Sunday 12 September 2010

My slide show from Newcastle Upon tyne

My Happy Life Slideshow: The Eye’s trip to Newcastle upon Tyne, United Kingdom was created by TripAdvisor. See another Newcastle upon Tyne slideshow. Create your own stunning free slideshow from your travel photos.

Thursday 22 July 2010

Regina Mwanza Chiluba Personal Interview with Amos Malupenga (Post Newspaper)

I AM THE GOOD THING IN FTJ'S LIFE

I am the good thing in Chiluba's life, said former MMD chairperson for women affairs Regina Mwanza yesterday.

In an exclusive interview, Regina, who divorced Eddie Mwanza of Indeni Oil Refinery last year, revealed that she got married to former president Frederick Chiluba on May 6, this year in Lusaka. 

This was at Lusaka Boma local court under the customary law.

"Of course Dr. Chiluba has found a good thing according to the Bible," Regina said. "The Bible says he who finds a wife, finds a good thing.

So I am sure I am the good thing in his life and I would want to be that forever."

Chiluba, who was found relaxing with Regina in one of the living rooms, excused himself because he didn't want to witness the interview. "I will leave the two of you to talk because she may be jittery if I am around, so I will join you when you are through," said Chiluba before he went to watch soccer on Supersport. And Regina was not jittery. She answered every question with a lot of confidence. She also laughed at a number of questions which she said people were creating through speculation.

Thank you for this opportunity but I know it has been long overdue. People yesterday (Friday) received news about your marriage with the former president with a sigh of relief as you know you have made a few public appearances together and people had made conclusions already but were just wondering when the official announcement would be done.

Some people were saying you would do things like Nelson Mandela who lived quietly with Gracia Machel until Bishop Desmond Tutu encouraged them to marry officially because they were setting a bad example to the young people in the nation. So my first question is: Why did it take this long to marry when both of you divorced a long time ago? 

Well, before I answer that question, I must say that you are most welcome to our home and we are very grateful that you were able to make time to come. I don't believe we took long to get married as you realise we both came from a background of divorce and divorce is a very traumatic situation, it is a very traumatic thing, especially where children are involved.

There are children involved in these marriages and in these divorces and you have to take them into account. You have to think of them before you jump into the next marriage. So we wanted first to look at those issues and make sure that the children were settled and ensure that they didn't feel divorced as well because when you divorce the father, children become a little bit confused, they don't know which way to go, they don't know whether to go to the mother or go to the father but we had to make them relax and feel that they were welcome in either home, either with their father or their mother.

And that is what we managed to do, we managed to relax our children. So only after that did we start thinking of getting married. And I had to make sure that I wasn't jumping into this marriage because of the divorce or for other reasons, I wanted to get into this marriage because I loved the man and that is what I wanted to do and that is what I wanted for myself. So we had to take time to do that.

For how long have you known each other now? 

It depends on what you mean, whether politically or courting.

Okay, start politically and end with courting. 

As a political leader, I have known him for a while since I was Constituency chairperson for Ndola Central where Hon. Eric Silwamba was my member of parliament then in 1993. I knew him not personally as such but just as my leader in the party. I can't even say we knew each other then because we would just welcome him and execute programmes for him and things like that. So I have known him quite a while in that area. As in courting (she laughs before proceeding with the answer) that's a good one. Not as speculated really. There were a lot of speculations. Our friendship was platonic for a long long time actually. It was just a working relationship. I knew him in that manner (courting) for the past two years really (since year 2000).

But there are allegations that your relationship started when both of you were married. I know you have touched on that but how true is that? 

It's not true. I separated with my husband five years ago, I think. I moved to another house but before I moved to another house, we were staying in the same house. But maybe that is a little bit too deep. I think I better leave that out. But it had nothing to do with Dr. Chiluba or politics.

How have the children from both sides taken this marriage? 

That's a difficult question to answer. From what we are seeing, they are settling down well but it is difficult to get into their hearts to really know what is in them but we are hopeful that everything will go well. They are trying to get to know one another, they meet.

This marriage has come at a time when Dr. Chiluba's life is in problems arising from politics. How does it feel to start the new marriage like this? 

Well, I came into this marriage fully aware of what was happening. I married Dr. Chiluba for himself, for him. You know when you take vows, I am sure you are familiar with the vows, for better for worse, for richer for poorer, in sickness and in health, in problems as well. So really the problems have no bearing on our marriage. And when you are in problems like this, we have given ourselves to prayer, a lot of prayer. My husband is quite prayerful and this has lightened our burden quite a lot. I just love the man (she laughs) so this really will not affect our marriage.

How did you marry? Was it under the Act or it was customary? 

We married under customary law, hoping to regularise as time goes on under the Act. We had our reasons for that.

What are those reasons? 

Well, looking at the political environment and circumstances we sat down and thought of not publicising this whole thing. If we had gotten married under the Act, it would have meant that we advertise for 21 days and people are a little bit funny sometimes; they could have put a caveat or something and we wanted to avoid all those things. We said that it would maybe draw a lot of comments, maybe good comments, maybe bad comments so we decided that we just get married quietly.

And how much did Dr. Chiluba pay as dowry? 

Oh, no. That's not a question for me to answer. That's for my parents. Surely, I wouldn't go snooping around asking how much did he bring?

Who did Dr. Chiluba see from your side and who was his bashibukombe? 

His bashibukombe was a Mr. Katebe from Chingola, he is the district chairman in Chingola and he saw my aunt in Kapiri Mposhi, my father's sister and my mother's sister. Both my parents are dead but I have a foster mother, a Mrs. Musuka.

We heard that there was Matebeto for Dr. Chiluba last year in Ndola, who ate with Dr. Chiluba? 

She laughs) Oh, God. Unfortunately that was pure speculation. There was nothing like that for the same reason I gave you that we didn't want to publicise this whole thing. But definitely there will be at an appropriate time and you will be invited.

How does it feel to be Mrs. Chiluba? A wife to the former head of state? 

Well, the office of the president never ceases really. It has its own responsibility to shoulder even after someone leaves the office. There are responsibilities commensurate to his status. I am trying to learn to cope with those like we have streams of people here, almost everyday. It's a big challenge. You receive all sorts of people from grassroots to up there and you just have to adjust and know how to treat people. It's a big challenge but I am enjoying it.

People said you left your humble husband for the King and that you used to yearn for State House so much, how true is that? 

Oh my God, that's not true. We had our own differences with my previous husband but I have a lot of respect for him and I know that he has a lot of respect for me. Our divorce was quiet, of course you picked it up but we decided to do it quietly because we have a lot of respect for each other, we knew what kind of differences we had. It has nothing to do with Dr. Chiluba, absolutely nothing. Of course there were speculations and I am sure you must have seen my previous husband sometimes standing for me but there was nothing between me and Dr. Chiluba that time. Yearning for State House? I would not yearn for State House because I knew that Dr. Chiluba was going to leave office at one time. State House does not make a person.

He didn't tell you about the third term? 

I knew about the third term but not at one time did he ever tell me that he wanted to go for it. The debate was there but he never told me that he was going to stand so I knew it. And knowing him the way I do, when he keeps quiet about something, I knew that he was not going to stand. So the question of continuing in State House was completely out. State House has its own benefits. If there is no one that you love there, it's just a building, it's just a building.

At what point did your marriage with Mr. Mwanza break? Is it Dr. Chiluba who broke it?

I think I have already answered that. I said Dr. Chiluba had nothing to do with the breaking of my marriage. We had our own differences which I would not like to go into. We separated, I think this is now the fifth year.

How do you see the future of your new marriage? Where are you headed with these unresolved problems around Dr. Chiluba? 

Well, like I have already said, when problems come you expect them to pass. The Bible says all these problems we are going through are but for a season. So I don't think we would go on and on with problems. But even if these problems continued really, it's Dr. Chiluba that I have married and I intend to have a successful marriage, very very successful marriage with him.

Are we expecting some formal function to officially celebrate your wedding?

Oh yes, when time comes. I cannot tell now but you will be informed.

Who is Regina? Tell us where you were born and when. How you grew up, your education background and about your family. 

I was born in Ndola Central Hospital on 23rd November 1962. I think I had a very wonderful childhood. Both my parents were Christians and disciplinarians, especially my father. My father was a businessman. He had a small shop in Lubuto, that's where I grew up in Lubuto.

He had two trucks which were bringing in fish from Mpulungu and the other side. I started school in 1968 at Kabushi 'A' Primary School. I was there for two years and then I moved on to Dambo Primary School. I jumped a grade. I don't know whether they thought I was a little bit clever, I am not sure, or because the teacher was my aunt.

