two options, 0 and 1. In this case didn't get cancer was coded at 0 and got cancer was coded as a 1.

Then a variable is entered into the equation to see if it can predict assignment to 0 or 1 more than random chance can. In this case the variable "uses cell phone" was entered into the equation. For simplicity of results, the variable was vector coded (or dummy coded) into a 0 or 1 instance as well. The 0 group was those who didn't use cell phones and the 1 group was those that did.

The end result of this division is that the statistical test will determine if assignment in the cell phone group makes you more likely to be in one of the cancer options.

The numeric output is called an odd-ratio (OR). In this case the odds ratio was 1.39, which means that those who used cell phones were 39% more likely to get this specific type of brain cancer on the side they used the cell phone.

Lets take this number a little further. What is not said is how many people got cancer in each group, we only know that the long term cell phone users had more cancer (39 % more). A 39% increase can be a lot or it can be very little. It depends on the prevelance of the basline risk. If normally only 1 in 100 people get this type of cancer, then out of 100 people who use cell phones for a long period of time, 1.39 out of 100 will get this cancer. If 50 out of 100 traditionally get this type of cancer, then using a cell phone increases the prevelance to 69.5 out of 100.

Further research on my own has found that the united states reports 20,000 cases of gliomas a year. The current US population is about 300 million. So the presentation of a glioma is 1 per 15,000 persons. If everyone were to use cellphones in a prolonged fashion, the instance of gliomas would increase to 1.39 in 15,000 or 27,800 cases a year.

But that isn't even what the study was about. The study only found associate with the tumor on the side of the brain that the cell phone was used in patients that already had cancer. I'll come back to this.

Now lets talk limitations of odds-ratios. An odds ratio's significance is generally based on 3 factors : the size of the odds ratio, the variance of the variable, and the sample size used. These three factors affect signifance in all staistical tests.

Basically an Odds-ratio of 10.0 is more signficant than an Odds-ratio of 2.0.

An odds-ratio of 2.0 that has a confidence interval of 1.9-2.1 is more significant than an odds-ratio of 2.0 with a confidence interval of 1.5-2.5.

Lastly, an Odds-ratio of 2.0 with a conidence interval of 1.9-2.1 based on a sample size of 1000 is more significant than an odds-ratio of 2.0 with a conidence interval of 1.9-2.1 based on a sample size of 100.

It tell you this because ANY odds-ratio will become statistical significant if you collect a large enough sample. Large samples such as the one taken in this study always return significant odds-ratios, just small ones that simply could be due to natural variablity that occurs in each sample. The odds-ratio reported in this study is a small one.

The end result is that this study did find an increased rate of cancer, and even if the numbers reported were used to demonstrate an increased rate of cancer, the increase is small.

What they did find is that in patients who already had cancer, the cancer was 39 percent more likely to be on the side of the brain that they held their phone up to.

But only in patients that had used cell phones for 10 years.

Then end result is that cell phone usage seems to predict the side of tumor development. Assuming even distibution of tumor location, right or left, a 39% increase would mean that the tumor was on the side the cell phone was used 69% of the time.

This study reaks of data mining. The results are not as shocking as the reporter claims, and the effect is small.

On a more opinioated note, the misuse of regression in the medical field is appalling. The lack of understanding and the misapplication of these numbers happens with great frequence. Regression has different rules and interpretations than effects testing, which doctors use all the time (properly too), and doctors fail to recognize this sometimes.

This is just fine by me, because it gets me publications and employment.
 
2007-01-25 10:58:37 AM  
FTFA- Prof Challis, who is negotiating funding for a long-term international study, said last night: "I agree with the authors that this is a hint that needs further exploration. It's further reason why a long-term study is necessary."

SuburbanCowboy
It is hard for me to believe that having something emitting microwaves close to your brain can possibly be healthy. I am also afraid of keeping it in my pocket all day.
I have started using the hands free kit, and keeoing the cell phone away from me whenever possible.


Unless that is a wired hand-free kit you're just sticking another radio transmitter right by your head.
 
2007-01-25 11:12:26 AM  
submitter: Newly released study says mobile phones cause brain cancer, but other studies conclude they don't cause brain cancer. I hope that clears things up for you

You forgot to add "Still no cure for cancer studies" to your headline.
 
2007-01-25 11:28:12 AM  
submitter: Newly released study says mobile phones cause brain cancer, but other studies conclude they don't cause brain cancer. I hope that clears things up for you


I wonder if this is the "cigarettes cause cancer" scare or the "electric power lines cause cancer" scare.
 
2007-01-25 11:46:00 AM  
soakitincider

How long does it take to cook the egg?

If it's in the order of years then maybe it's legit.
 
2007-01-25 12:14:49 PM  
hockeyfarker

Cell phones use radio waves. You know what else produces radio waves? The Sun. Let's get the scientists working on turning the Sun off.

The radiaton from the sun also produces warmth. Does that mean its safe to crawl into a 400 degree oven?
 
2007-01-25 12:17:12 PM  
No submittard the difference between this study and the others was this was a LONG TERM study while the others were SHORT TERM. Do you understand or do I need to capitalize more words?

The conclusion is that there need to be more long term studies.

And I too keep my phone in my front pocket Mugato, and ball cancer is definitely something I'm a bit concerned about.
 
