Cell phones lower sperm count?

[cogreslab]

chrisberez said:

Research that has NOT been peer-reviewed. Hmmm, I wonder why that is...


Well, here are a few possibuilities:

1. The referees rejected the study.

2. It was never submitted

3. The authors did not wish to give away commercial advantage

4. There are substantive flaws in the peer reveiw system.

5. There was no pressure on the authors to submit.

In our case I do not recall we ever bothered to submit, but when we did the work was accepted and published. A scientist at the NRPB has recently suggested we should publish more of our work, and we may reconsider our policy.

 

[Bjorn]

Sperm count or not, it seems like cell phone users have more sex:

Mobile phones, web chat, and sex: a study of Norwegian adolescents based on a representative sample.
Willy Pedersen

Background. We investigated the associations between new interactive technology for communication, such as web chat or mobile phones, and sexual behaviour among Norwegian adolescents.

Materials and methods. A representative sample of adolescents (age 13 - 18, N = 10 926) filled in a questionnaire during school hours; the response rate was 92 %.

Results. Most adolescents have access to communication technology, but how much they use it varies. In particular with regard to mobile phones, a strong association to sexual behaviour was found. Among those who did not use the new technology, less than 10 % reported having had intercourse while two out of three of the most active users reported intercourse. The associations remained significant when controls were made for age and a range of contextual, family, peer and individual factors.

Interpretation. Norwegian adolescents have changed their sexual behaviour over the last decade. The introduction and widespread use of new communication technology is one of the most salient changes over the same period. The findings suggest that this technology may in fact be of importance to teenagers' sexual socialisation.

Let me add that "Tidsskrift for Den norske lægeforening" is absolutely not a woo-woo magazine. I can, however, think about a lot of reasons why those who have sex more often also have cell phones ... or was it the other way around?

Only this summary has been translated into English.
http://www.tidsskriftet.no/pls/lts/pa_lt.visSummary?vp_seks_id=1038905

 

[Vitnir]

The associations remained significant when controls were made for age and a range of contextual, family, peer and individual factors.

So that means that the mobile effect isn't just that those with mobile phones could be more outgoing and have more friends and thus have a greater chance (risk?) of meeting a partner?

I like the Tidsskrift for Den norske lægeforening too, I copied an article where they tested dowsing for energy-lines. Four dowsers had a go where they tried to find energy-lines independently in a large room. Guess what, they all got different results.

 

[Stumpy]

Originally posted by Vitnir
Well I'm sorry but that is strictly John Brignells opinion, there is no such defined limit for relative risks.

Not true, there has been a gradual degradation of the standard, allowing all manner of junk science to get published. Another view can be found here.

most epidemiologists consider a single study with a relative risk ratio less than 3 as not significant

Another view is available here.

In epidemiologic research, relative risks of less than 2 are considered small and usually difficult to interpret. Such increases may be due to chance, statistical bias or effects of confounding factors not evident

Originally posted by Vitnir In addition it depends entirely on the study if you can achieve such high relative risks. If you have a yes/no exposure then yes you can get that high.

Even a RR of 3 is not "high", I believe that the RR for active smoking is 20.

Originally posted by Vitnir One step in increased exposure (exposure vs no exposure) will then get you a relative risk of maybe as high as 3. But imagine that you have continous exposure data and one step of increased exposure is one year, to demand an increased risk of 3 for each year of exposure, well you soon will realize that you will come close to a infinite risk.

The main issue is what is an acceptable MINIMUM risk ratio, not the maximum one. What do you consider to be an acceptable RR then?

 

[Vitnir]

Acceptable RR? I don't know what you are talking about. If the RR is separated from 1.0 using a 95% confidence interval its a significant association. After that we can talk about why it is elevated, confounding, bias, or exposure etc. Depending on the measured data we can debate on how certain the results are, for questionnaire data you might like to have a high RR to be more sure of your conclusions but there are other occasions where a RR of 1.40 is considered strong.

There are people that doesn't like epidemiology and always says it doesn't prove anything. Well if it hadn't been for epidemiology we may still not have known that tobacco gives cancer. There are lots of occasions where you can't use double blind clinical studies for economical or ethical reasons. It is true that epidemiology has got a bad reputation since journalists tend to scream their head off when a new association is published but that doesn't change the underlying science. To arbitrary reject results like that sounds very strange.