From grade two I went to grade four. After that I went to Ibenga Girls Secondary School in 1974. I was at Ibenga until 1976 when I went to the Convent. I finished school at Ndola Convent in 1978. Thereafter, I went to Luanshya Trades for a secretarial course. After that, I joined Indeco Properties here in Lusaka where I worked for a year and went back to Ndola and got married in 1983 to my previous husband.

I had three children with him, one is 17 - the girl, the boy is 15 and another girl is 13. That's Bwalya, Mando and Malama. I have two other children that I have kept from childhood, my late brother's children so I normally say that I have five children because I have kept those with me since both their parents died, that's Makasa and Lombe. In the family, there were seven of us. We lost the first born, Bernard.

The second born, Eric, is in America. He has been abroad for about 15 years now. The third born, Fred, is the father to those children that I live with. I was the fourth born. The fifth born Charles is in Ndola, he is running my farm. Before that he was running my other businesses. Then there is my sister, Catherine, my only sister.

The last one was Frank, the one who just died recently. My brother in America is an occupational therapist. He runs his own clinics and Charles used to run my businesses in Ndola. Some of them have closed down but some are still there. I am only left with two brothers now and one sister. My sister is running my shop and my salon. My brother Charles is running my farm.

Is it true that you have a child with Dr. Chiluba? 

(She laughs) Oh my God, speculation, speculation, speculation and speculation again. You know it's so funny because one day I was in town and I was driving near the robots at Holiday Inn where those kids sell newspapers and I looked out of the window, I see this boy holding a paper, the pink one then; ah, ah: 'Regina Mwanza pregnant, what, what in Bahamas' or whatever I can't remember what was there. And I was with my daughter so she says 'Mummy what is this?'

So I asked the boy saying : iwe, walibeshiba aba balembele apa? (You, do you know this one who has been written about?) So the boy said ee, baliya, baliya. Niba Regina Mwanza aba. Baliya ku Bahamas (the boy said yes, that's Regina Mwanza, she is gone to the Bahamas). I said really, do you know her? He said ee twalibeshiba niba mukaba Chiluba aba (Yes, that's Chiluba's wife). And then I said but I am Regina Mwanza. The boy shook, he even dropped the newspapers. I even felt sorry for the boy but I just laughed it off because once you decide to be a public figure you should expect some of these things. And even your newspaper wrote that.

I don't know where you got that from. I would love to know really because when you were writing that, I think you wrote it a day after (George) Mungwa's death. I was at Mungwa's funeral and one boy from the Daily Mail came to me and said, 'ah but we thought you were in America'. I said I am here, very much here. I am not pregnant and I don't intend to be pregnant. The boy said say something but I said I don't intend to say anything. So where did you get that story? Malupenga: You will have to conduct an interview with me, so make an appointment and we will discuss that in detail. Today, it's me who is interviewing you.

But do you intend to have any children with Dr. Chiluba? 
Well, we haven't thought about that yet. I think we are still enjoying the honeymoon.

 
Your close friends say Regina is now the driving force behind Dr. Chiluba, they say you are a very intelligent young lady and without you, Dr. Chiluba's current difficulties could have worsened. How true is that?

Well, I must first thank those friends of mine who think so highly of me. I am very grateful and I feel very humbled. But to say that I am the driving force behind Dr. Chiluba I think that's not true. You see in marriage, you complement one another, you don't drive each other. You can drive one into the ditch, you can drive one wrongly.

And you see, between me and my husband, I am a little bit more emotional. He is very restrained. He is very restrained to his credit and that is something I am struggling to learn from him. He is of a very sober character. He is one person who will not react almost to everything. He will be there, this morning me I pick up a newspaper almost flicking but he says no don't worry those are just papers, leave them.

He has been there for me and I am there for him. We are there for each other. Even you, can you imagine if you are going through these problems this man is going through and you are alone? I mean, you need somebody to share with, you need somebody to talk to. You need to be with somebody, so we have been there for each other. But being a driving force, no.

Talk about your political career. Have you also retired from active politics like Dr. Chiluba? And how, when and why did you join politics? 

Well, my husband has retired and I am resting with him. We are resting at home.

What is your relationship with the former first lady Vera? 
I have never really known her and there has never been a time when we have ever met to like exchange words or anything, no. And I never knew her before and I have never known her now. But I have no ill feelings towards her or any ill will, no.

Have you accepted Dr. Chiluba's (nine) children, I know you talked about that but there is Castro who is generally perceived by the public to be a lumpen? 
(She laughs) I don't take kindly to calling my son that, a lumpen. Actually, Castro is a very good friend of mine. Two weeks ago he was here with us for a week and he is a very, very intelligent young man. If you sit down with him, he has very constructive conversations, you will be very surprised. He is extremely intelligent and we get on very well.

And the rest of the children? 

Oh well, I get on with them well. They come here quite often. But since you particularly mentioned Castro, I had to expand on that.

What about Dr. Chiluba, has he accepted the children from your previous marriage? 
Yes, very much. But I can't answer for him. The kids are actually here, most of them are in boarding. They just closed schools but they are free to either stay with their father or here, we don't have any problem. They just went out to the shops but they are here with us.

You used to go round with Dr. Chiluba on campaign trails, how did Mr. Mwanza take it? Wasn't this a source of problems in your marriage?
In fact, I never used to go round with Dr. Chiluba on the campaign trails. If you remember, during the time of the third term, that's when we had provincial conferences and during those provincial conferences that's the time you campaign for your position at the convention because after the provincial conferences then you go to the convention.

So I was gunning for the national chairperson, that's how come I was found in all those provincial conferences. I was campaigning and I had my own program. Dr. Chiluba had his own program, he had to open the conferences. So it's not like I went round with him. I didn't even go in the same car with him or did I travel in that motorcade.

He used to find us there because we could go there a week to campaign. I was campaigning for myself. And as for Mr. Mwanza, to his credit, Mr. Mwanza actually used to encourage me a lot in my political career. In that area, he was a good husband. So we didn't have any problems.

So much has been written and talked about your relationship with Dr. Chiluba, just how did you take some of the negative publicity about your relationship? And how did your families react to such publicity? 

I will be very honest with you because I am human and definitely negative publicity will affect one somehow, it doesn't matter which way. And sometimes I was affected badly, sometimes I laughed it off. But as a public figure you expect things like that. You know, it's so strange because I remember when your paper carried a story about me being in China with Dr. Chiluba and [Sheikh] Chifuwe had come to Ndola during the time Dr. Chiluba was in China and he spoke to me. So when the story came out I said this is stranger than fiction because you found me and I spoke to you and you knew I did not go to China, so why did you write that? He said no, no, the story is from the courts (Chifuwe was not the author of that story). I said some of these things you would know that they are lying because that was pure lies.

But if you followed that story, there was a ZRP official who said at some campaign function that Dr. Chiluba was a womaniser who was flirting with a married woman Mrs. Mwanza and he deliberately included her in his entourage to China. And that gentleman appeared in court, that is where we picked the story from and reported. So Chifuwe was right with his explanation.

You reported and your reporting was okay, but your headline was bad. It was terrible and it affected my children. They were in school then and they knew I hadn't gone to China and the way you wrote it, oh. That's why I called Chifuwe because I was a bit surprised knowing that I had spoken to him and he knew that I didn't go, I was at home because he kept calling me oh, this or that and then to come and see the headline like that, surely there must be some...I don't know how to put it.

But the point is we didn't manufacture that, someone said those words and he appeared in court so we had the duty to inform members of the public on what was going on, don't you agree with me?

But you took advantage and used that to sell the paper. You certainly used that to sell the paper. Even if the story was coming from court, was it a big issue when one of you knew that it was not true? So why make it such a big issue? There was no need really, absolutely no need.

Okay, away from that, what are you doing currently? 
You know my husband loves reading, he loves reading such that sometimes when I just came here I used to feel like he just probably didn't want to talk to me because he would read sometimes the whole day. He would have this book finish, take the other book. So I made a deliberate decision to go back to school. I am currently studying privately with some university in London through correspondence but I won't tell you which subject I am studying until I write the exams. These are some of the things I am learning from my husband. He loves praying, he is very very prayerful. I love praying too. That's what has seen us through all those problems we are in. But we are hopeful and prayerful that all these things will go away.

Anything you would want to say? 
The other thing, a lot of people have wrong ideas about me wanting to marry Dr. Chiluba maybe for money or for whatever. You see, all my life I have been very hard working.