2007-01-25 12:41:21 PM  
Republican,
This was not a long term study, it was a chart review. It looked back on person's long term cell use.

There is a difference between following a group of people that use cell phones for 10 years and then observing the outcome and simply asking everyone with cancer how often they use a cell phone.

If you are worried about cell phones and cancer based on this study, you don't understand statistics.

Again, I work in cancer research. I am assigned to a non-hodgkin's lymphoma group.
 
2007-01-25 12:44:17 PM  
Hi, how ya doing? Yeah, great. If anyone is so inclined, feel free to stop using the phone. And don't live under power lines. And don't run a microwave. But to be consistent, you'll have to stop exposing yourself to all UV radiation, radio waves, and every dietary GI irritant.

Mythbusters couldn't even *warm* a turkey after an hour stuck on a radar dish and they showed that microwaves heat from the surface inward. Further, radio waves interfere ZERO percent of the time with your cells. Microwaves won't damage your DNA except by the generation of heat caused by water molecules with ants in their pants. But you'd know if you were getting that kind of dose, because your flesh would burn. And for that matter, the heat would just cause strand separation, not point/frameshift mutations.

Your cell phone will not, in 1000 years, kill you. On the other hand, your automobile just might. Still gonna drive?
 
2007-01-25 12:58:12 PM  
My personal crusade is against the lay interpretation of the word "significant," Statjunkie. What's discouraging is how many researchers don't get it. All the time I see results posted with the various p values of the data--as if coming in EVEN MORE under alpha makes it EVEN MORE likely to be true. And when did they stop teaching about power in stats? Yes, I'm staggered that in a sample with a n=300,000 your multiple R was significant. What's the variance? Oh wow--1% huh? Yeah, you're really on to something, boyo.
 
2007-01-25 1:20:37 PM  
Sleepy,

Nice to know someone else out there knows this phrase:

"With enough power, everything is significant."
 
2007-01-25 2:00:22 PM  
Yep. Between that and "all outcomes are possible by chance alone" you've pretty much taken the non-symbolic portion of my statistical theory course.

Anyway, keep your hands up and keep throwing punches.


Cheers,
SG
 
2007-01-25 2:27:35 PM  
I'm an electrical engineer. Your cell phone will not give you brain cancer. It's a radio transmitter, not an X-ray machine. It doesn't have near the power to split a DNA molecule.

You get more than a hundred times as much radiation by simply going outside in the sun.

So if you're using a cellphone indoors, you'd be getting less than 0.01 of the radiation you'd be getting if you were outside, instead.
 
2007-01-25 3:38:48 PM  
StatJunkie
Again, I work in cancer research. I am assigned to a non-hodgkin's lymphoma group
Sim TreeI'm an electrical engineer

I'll bet you are both getting a kick out of these replies.
 
2007-01-25 6:19:20 PM  
I am 19 and recovering from brain cancer.

I have a cell phone and use it like maybe 10 minutes a day at most.

So I'm really getting a kick out of some of these replies.
 
2007-01-25 8:29:12 PM  
Sim Tree
Its true that radiation in the microwave spectrum can´t ionize a covalent bond and thereby directly damage DNA, but microwaves as we all know cause molecules to vibrate rapidly and heat up. CANCER is a generic term and there is no known single trigger, but the chances of any given cell becoming malignant are increased by most kinds of stress, including heat stress.
While cell phones may not cause cancer, I would certainly put them in the catagory of likely not to help!

You know the egg thing talked about earlier in this thread? I´ve done it, and it works. WHY it works exactly is not clear, cell phones do produce heat directly when transmitting, not just microwaves.
 
2007-01-25 10:25:20 PM  
Farkers who use cell phones and don't have cancer yet are just like smokers who say they they haven't got lung cancer yet. Aw, guess it's okay then. Suckers.
 
2007-01-26 1:11:23 AM  
John, you undergo greater heat stress while taking a shower, wearing a coat, or walking to the mailbox. Not only will neither form of radiation cause mutations to occur at all, but your body has multiple error corrections systems constantly scanning for and fixing mispaired or methylated bases, dimers, etc. Look up the power rating for your mobile and work out how long you'd have to keep that thing going to equal the one second output of a microwave oven on high. Then remember that it doesn't matter anyway, because power isn't an expression of total energy, but rather energy / second.

YOU. WILL. NOT. GET. CANCER. FROM. YOUR. CELL. PHONE.

/unless you smear it with peanut butter and radium and swallow it.
 
2007-01-26 2:24:30 AM  
The people who are on cell phones most also typically have the most stressful lifestyles, and stress leads to a weaker immune system, which promotes cancer. I'm betting that even if there's a link between cell phones and cancer, it's correlation, not causation - unless you can show that the incidence of cancer on the side of the head closest to the cell phone is significantly higher, and I haven't heard anything like that.

The fields put out by a cell phone are so weak as to be practically nonexistent. Anyone who thinks cell phones are a health risk (aside from the added stress they cause) is a complete moron and obviously never took physics.
 
Displayed 40 of 40 comments


This thread is archived, and closed to new comments.

Continue Farking





On Twitter



  1. Links are submitted by members of the Fark community.

  2. When community members submit a link, they also write a custom headline for the story.

  3. Other Farkers comment on the links. This is the number of comments. Click here to read them.

  4. Click here to submit a link.