 

[Stumpy]

Acceptable RR? I don't know what you are talking about. If the RR is separated from 1.0 using a 95% confidence interval its a significant association.

I have not the words!!!!! So, one extra person exposed to a putative cause, suffering an effect is significant???????
You study 10,000 people, divided into 2 groups, to study the association between eating burnt toast and getting eyebrow cancer. Your control group of 5,000 don't eat burnt toast but 500 of them have contracted eyebrow cancer. Within the burnt toast eaters you find that there is 501 cases of eyebrow cancer giving a risk ratio of 1.002...YOU THINK THIS IS SIGNIFICANT!?!?! Not withstanding the fact that your 95% confidence interval means that the result had a 1 in 20 probability of being obtained by chance alone, do you honestly not think that the 501 figure falls well within the range of numbers expected by a normal random distribution? Do you honestly think that if the national rate of eyebrow cancer is 10% for every 5,000 people that you study there will be exactly 500 eyebrow cancer cases??!?!?!?

This sums up exactly everything that is wrong with epidemiolgy today.

After that we can talk about why it is elevated, confounding, bias, or exposure etc. Depending on the measured data we can debate on how certain the results are, for questionnaire data you might like to have a high RR to be more sure of your conclusions but there are other occasions where a RR of 1.40 is considered strong.

Name one occasion!!!!! Run a simulation for yourself with mathcad, you will be hard pressed to find a RR of 2.0 by chance, but you will find many RR values in the 0.75 - 1.50 range, BY CHANCE ALONE!!!!

There are people that doesn't like epidemiology and always says it doesn't prove anything.

Probably because it's supporters think that a result that falls well within the range of chance is "significant"!!!!

Well if it hadn't been for epidemiology we may still not have known that tobacco gives cancer.

Huh??? Name ONE study that shows that tobacco gives cancer. Another fallacy perpetuated by the epidemiologists! The nicotine content of potatoes, tomatoes and aubergines is thousands of times higher than a cigarette.

There are lots of occasions where you can't use double blind clinical studies for economical or ethical reasons. It is true that epidemiology has got a bad reputation since journalists tend to scream their head off when a new association is published but that doesn't change the underlying science. To arbitrary reject results like that sounds very strange.

To accept a result that falls well within statistical chance as significant is waaaaayyyyyyyyy stranger.

 

[steenkh]

Huh??? Name ONE study that shows that tobacco gives cancer. Another fallacy perpetuated by the epidemiologists! The nicotine content of potatoes, tomatoes and aubergines is thousands of times higher than a cigarette.

I am not a medical type, and I am sure that if I googled for such a cancer study, you would probably be able to tear it apart, and I would not know what hit me! So you are stating that there is not more chance of getting lung cancer if you smoke than if you do not smoke? I had the clear impression that this correlation was undisputed. At least here in Denmark. Or are you saying that most of those people who have got lung cancer can blame potatoes and not cigarettes?

I am very interested in this, because the father of one of my friends is touring schools to tell the pupils to quit smoking, using himself as the bad example: he has had one lung removed because of lung cancer.

 

[Rolfe]

Huh??? Name ONE study that shows that tobacco gives cancer. Another fallacy perpetuated by the epidemiologists! The nicotine content of potatoes, tomatoes and aubergines is thousands of times higher than a cigarette.

Nicotine does not cause cancer. However, the effects of nicotine cause people to become addicted to smoking cigarettes, the carcinogenetic tars of which cause cancer.

Epidemiology of smoking and lung cancer (BBC news link)

Rolfe.

 

[Stumpy]

I am not a medical type, and I am sure that if I googled for such a cancer study, you would probably be able to tear it apart, and I would not know what hit me! So you are stating that there is not more chance of getting lung cancer if you smoke than if you do not smoke? I had the clear impression that this correlation was undisputed. At least here in Denmark. Or are you saying that most of those people who have got lung cancer can blame potatoes and not cigarettes?