When I was working for Zamtel in Ndola, I started my business on the side. I worked for Zamtel until 1989. Even at the time I was working for Zamtel, I was doing some business on the side. I was buying groceries and cooking oil. You know that time, there were shortages of essential commodities. There were terrible shortages during those times of the vigilantes and I am sure you remember, I don't know whether you were grown up by then.

We talked about joining politics and that's what drove me to get into politics because one of my friends lost a sister who died whilst on the line. The girl collapsed because she was in that queue close to two days queuing up for cooking oil or bread, I can't remember correctly.

It became so bad and when I started business like buying and selling, I would go to Swaziland bringing shoes to sell to Indians. I would flood almost the whole Copperbelt with shoes.

Just going to Swaziland, you had to hide money I don't know where, I cannot explain but it is very embarrassing. And all those things just used to eat me up. And I said to myself 'when are we going to have change in this country?'.

So when the Movement for Multi-party Democracy started, I registered just as an ordinary member in Mitengo branch until 1993 when a situation arose. This was when we didn't have water for close to a month in that area.

I just woke up one day about 04:00 hours in the morning, fed up with the whole situation and I went to almost every door to wake up every woman.

I said this is not the way we should live. We must voice up and tell these people there is no way we are going to keep our homes like this. At that time our member of parliament Hon. Silwamba was still in Ndola. By 06:00 hours, we marched to Hon. Silwamba's home with those containers because I felt that since he was the MP he could have water in his house.

So we went and said we haven't had water for sometime and to his credit, he just looked at us and he said this is what I have been singing about to these people at the council so let's go to Chenda's home (the then town clerk).

And we decided to walk with him. We were very happy that our MP was able to join us, we walked to Chenda's home and he called the councillor. The councillor then was Dorothy Kazunga who was also very happy that we did what we did because she had been pushing for that. She had gone to almost every office. Our friends at the council sometimes would relax when they had water in their homes.

So we went to Chenda's home and arranged for a meeting with the Director of Water Works and after about two days, our MP and the councillor pushed and we had water. So Mrs. Kazunga was impressed with the way I organised the women and she encouraged me to stand for a position at the constituency. Remember I was at the branch level then. So I stood and I won. That's how I started rising. I rose from the branch to the constituency. I never went to the district. I just went straight to the province as a youth leader. After that, that's when I became MMD national chairperson for women's affairs.

From 1989, I have been in business. I was selling shoes and I was importing parts for motor vehicles from that time and then I decided to go into workers' cafeterias or canteens.

In 1990, I opened up a cafeteria at Contract Haulage and there you have captive market so you get cheques at the end of the month. I kept that and expanded when I opened another one at Drum and Can. I later opened another one at TAZAMA, from there I opened one at Zamtel. And at Zamtel, we were preparing five meals per day.

The latest one was the one I had at Bank of Zambia. At Zamtel, we would net almost K40 million per month. So surely, I could feed myself and I could do all that I wanted to do because from all those canteens we were getting cheques at the end of the month.

I have a shop where I sell clothes. All these clothes you are seeing on these pictures, I have them in the shop. I had a salon as well. And then I decided to venture into farming. Right now, I rear chickens.

Is it true that before you were married to Dr. Chiluba, Castro would walk into your boutique and pick one or two items saying 'my father would pay'? 

No, it has never happened. Speculation, again.

Q: Some people say Mr. Mwanza never made any noise about you, in fact he surrendered you to Dr. Chiluba because Dr. Chiluba looked after him very well, he bought him vehicles and gave him a lot of money just to shut him up. How true is that? 

Speculation, again. Oh, my God. You know, my previous husband has been driving vehicles from work. The RAV4 which they thought he (Chiluba) bought is for ZNOC (Zambia National Oil Company) which he drives up to now. And he bought himself a Lexus, you know these vehicles which are advertised on the internet, those for US $3,000 or US $4,000 which almost everyone can afford. He bought that and as a family from our businesses we bought the red RAV4 which is of course with the Task Force now. And we bought a few vehicles but second hand vehicles. Like some of the cars that they (Task Force) took, they had to tow them because they are old. But to think that Mr. Mwanza just gave me away is wrong. Which man would do that? Tell me, even for a million dollars which man would give away a wife just like that? Would you if you were paid? Tell me. Which man? No, we just had our own differences actually. We had our own differences and our parents came together and we discussed. The problems had been going on for a long time so we just could not go on. It was affecting the children but it had nothing to do with Dr. Chiluba, absolutely nothing.

How would you describe Dr. Chiluba as a leader and Dr. Chiluba as a husband? 
Dr. Chiluba as a leader is a very patient man, very restrained like I said. He is not one that would react harshly to anything. Of course he is human, I cannot say there are no times when he would get upset. But I never really worked that closely with him like in Cabinet apart from him being my party leader. As a husband, he is a very caring person. He would always ask what you need, what you would want to do that day. Of course he has his own hobbies and I have mine. We are trying to marry the hobbies together. I never used to watch football but now I love football because of him. He loves watching football over the weekend. And he loves reading. There are certain films that I like watching. When I came here, he was a bit sceptical about that. He would say 'oh, no, no, let's not watch that film.' You know these little soaps like Generations, Isidingo, he doesn't like those things. He says ah, that's childish but now he has accepted, he would watch with me and we would laugh. Generally, he is a very nice person and I must say that I am blessed to be his wife because first and foremost he is very prayerful and very forgiving that sometimes I say, eh, this is too much. But he says no. He is very forgiving, that's my husband. It's good for all of us and I have to learn to get to that. He is forgiving and very giving. He is one person that wants to give all the time. He wants to give of himself, he wants to give of his time, he wants to give of anything that he has. That's him. Those are very rare qualities that he has. Of course, he is human he could have his own weaknesses.



COPYRIGHT 2002 Financial Times Ltd.
(From The Post (Zambia) - AAGM)

Byline: Amos Malupenga

Wednesday 14 July 2010

Fantastic article on Mobile Phones and brain Tumors

Brain tumour risk in relation to mobile telephone use: results of the INTERPHONE international case–control study

The INTERPHONE Study Group*Corresponding author. Elisabeth Cardis; CREAL, Doctor Aiguader 88, 08003 Barcelona, Spain. E-mail: ecardis@creal.cat


    Abstract
 Top Abstract
 Introduction Methods Results Discussion Conclusion Supplementary data Funding Appendix Acknowledgements References

Background The rapid increase in mobile telephone use has generated concern about possible health risks related to radiofrequency electromagnetic fields from this technology.
Methods An interview-based case–control study with 2708 glioma and 2409 meningioma cases and matched controls was conducted in 13 countries using a common protocol.
Results A reduced odds ratio (OR) related to ever having been a regular mobile phone user was seen for glioma [OR 0.81; 95% confidence interval (CI) 0.70–0.94] and meningioma (OR 0.79; 95% CI 0.68–0.91), possibly reflecting participation bias or other methodological limitations. No elevated OR was observed ≥10 years after first phone use (glioma: OR 0.98; 95% CI 0.76–1.26; meningioma: OR 0.83; 95% CI 0.61–1.14). ORs were <1.0 for all deciles of lifetime number of phone calls and nine deciles of cumulative call time. In the 10th decile of recalled cumulative call time, ≥1640 h, the OR was 1.40 (95% CI 1.03–1.89) for glioma, and 1.15 (95% CI 0.81–1.62) for meningioma; but there are implausible values of reported use in this group. ORs for glioma tended to be greater in the temporal lobe than in other lobes of the brain, but the CIs around the lobe-specific estimates were wide. ORs for glioma tended to be greater in subjects who reported usual phone use on the same side of the head as their tumour than on the opposite side.
Conclusions Overall, no increase in risk of glioma or meningioma was observed with use of mobile phones. There were suggestions of an increased risk of glioma at the highest exposure levels, but biases and error prevent a causal interpretation. The possible effects of long-term heavy use of mobile phones require further investigation.


Keywords Brain tumours, mobile phones, radiofrequency fieldsAccepted 8 March 2010

    Introduction
 Top Abstract Introduction
 Methods Results Discussion Conclusion Supplementary data Funding Appendix Acknowledgements References

Mobile phone use has increased dramatically in many countries since its introduction in the early-to-mid 1980s. The expanding use of this technology has been accompanied by concerns abouthealth and safety. In the late 1990s, several expert groups critically reviewed the evidence on health effects of low-level exposure to radiofrequency (RF) electromagnetic fields, andrecommended research into the possible adverse health effects of mobile telephony.1–4 As a result, the International Agency for Research on Cancer (IARC) coordinated a feasibility study in 1998 and 1999, which concluded that an international study of the relationship between mobile phone use and brain tumour risk would be feasible and informative.5,6
INTERPHONE was therefore initiated as an international set of case–control studies focussing on four types of tumours in tissues that most absorb RF energy emitted by mobile phones: tumours of the brain (glioma and meningioma), acoustic nerve (schwannoma) and parotid gland. The objective was to determine whether mobile phone use increases the risk of these tumours and, specifically, whether RF energy emitted by mobile phones is tumourigenic.
This article presents the results of analyses of brain tumour risk in relation to mobile phone use in all INTERPHONE studycentres combined. Analyses of brain tumours in relation to mobile phone use have been reported from a number of cohort7–9 and case–control studies, including several of the national components of INTERPHONE.10–25 No studies, however, have included as many exposed cases, particularly long-term and heavy users of mobile phones, as this study.

    Methods
 Top Abstract Introduction Methods
 Results Discussion Conclusion Supplementary data Funding Appendix Acknowledgements References

Study design
The INTERPHONE study is an international, largely population-based case–control study. The common core study protocol is described in detail elsewhere.5,26 Sixteen study centres from 13 countries (Australia, Canada, Denmark, Finland, France, Germany, Israel, Italy, Japan, New Zealand, Norway, Sweden and the UK) were included. To maximize statistical power, the INTERPHONE study focussed on tumours in younger people, 30–59 years of age, as they were expected to have had the highest prevalence of mobile phone use in the previous 5–10 years, and on regions likely to have the longest and highest use of mobile phones (mainly large urban areas).
Eligible cases were all patients with a glioma or meningioma of the brain diagnosed in the study regions during study periods of 2–4 years between 2000 and 2004. Cases were ascertained from all neurological and neurosurgical facilities in the study regions (except in Paris and Tokyo where some did not agree to participate), and in some centres also from cancer registries. All diagnoses were histologically confirmed or based on unequivocal diagnostic imaging. To facilitate interviews soon after diagnosis, cases were ascertained actively within treatment facilities wherever possible. Completeness of ascertainment was checked through secondary sources, such as population- or hospital-based cancer registries, medical archives and hospital discharge or billing files.26
One control was selected for each case from a locally appropriate population-based sampling frame, except in Germany where two controls were chosen. The sampling procedure involved individual matching in seven centres (Canada – Ottawa, Canada – Vancouver, France, Israel, Japan, New Zealand and UK North) and frequency matching elsewhere. The matching variables were age (within 5 years), sex and region of residence within each study centre. In Israel, the subjects were also matched on ethnic origin. Where stratified matching had been used, individual matching was conductedpost hoc, with cases being assigned one control (two in Germany), interviewed as close as possible in time to the case, from those who fitted the matching criteria.
Detailed information on past mobile phone use was collected during face-to-face interviews with the study subject, or aproxy, if the subject had ever been a regular user of a mobile phone (had an average of at least one call per week for a period of ≥6 months).26 A proxy was sought when the study subject had died or was too ill to be interviewed. The interviews were conducted by a trained interviewer using a computer-assisted questionnaire, except in Finland where a paper version was used. The questionnaire also included sections on socio-demographic factors, occupational exposure to electromagnetic fields and ionizing radiation, medical history (subject’s and family), medical ionizing and non-ionizingradiation exposure and smoking. For cases, information was also collected on the anatomic location and histological type of the tumours. Where possible, location data were obtained from magnetic resonance imaging (MRI) reports or images; they were otherwise obtained from surgical records or clinical notes. Details of the specific source for each case were not recorded in the INTERPHONE database. Those collecting the data did not know the reported mobile phone use of individual cases.
Statistical methods
Data from countries with multiple centres were combined for the analyses, except in the UK where the UK South and UK North, each with large numbers of subjects, were kept separate. The word ‘centre’ in the remainder of this article isused to refer to the 14 analytic entities (12 countries, UK North and UK South). All analyses were carried out for all centres combined and for each centre separately. Formal tests for heterogeneity of risk across centres were conducted by allowing for an interaction between centre and the exposure variables.
The analyses presented here focus on past mobile phone use as reported by or for the study subjects. The main analyses were based on conditional logistic regression for matched sets.27 The date of diagnosis of the case was used as the reference date for cases and controls in each matched set. For the main analyses, the reference category for odds ratios (ORs) was the set of subjects who reported that they had never been regular users. Exposure variables included ever having been a regular user (as defined above), time (years) since first regular use, cumulative number of calls and cumulative duration of calls. To allow for a latency period of 1 year, the year before the reference date was included in the reference category for time since first regular use and all other exposure variables were censored at 1 year before the reference date. Cumulative number and duration of calls were analysed as categorical variables, based on deciles of the distribution of these variables among all controls who were regular users, including those matched to patients with an acoustic neuroma or a parotid gland tumour, so that the same cut-off points are used in all analyses.26 Cumulative use excluded use of mobile phones with hands-free devices: for all time periods for which the subject reported the use of hands-free devices the amount of use was reduced by 100, 75, 50 or 25% depending on whether hands-free deviceswere used always or almost always, more than half, about half or less than half of the time, respectively. For ease of presentation, some results are shown for the following grouping of deciles: 1, 2–5, 6–7, 8–9 and 10, chosen post hoc toreflect the spread of the highly skewed distribution of these variables. For convenience, we will systematically use the term‘regular user’ in text and tables to refer to ever having been a regular user.
The reference group for these analyses, never regular users, included people who had some mobile phone use but never as much as one call a week on average for ≥6 months (~32% of meningioma and 26% of glioma cases, and 30% of meningioma and 26% of glioma controls) and people who had never used a mobile phone (~11% of meningioma and 9% of glioma cases, and 8% of meningioma and 6% of glioma controls). These percentages are approximate because never use and never regular use were defined at different dates; the reference date and the date of interview, respectively. We are not able to determine whether inclusion of subjects with some occasional mobile phone use in the reference group hada material effect on our results because this difference in definition dates prevented us from distinguishing participants with only occasional use from those with no use at all at their reference dates. Moreover, because numbers of never users atthe date of interview were small, particularly in certain age- and gender-specific sub-groups (such as young men), never users were not a suitable reference group for this analysis.
All analyses were adjusted for educational level; an a priori decision had been made to adjust for it as a surrogate for socio-economic status (SES). Creation of consistent educational levels across the 13 countries is described elsewhere.26In practice, this adjustment had little impact on OR estimates, changing their values by ≤2% in most instances and in all cases by <5%. Using a 10% change-in-estimate criterion for confounding,28 no other covariate among those collected (see list above) was included in the main analyses. The interval between the start date of interviews in the study centre and the date of each subject’s interview was modelled by fitting the interaction of this interval with study centre.
A common protocol was applied to impute missing data for cases and controls.26 The study questionnaire allowed ranges to be given instead of exact answers to a number of questions, including number and duration of calls and dates of start and end of mobile phone use; in such instances, the main analyses in this article were based on the mid-point of the reported range.
Because absorption of RF energy from mobile phones is highly localized,29 three different types of analyses were conducted to account for tumour location. First, analyses were conducted by the anatomical lobe of the brain in which the tumour occurred. Secondly, separate analyses were conducted for the subjects who reported using the mobile phone mainly on one or the other side of the head, and the preferred side was compared with the side on which the tumour occurred. For this, each control was assigned the location of the tumour of his or her matched case. Exposure was considered to be ipsilateral if the phone was used predominantly on the same side as the tumour or on both sides of the head, and contralateral if used mainly on the side of the head opposite to the tumour. Laterality was not assigned if the tumour was reported to be centrally located (i.e. it crossed the midline of the brain); these cases were excluded from laterality analyses. Thirdly, case–case analyses were carried out on the concordance between tumour side and laterality of phone use using the method proposed by Inskip and collaborators.18
Sensitivity analyses
To complement these primary analyses, we undertook sensitivity analyses to try to determine whether any of the following might have biased the results: (i) any study centre; (ii) required mention of mobile phones in the introductory letter to subjects in some centres; (iii) centres with a hospital-based design or particularly low participation rates; (iv) respondents whose interviews were considered by the interviewer to be of poor quality; (v) subjects for whom proxies provided the responses or a telephone interview was given; (vi) interviewers who had little experience or who had unbalanced case to control workloads; (vii) difference between the interview dates of cases and their matched controls (on average, each control was interviewed 3 months later than its matched case26 and mobile phone use was increasing rapidly during the study period); (viii) subject’s choice between two ways of responding to call time questions (time per day, week or month, or time per call); (ix) subjects who reported implausibly high amounts of mobile phone use (by excluding them or by retaining them and truncating their use at a specific lower value when they reported a higher one); (x) method of calculating accumulated call time; (xi) use of matching and conditional analysis; (xii) the choice of a particular imputation algorithm; and (xiii) adjustment for possible confounders.