I am very interested in this, because the father of one of my friends is touring schools to tell the pupils to quit smoking, using himself as the bad example: he has had one lung removed because of lung cancer. [/B]

I think that this highlights the limits of most epidemiological studies. Vitnir stated that tobacco gives cancer. This is incorrect, you correctly point out that there is a strong correlation between cancer and active smoking. You may think that this is a subtle difference, in fact it is quite an important one and certainly one that is often overlooked in the field of epidemiolgy and ALWAYS in the popular press. Corrrelation is NOT causation.

AFAIK there is no experimental data in the world that proves that ANY substance in the world CAUSES cancer.

 

[Bjorn]

So that means that the mobile effect isn't just that those with mobile phones could be more outgoing and have more friends and thus have a greater chance (risk?) of meeting a partner?

I could guess that those who (at the time of the research) were in love (and had sex) called (each other) more than those who were single? Or maybe they felt a need to discuss the possibility of pregnancy ...

In other words, I think those who have sex will use the cell phones more, not the other way around.

 

[Rolfe]

AFAIK there is no experimental data in the world that proves that ANY substance in the world CAUSES cancer.

Asbestos, for one?

Rolfe

 

[Rolfe]

Oh, here's an interesting one.

Perhaps I was wrong about what I said above about nicotine not causing cancer.

All these papers about "known carcinogens", and quantifying and reducing their presence in so many areas, and none of them has ever been conclusively proved to cause cancer?

I think this may be some new definition of "proof" that might make even the strongest sceptic blanch a little.

Rolfe.

 

[Pragmatist]

I could guess that those who (at the time of the research) were in love (and had sex) called (each other) more than those who were single? Or maybe they felt a need to discuss the possibility of pregnancy ...

In other words, I think those who have sex will use the cell phones more, not the other way around.

I think it's a side effect of all these tobacco/cancer scares. In the old days you would have sex and then enjoy a nice cigarette. These days the cigarette is out, so you have to substitute a mobile phone call instead!

"Was it good for you honey? Yes? Great! Pass me the mobile would you...?"

 

[Vitnir]

Well I was puzzled at first but now I realize that I assumed that you knew what relative risk was. To use your example the RR would be 1.002 but the confidence interval would be 0.891-1.127 so its not separated from 1.0

If you instead think that 501 toast eaters get cancer out of 5000 people compared to 400 who didnt eat toast from the control group of 5000. That gives a RR of 1.253 with a confidence interval of 1.105 - 1.420

101 extra cases of cancers, do you think thats insignificant?

And obviously its a 1 out of 20 chance its wrong. Who said otherwise?

Edited: Because I wrote some wrong figures before.

 

[Vitnir]

Calculator
And here is a Java-calculator where you can knock yourself out calculating relative risks.

This is a simple yes/no exposure example. If you want more complex exposures, maybe weigh in the number of years they ate toast etc it gets more complicated.

 

[Vitnir]

Vitnir stated that tobacco gives cancer.

Yes I was thinking about smoking but I didn't write it.

 

[Stumpy]

101 extra cases of cancers, do you think thats insignificant?

Statistically yes, it is indeed insignificant - I trust we are not about to muddy the water by protesting the devasting effect on the sufferers of this disease, we are talking statistics here. The number falls well within the bounds of a random distribution. A risk ratio of 1.2 is utterly meaningless.

Yes I was thinking about smoking but I didn't write it.

Okey-dokey, where can I read about the epidemilogical study that proves that "tobacco gives cancer"?

 

[Vitnir]

No further comment needed

 

[Stumpy]

No further comment needed [/QUOTE

...does that mean you aren't going to back up your assertion that "tobacco gives cancer"?

 

[Lucianarchy]

Stumpy, I believe Vitner meant 'smoking causes cancer'.

But asking people to back up their claims is a bit rich coming from you, considering you made a bold, clear claim about some police guidelines for dealing with psycics, over a year ago

BTW,. the most at risk from cellpohnes are the young and old. You claim they are the least likely to use them. Do you have any evidence of this? Do you not ever see kids with mobiles these days or something? Being vulnerable groups in terms of informed responsibility, don't you agree that those two groups need protections against such risks?