    Results
 Top Abstract Introduction Methods Results
 Discussion Conclusion Supplementary data Funding Appendix Acknowledgements References

During the study period, 3115 meningioma and 4301 glioma cases, and 14 354 potential controls were identified. Interviews were completed with 2425 meningioma cases (78%; range 56–92%),2765 glioma cases (64% participation; range by centre 36–92%) and 7658 controls (53%; range 42–74%; Appendix 1Table 1Supplementary data are available at IJE online). The mostcommon reasons for non-participation were subject refusal (11% of meningiomas, 11% of glioma cases and 30% of controls); illness, death or physician refusal (4% of meningiomas, 20% of gliomas and 1% of controls); and inability to contact the subject (7% of meningiomas, 5% of gliomas and 15% of controls).
The main analyses, based on matched sets only, included 2409 meningioma cases with 2662 matched controls and 2708 glioma cases with 2972 matched controls. Among meningioma cases, 24% were men and 76% women; among glioma cases, 60% were men and 40% women (Table 1). Although the median age of meningioma cases was only slightly older than that of glioma cases (51 and 49 years, respectively), 23% of glioma cases were diagnosed before the age of 40, compared with 13% of meningioma cases.

View this table:
[in this window]
[in a new window]

 
Table 1 Selected characteristics of meningioma and glioma cases included in the main analysesa

The proportion of proxy interviews was higher in glioma cases (13%) than in controls (1%) or meningioma cases (2%). Whereas 17% of glioma cases who were regular users had imputations because of missing information in at least one of their mobile phone-related variables, the corresponding fractions were 9% among regular user meningioma cases and 8% among regular user controls. The proportion of subjects who answered questions about mobile phone use by giving a range of values rather than a particular amount of use (for any of the use dimensions) was very similar (~42%) for meningioma cases, glioma cases and controls.The prevalence of regular mobile phone use 1 year before the reference date was 52% for meningioma cases (ranging from 34 to 73% across study centres) and 56% in matched controls (35–78%). It was higher for glioma cases (62% overall, range: 42–80%) and controls (64% overall, range: 45–84%), reflecting the different sex distributions of these tumours.
The majority of subjects in the study were not heavy mobile phone users; the median lifetime cumulative call time amongmeningioma controls using mobile phones was ~75 h, with a median call time of ~2 h/month and a median lifetime number of calls about 1500. Corresponding values for glioma controls were ~100 h lifetime, 2.5 h/month and about 2000 calls. The distributions of time since start of mobile phone use and cumulative call time were highly skewed, with few long-term and heavy users, and varied across study centres and by age and sex (not shown).
Relation between mobile phone use and risk of brain tumours
Meningioma
A reduced OR of meningioma was found for regular mobile phone use in the past ≥1 year, OR 0.79 [95% confidence interval (CI) 0.68–0.91; Table 2]. There was some suggestion of heterogeneity of risk across centres (P = 0.08) with ORs <1.0 except in Canada, Denmark, Germany and Italy (data not shown). ORs were <1.0 for regular users in all categories of time since start of use and cumulative number of calls. Analyses by cumulative call time showed ORs <1.0 in the first nine deciles and an OR of 1.15 (95% CI 0.81–1.62) in the highest decile. Analyses of cumulative call time among recent-, medium- and long-term users (Table 3) showed no indication of excess risk except in the highest call time category in those who started phone use 1–4 years before the reference date: OR 4.80 (95% CI 1.49–15.4).

View this table:
[in this window]
[in a new window]

 
Table 2 ORs between mobile phone use and brain tumours (meningioma and glioma separately) by regular use, time since start of use, cumulative call time and cumulative number of calls—excludes use with hands-free devices


View this table:
[in this window]
[in a new window]

 
Table 3 ORs between mobile phone use and brain tumours (meningioma and glioma separately) by cumulative call time, stratified by recency of starting regular use—excludes use with hands-free devices

There was no appreciable effect modification by age or sex on any of these results (data not shown).In analyses by anatomical location of the meningioma, the OR for temporal lobe tumours with regular use was 0.55 (95% CI 0.36–0.82) and the ORs were <1.0 in all categories of time since start of use, cumulative call time and cumulativenumber of calls. ORs for other lobes were also mostly <1.0 (Table 4).

View this table:
[in this window]
[in a new window]

 
Table 4 ORs between mobile phone use and brain tumours (meningioma and glioma separately) by anatomical location of tumoura and by regular use, time since start of use, cumulative call time and cumulative number of calls – excludes use with hands-free devices

The OR for mainly ipsilateral use among regular users was 0.86 (95% CI 0.69–1.08), and that for contralateral use was0.59 (95% CI 0.46–0.76; Table 5). The ORs by time since start of use were <1.0 in all categories of ipsilateral andcontralateral use. When analysing by any of the exposure metrics in Table 5, the ratios of the ORs for ipsilateral use to contralateral use were always one or above one regardless of level of exposure and they were highest (~2 or 3) for the two highest categories of cumulative call time and the second highest category of cumulative number of calls. A case–case analysis, based on Inskip’s method, showed an OR of 1.07 (95% CI 1.00–1.16; Appendix 1Table 2Supplementary data are available at IJE online) for ipsilateral use.The OR for those who reported regular use of only an analogue phone was 0.81 (95% CI 0.65–1.03) and for only a digital phone it was 0.79 (95% CI 0.68–0.92). Focussing on the highest decile of cumulative call time, the OR among those who used only an analogue phone was 0.50 (95% CI 0.25–0.99); among those who used only a digital phone it was 1.84 (95% CI 1.17–2.88); and among those using both 4.43 (95% CI 1.42–13.9; Appendix 1Table 3,Supplementary data are available at IJE online).

View this table:
[in this window]
[in a new window]

 
Table 5 ORs between mobile phone use and brain tumours (meningioma and glioma separately) by side of use of mobile phones and by regular use, time since start of use, cumulative call time and cumulative number of callsa

Glioma
A reduced risk of glioma was seen for regular mobile phone use in the past ≥1 year (OR 0.81, 95% CI 0.70–0.94; Table 2) relative to never regular users. There was little evidence of heterogeneity in results across centres (P = 0.68). ORs were <1.0 in all study centres except Australia, France and New Zealand.Analyses by time since start of use showed a reduced OR in all categories of use; the OR for ≥10 years since start of use was 0.98 (95% CI 0.76–1.26; Table 2). The pattern of results by duration of mobile phone use was similar (data not shown).
Analyses by categories of cumulative call time showed decreased ORs in eight of the first nine deciles (five of which had upper confidence bounds <1.0) and an increased OR of 1.40 (95% CI 1.03–1.89) in the highest exposure category, ≥1640 h. Analyses by cumulative number of calls showed ORs <1.0 in all categories, with the OR in the highest decile approaching unity.
Analyses of cumulative call time stratified by short-, medium- and long-term use (Table 3) showed reduced risks in the lower call time categories in all strata of time since start of use and ORs >1.0 in the highest category (≥1640 h cumulative call time) in each of the three strata. The highest OR was in the short-term users but its CI was very wide.
The lobe of the brain in which the tumour was located was known for 96% of cases. The OR for temporal lobe tumours with regular use was 0.86 (95% CI 0.66–1.13; Table 4). The point estimates for the highest categories of cumulative call time, cumulative number of calls and time since start of use were higher for tumours in the temporal lobe than in other locations, but their 95% CIs were wide. Only for the highest decile of cumulative call time was the OR for temporal lobe tumours appreciably elevated (1.87, 95% CI 1.09–3.22).
For regular use in the past ≥1 year, the OR for ipsilateral mobile phone use was 0.84 (95% CI 0.69–1.04) and that for contralateral use was 0.67 (95% CI 0.52–0.87; Table 5). The ORs by time since start of use were <1.0 in all categories, except for ipsilateral use beginning ≥10 in the past (OR 1.21, 95% CI 0.82–1.80). The ORs by cumulative number of calls were <1.0 irrespective of laterality and exposure level, except for ipsilateral use in the two highest categories. The results by cumulative call time showed a similar pattern, but the OR for ipsilateral use in the highest category was appreciably elevated (OR 1.96, 95% CI 1.22–3.16) and that for contralateral use was slightly elevated (OR 1.25, 95% CI 0.64–2.42). The ratios of the ipsilateral ORs to the contralateral ORs were all above one with one exception (0.99 for 2–4 years since start of use) and the highest (~2) were in 1–1.9 and ≥10 years since start of use, the lowest category of cumulative call time, and the highest category of cumulative number of calls. For cumulative number of calls, there was a consistent trend towards increasing ratios with increasing exposure.
Case–case analyses of the concordance between tumour side and preferred side of phone use using the Inskip method showed an elevated risk for ipsilateral use among regular users (OR 1.27, 95% CI 1.19–1.37) and among those in the highest decile of cumulative call time (OR 1.55, 95% CI 1.24–1.99; Appendix 1Table 2Supplementary data are available at IJE online). When stratified on time since first use, the point estimate of the OR using Inskip’s method in the highest decile was higher among short-term heavy users (OR 2.37, 95% CI 0.93–8.59) than among medium (OR 1.40, 95% CI 1.04–2.01) and long-term (OR 1.57, 95% CI 1.13–2.30) heavy users, resembling an analogous pattern in Table 3.
The OR for ever use of an analogue phone was 1.00 (95% CI 0.83–1.21) and for ever use of a digital phone 0.76 (95% CI 0.66–0.88). Increased ORs were seen in the highest decile of cumulative call time with analogue phones (OR 1.95, 95% CI 1.08–3.54) and with digital phones (OR 1.46, 95% CI 0.98–2.17; Appendix 1Table 3Supplementary dataare available at IJE online).
There was no evidence of heterogeneity of effects across centres for cumulative call time, cumulative number of calls, time since start of use or ipsilateral or contralateral use. Nor was there any appreciable effect modification by age or sex in any of the results mentioned above (data not shown).
Sensitivity analyses
Selected findings of sensitivity analyses are shown in Table 6 and Appendix 1Table 4 (Supplementary data are availableat IJE online). Because of a hint of a possible excess risk in subjects in the highest decile of mobile phone cumulative call time, for glioma (OR 1.40) and to a lesser extent for meningioma (OR 1.15), we focus presentation of sensitivity analyses on the findings in this highest decile, corresponding to 1640 or more cumulative hours of use.

View this table:
[in this window]
[in a new window]

 
Table 6 Results of sensitivity analyses on ORs between mobile phone use and brain tumours (meningioma and glioma separately) for the highest decile of cumulative call time, covering possible indicators of sample representativeness and response quality

For meningioma, some point estimates differed from the OR of 1.15 from the main analysis, but the estimates were imprecise and, with one exception based on nine cases and four controls, fell well within the CI of this ‘benchmark’ value.For glioma, results from the various sensitivity analyses were generally similar to those from the primary analysis. All theOR estimates, except one based on nine cases and three controls, are well within the 95% CI of the OR from the main analysis. When subjects with high reported use were included, but with use truncated at 5 h/day, the OR was hardly affected. When subjects who reported >5 h call time/day (38 cases and 22 controls) were excluded altogether, on the premise that such responses were unreliable, the OR decreased to 1.27 (95% CI 0.92–1.75).
Results of sensitivity analyses focusing on the OR for regular use in the past ≥1 year are shown in Appendix 1Table 5(Supplementary data are available at IJE online). All the OR estimates, except two ORs for meningioma relating to the presentation of the study, are well within the 95% CI of the OR from the main analysis.

    Discussion
 Top Abstract Introduction Methods Results Discussion
 Conclusion Supplementary data Funding Appendix Acknowledgements References

The INTERPHONE study is the largest case–control study of mobile phones and brain tumours conducted to date, including the largest numbers of users with at least 10 years of exposure and the greatest cumulative hours of use of any study. An exhaustive analysis of this large data set involved estimation of hundreds of ORs; rather than focus on the most extreme values, the interpretation should rest on the overall balance of evidence. The null hypothesis of no association would be expected to produce an approximately symmetric pattern of negative and positive log ORs. A skewed distribution could be due to a bias or to a true effect. Our results include not only a disproportionately high number of ORs <1, but also a small number of elevated ORs. This could be taken to indicate an underlying lack of association with mobile phone use, systematic bias from one or more sources, a few random but essentially meaningless increased ORs, or a small effect detectable only in a subset of the data.
For meningioma, there is little evidence to counter a global null hypothesis, and we conclude that INTERPHONE finds no signs of an increased risk of meningioma among users of mobile telephones.
For glioma, an increased OR was seen in analyses in the highest decile of cumulative call time, including tumours in the temporal lobe and subjects who reported having used the mobile phone mainly on the same side as where the tumour occurred. Still, the evidence for an increased risk of glioma among the highest users was inconclusive, as the increase could be due to one or more of the possible sources of error discussed below.
In the following sections, we explore possible explanations for the apparently decreased risk of meningioma and glioma for regular users compared with never regular users, and the apparently increased risk of glioma in a subset of users.
Decreased risk with ever regular use of a mobile phone
An apparently decreased risk of brain tumours with ever regular use of a mobile phone (relative to never regular use) has been seen in other studies.18,23 Putting aside a genuine protective effect as implausible, we have considered other reasons for these observations.
Sampling bias
In all but two centres, a population-based design was used. This requires that the cases in the study were representative of all cases in the respective population and that the controls represented all non-cases, within matching strata. In practice, it is difficult to demonstrate that these conditions have been fulfilled in any case–control study. Cases may be missed due to lack of detection, misdiagnosis or incomplete registration (such problems may be more likely for meningioma than for glioma). It is uncertain whether the sampling frames used to select controls represented the study base in some countries. To the extent possible, we conducted sensitivity analyses that examined the effects of different recruitment strategies between centres; they did not show substantial changes in the results (Table 6).
Levels of participation
Constrained by the requirements of ethical review committees and facing the population’s increasing reluctance to participate in interview studies, we attained participation rates of 78% among meningioma cases, 64% among glioma cases and 53% among controls.26 Although such proportions are not unusually low, they raise the possibility of selection bias with respect to mobile phone use.
Controls in 11 centres and cases in 9 centres who refused the full interview were asked to respond to a brief non-respondent questionnaire on mobile phone use. The cases and controls who complied with this short inquiry reported a lower lifetime prevalence of ever regular use of a mobile phone than did respondents to the full interview, implying that information from those who participated in the full interview may overestimate prevalence among all eligible subjects. Because participation and refusal differed between cases and controls, such non-representativeness may have distorted the OR estimates.30 Although caution is required in extrapolating from the findings of the sub-study, we estimated, in the more plausible scenarios, that non-participation bias may have led to a reduction in the ORs for regular use of 5–15%,30which is less than the observed reductions below the null in the ORs in ever regular mobile phone users for meningioma(21%, 95% CI 32–9) and glioma (19%, 95% CI 30–6; Table 2).
Prodromal symptoms
Prodromal symptoms of a brain tumour could dissuade subjects from becoming phone users or reduce their use before diagnosis (reverse causation). Glioma is typically diagnosed quite soon after first symptoms. Although prodromal symptoms might result in lowered ORs among very recent users (e.g. <2 years since starting use), these are unlikely to explain the reduction in ORs observed among the vast majority of the users in our study population who started using mobile phones 2–10 years before disease onset.
Timing of interviews
As the use of mobile phones has become more common over time, the later interview dates of controls could have spuriously increased the prevalence of exposure in the control group. However, restricting analyses to matched sets in which the cases and controls were interviewed within 1 month of each other resulted in very little change in the OR for regular use ≥1 year in the past (Table 6) and hence seems unlikely to explain the low ORs overall. Further, the use of a common reference date for each case and its matched control should have minimized any bias induced by differential timing of interviews.
Confounding
Higher socio-economic status has been associated with a higher risk of brain cancer in some but not all relevant studies,31,32 and with mobile phone use, particularly when the technology was new.9 We adjusted for education level in all analyses, but acknowledge this is an imperfect indicator of SES. Otherwise, there are few well-established risk factors for brain tumours; analyses adjusting for measured potential confounders had little impact on the ORs (Appendix 1Table 4,Supplementary data are available at IJE online).
Low overall risks among mobile phone users
The reduced OR for regular users compared with never regular users seems unlikely to reflect a genuine protective effectand makes our results difficult to interpret.33 It could result from the sources of error discussed above, although based onthe evidence we have regarding their magnitude and effects30,34 they may not account fully for the observed reduction inrisk.
It might be possible to correct, at least crudely, for assumed downwards bias in the ORs for mobile phone use by undertaking a series of analyses using the lowest category of users as the reference category for OR estimates in higher categories. Results of such an analysis of the mobile phone use variables in Table 2 are shown in the Table of Appendix 2 (see Supplementary data available at IJE online), accompanied by a discussion of the strengths and weaknesses of this approach. We have also done some work to characterize possible sources of bias30,34 and are currently exploring the possibility of correcting the OR estimates mathematically for their effects.
Elevated risks of glioma among heavy users
There was some evidence of an elevated risk of glioma in the highest decile of cumulative call time, with the highest pointestimates seen for tumours in the temporal lobe and for subjects who reported having used their mobile phone mainly on the same side as that on which the tumour occurred. We explore here possible interpretations of these findings.
Biases related to possible differential quality of exposure data
When compared with controls, glioma cases had a higher proportion of proxy respondents, a higher number of imputations for missing values, and a higher proportion of subjects judged by their interviewer to be non-responsive or having poor memory (data not shown). However, sensitivity analyses showed that these differences, on their own, did not explain the results seen in the highest decile of cumulative call time (Table 6).
Differential error between cases and controls in reporting of mobile phone use could substantially affect our results; suchinformation bias could arise from several sources. First, a brain tumour, particularly in the frontal or temporal lobes, may adversely affect cognition and memory.35 Secondly, cases may be more motivated to recall and report a publicized potential risk factor for their disease.
To investigate the accuracy of self-reported phone use, two validation sub-studies were conducted in some of the INTERPHONE centres. Amongst healthy volunteers using software-modified phones (recording number and times of calls), phone use in the past year was reported with substantial random error; with over- and under-estimation both frequent.36 Errors were larger for duration of calls than for number of calls, and phone use was under-estimated by light users and over-estimated by heavy users. In another sub-study, records of mobile phone use up to 6 years previously were obtained for some participants in three INTERPHONE centres, allowing us to compare the interview responses with the records.37 Overall, there was little evidence that recall quality differed between cases and controls, but there was some indication of greater over-reporting by cases than by controls for the period 3–5 years before interview. These sub-studies provide no information regarding differential reporting error for periods more distant than 5 years before interview.
Some subjects reported very high daily average call times and this was more common among cases than controls. Thirty-eight cases and 22 controls reported >5 h use/day and 10 cases and no controls reported ≥12 h/day. There is reasonable doubt about the credibility of such reports. Excluding all subjects who reported >5 h use/day reduced the ORs in the highest decile of cumulative time from 1.40 to 1.27 (95% CI 0.92–1.74). In contrast, truncating the average call time to 5 h/day had little effect on the OR. It is not clear which of these two approaches (if either) is more appropriate. However, the key question is whether these cases with unreasonably high values reflect a general tendency for cases to overestimate more than controls, which could contribute to the apparent excess risk in the highest decile. As noted earlier, there is evidence that cases tended to overestimate their past exposure more than controls did.37
Non-differential error (random variability or uncertainty in the exposure estimates) may also affect the findings. With dichotomous exposure indicators such bias is towards the null, but for polytomous variables the effect is difficult to predict.38–40
Location of tumours and laterality of use of phones
Absorption of RF energy from mobile phones is highly localized.29 Thus, an association of phone use with tumours occurring near the location of the phone would constitute stronger evidence for aetiology than an association with more distant tumours.
Ipsilateral ORs were almost always greater than contralateral ORs. There was no consistent pattern with regard to level ofexposure, although a trend towards a stronger effect of ipsilateral use relative to contralateral use with increasing exposure was observed for cumulative number of calls. Results of case–case analyses (using Inskip’s method18) also suggested higher risks of gliomas with ipsilateral phone use, but again no consistent trend with increasing exposure. The observation of an unlikely ipsilateral effect in low exposure categories suggests that cases might have over-reported use on the side of the tumour.
There is, though, evidence of lack of such reporting bias from a sub-study. In three centres (Australia, Canada and Japan), participants (172 glioma and 160 meningioma cases and 340 controls who were regular users) were asked at the end of their interview to put a mobile phone to their ear as if answering a call. The concordance between the reported side of use of the phone and the side where it was held was lower for cases (72% glioma cases, 66% meningioma) than controls (95%). The greater degree of concordance among controls suggests differential reporting quality. Among cases, however, there was as much discrepancy in the contralateral direction (52 instances) as in the ipsilateral direction (48instances). Thus, it is possible that the ipsilateral effect is a true effect, is due to reporting bias or is a mixture of both.
Few studies have related field strength to anatomic structures, but a recent investigation of 110 phone models found that exposure is generally highest in the temporal lobe.29 While laterality analyses may be biased by the respondent’s knowledge of the side of the tumour, results for tumours in different lobes are probably less susceptible to reporting bias. ORs for glioma in the highest exposure categories were higher for tumours in the temporal lobe than in other lobes, but the CIs around the lobe-specific estimates for each measure were wide.
Coherence and consistency
The strongest evidence of an increased risk of glioma was found for cumulative call time, which is a function of the numberand duration of calls. Conceptually, cumulative call time might be the most relevant measure of exposure. However, in validation studies, the number of calls was recalled more accurately than the duration of calls.36,37 For the cumulative number of calls, the ORs, while highest in the highest deciles, were consistently below one. In the absence of a known biological mechanism, it is hard to know whether more weight should be put on results from the more accurate or the conceptually preferred exposure measure.
The apparently increased risk of glioma for cumulative call time was restricted to the top decile, ≥1640 h. There was noupward trend across the first nine deciles of cumulative call time. In contrast with the excess risk seen on the scale ofcumulative call time, risk did not appear to be increased by length of time since first exposure or by duration of exposure.The pattern of point estimates of ORs in the high call time categories in three strata of time since exposure started—3.8 in the most recent and 1.3 in the more distant ones (Table 3)—is not what one would expect if there were a causal association; although the CI in the newest users was wide and encompassed the point estimates for heavy use in the two longer use groups. By analogy with known carcinogens, the lack of a consistently increasing risk with dose, duration of exposure and time since first exposure weigh against cause and effect. Nevertheless, given the uncertainty surrounding possible effects of RF on the brain, no strong case can be made for the plausibility or implausibility of any observed exposure response pattern.
Comparison of meningioma and glioma results
While the ORs for meningioma were lower than that for glioma in high exposure subgroups, there were some similar patterns. First, the OR for all regular users compared with never regular users was very similar. Secondly, there was no trend in relation to cumulative call time except for an elevated OR in the highest decile. Thirdly, the increase in the last decile was more pronounced for cumulative call time than number of calls. Fourthly, the highest OR for cumulative call time was seen among subjects who had recently started regular use. Fifthly, the ORs were greater for ipsilateral than contralateral use and the ratios of ipsilateral ORs divided by their corresponding contralateral ORs were of a similar magnitude. However, while there was evidence of a higher risk of gliomas in the temporal lobe than elsewhere with several different exposure metrics, there was no such evidence for meningioma. Although ORs for meningioma were generally lower than that for glioma, the otherwise similar patterns of associations of mobile phone use with meningioma and glioma could indicate shared aetiology or shared bias.
Interpretation of these findings
We have no certain explanation for the overall reduced risk of brain cancer among mobile phone users in this study, although selection bias is almost certainly a contributor. There is some evidence that very high users experienced excess risk of glioma, but that evidence is inconclusive because of possible bias. Further light may be shed on dose–response relations by work now being undertaken with the INTERPHONE data using precise coordinate localization of tumours within the brain in relation to estimates of absorbed RF energy.
The possibility of raised risk in heavy users of mobile phones is an important issue because of their ever-increasing use.Moreover, few subjects in our study had used mobile phones for >12 years; therefore, our results are uninformative withrespect to lag periods longer than this.
Consistency with previous research
Our results are consistent with most of the research published to date. A large Danish cohort study of mobile telephone subscribers,8,9 with an average follow-up time of 8.5 years, found no increased risk of brain tumours in subscribers of ≥10 years. The first case–control studies conducted included cases diagnosed in the mid-to-late 1990s and therefore could only address possible risks among short-term mobile phone users.10,12,18,23 In addition, the highest cumulative call times in these studies were much less than in ours. Generally, these studies reported ‘negative’ results. In contrast, increased risks of malignant brain tumours at higher levels of accumulated use of analogue and digital mobile phones and cordless desktop phones were reported from a sequence of three case–control studies from the same authors with cases in the last diagnosed as late as 2003.13–15 However, the methods of these studies have been questioned.41
Some of the INTERPHONE centres have published their results for brain tumours11,16,17,19,22,24,25 and two pooled analyses from Northern European centres have also been published.20,21 Most cases in these reports are included in the present analyses and constitute 69% of gliomas and 57% of meningiomas. The centre-specific analyses are consistent with our all-centre results.
Much biological research has been done in recent years on possible biological effects of RF fields. This work covers in vitro and in vivo exposure, alone and in combination with other physical or chemical agents, and has found no evidence that RF fields are carcinogenic in laboratory rodents or cause DNA damage in cells in culture.42 Possible effects of RF fields on other biological endpoints are still being explored.
The possible effects of long-term heavy use of mobile phones on risk of brain tumours require further investigation, givenincreasing mobile phone use, its extension to children and its penetration worldwide. The problems presented by selection and information bias in this and probably other studies suggest that new studies should, in general, only be done if they cansubstantially reduce or eliminate selection bias, obtain detailed and high-quality exposure information over the full period of use and offer sufficient statistical power to detect comparatively small effects in people with heavy or long continued exposure. Monitoring of age- and gender-specific incidence rates may also be valuable, particularly if informed by good longitudinal data on mobile phone use by age and sex, and having regard to features such as brain tumour location that may allow more specific inferences about possible mobile phone use effects.

    Conclusion
 Top Abstract Introduction Methods Results Discussion Conclusion
 Supplementary data Funding Appendix Acknowledgements References

This is the largest study of the risk of brain tumours in relation to mobile phone use conducted to date and it included substantial numbers of subjects who had used mobile phones for ≥10 years.Overall, no increase in risk of either glioma or meningioma was observed in association with use of mobile phones. There were suggestions of an increased risk of glioma, and much less so meningioma, at the highest exposure levels, for ipsilateral exposures and, for glioma, for tumours in the temporal lobe. However, biases and errors limit the strength of the conclusions we can draw from these analyses and prevent a causal interpretation.

    Supplementary data
 Top Abstract Introduction Methods Results Discussion Conclusion Supplementary data
 Funding Appendix Acknowledgements References

Supplementary data are available at IJE online.

    Funding
 Top Abstract Introduction Methods Results Discussion Conclusion Supplementary data Funding
 Appendix Acknowledgements References

This work was supported by funding from the European Fifth Framework Program, ‘Quality of Life and Management of Living Resources’ (contract QLK4-CT-1999901563) and the International Union against Cancer (UICC). The UICC received funds for this purpose from the Mobile Manufacturers’ Forum and GSM Association. Provision of funds to the INTERPHONE study investigators via the UICC was governed by agreements that guaranteed INTERPHONE’scomplete scientific independence. The terms of these agreements are publicly available athttp://www.iarc.fr/en/research-groups/RAD/RCAd.html.
The Australian centre was supported by the Australian National Health and Medical Research Council (EME Grant 219129) with funds originally derived from mobile phone service licence fees; Bruce Armstrong was supported by a University of Sydney Medical Foundation Program Grant and Julianne Brown by an AustralianPostgraduate Award. The Cancer Council NSW and The Cancer Council Victoria provided most of the infrastructure for the project in Australia. The Canada – Montréal study was primarily funded by a grant from the Canadian Institutes of Health Research (project MOP-42525). Additionally, Dr Siemiatycki’s research team was partly funded by the Canada Research Chair programme and by the Guzzo-CRS Chair in Environment and Cancer. Dr Parent had a salary award from the Fonds de la recherche en santé du Québec. The other Canadian centres were supported by a university–industry partnership grant from the Canadian Institutes of Health Research (CIHR), the latter including partial support from the Canadian Wireless Telecommunications Association. The CIHR university–industry partnerships program also includes provisions that ensure complete scientific independence of the investigators. D. Krewski is the NSERC/SSHRC/McLaughlin Chair in Population Health Risk Assessment at the University of Ottawa. The Danish centre was supported by the Danish Cancer Society and the Finnish centre by the Emil Aaltonen Foundation and the Academy of Finland. Additional funding for the study in France was provided by l’Association pour la Recherche sur le Cancer (ARC) (Contrat N°5142) and three network operators (Orange, SFR, Bouygues Télécom). The funds provided by the operators represented 5% of the total cost of the French study and were governed by contracts guaranteeing the complete scientific independence of the investigators. In Germany, additional funds were received from the German Mobile Phone Research Program (Deutsches Mobilfunkforschungsprogramm) of the German Federal Ministry for the Environment, Nuclear Safety, and Nature Protection; the Ministry for the Environment and Traffic of the state of Baden-Württemberg; the Ministry for the Environment of the state of North Rhine-Westphalia; the MAIFOR Program (MainzerForschungsforderungsprogramm) of the University of Mainz. The study conducted in Japan was fully funded by the Ministry of Internal Affairs and Communications of Japan. In New Zealand, funding was provided by the Health Research Council, Hawkes Bay Medical Research Foundation, the Wellington Medical Research Foundation, the Waikato Medical Research Foundation and the Cancer Society of New Zealand. The Swedish centre was also supported by the Swedish Research Council and the Swedish Cancer Society. Additional funding for the UK North and UK South studies wasreceived from the Mobile Telecommunications, Health and Research (MTHR) program, and the UK North study received funding from the Health and Safety Executive, the Department of Health, the UK Network Operators (O2, Orange, T-Mobile, Vodafone, ‘3’) and the Scottish Executive. The Institute of Cancer Research acknowledges National Health Service (NHS) funding to the NIHR Biomedical Research Centre.

    Appendix
 Top Abstract Introduction Methods Results Discussion Conclusion Supplementary data Funding Appendix
 Acknowledgements References

IARC: Cardis E,1,2,* Deltour I,1,3 Vrijheid M,1,2 Combalot E,4 Moissonnier M,1 Tardy H5; Australia: Armstrong B,6 Giles G,7 Brown J6; Canada – Montreal: Siemiatycki J,8 Parent ME,9Nadon L9; Canada – Ottawa/Vancouver: Krewski D,10 McBride ML11; Denmark: Johansen C,3Collatz Christensen H3; Finland: Auvinen A,12,13 Kurttio P,13 Lahkola A,13 Salminen T13; France: Hours M,5 Bernard M,5,14 Montestruq L5,14; Germany: Schüz J,15,3 Berg-Beckhoff G,16Schlehofer B,17 Blettner M15; Israel: Sadetzki S,18,19 Chetrit A,18 Jarus-Hakak A18; Italy: Lagorio S,20 Iavarone I21; Japan: Takebayashi T,22 Yamaguchi N23; New Zealand: Woodward A,24 Cook A,25 Pearce N26; Norway: Tynes T,27,28,29 Blaasaas KG,30,31 Klaeboe L27,29; Sweden: Feychting M,32 Lönn S,33 Ahlbom A32; UK-North: McKinney PA,34 Hepworth SJ,34Muir KR35; and UK-South: Swerdlow AJ,36 Schoemaker MJ36
1Formerly (except M.M.) at International Agency for Research on Cancer, Lyon, France, 2Centre for Research in Environmental Epidemiology (CREAL), Municipal Institute of Medical Research (IMIM), CIBERESP, Barcelona, Spain,3Institute of Cancer Epidemiology, Danish Cancer Society, Copenhagen, Denmark, 4GELARC, CHU Lyon Sud, Pierre Bénite, France, 5Université de Lyon, Institut National de Recherche sur les Transports et leur Sécurité, Institut national de Veille Sanitaire, Unité Mixte de Recherche épidémiologique et de Surveillance Transports Travail Environnement T9405, Lyon, France, 6Sydney School of Public Health, The University of Sydney, Sydney, Australia, 7Cancer Epidemiology Centre, The Cancer Council Victoria, Melbourne, Australia, 8University of Montreal School of Public Health, Montreal, Canada, 9INRS-Institut Armand-Frappier, University of Quebec, Laval, Canada, 10McLaughlin Centre for PopulationHealth Risk Assessment, University of Ottawa, Ottawa, Canada, 11BC Cancer Research Centre, BC Cancer Agency, Vancouver, Canada, 12Tampere School of Public Health, University of Tampere, Tampere, Finland, 13STUK - Radiation and Nuclear Safety Authority, Helsinki, Finland, 14Now at Observatoire Régional de la Santé Rhône-Alpes, Lyon, France, 15Institute of Medical Biostatistics, Epidemiology and Informatics, Johannes Gutenberg-University of Mainz, Mainz, Germany (J.S. – formerly), 16Department of Epidemiology and International Public Health, Faculty of Public Health, University of Bielefeld, Bielefeld, Germany, 17Unit of Environmental Epidemiology, German Cancer ResearchCenter, Heidelberg, Germany, 18Cancer & Radiation Epidemiology Unit, Gertner Institute, Chaim Sheba Medical Center, Tel-Hashomer, Israel, 19Sackler School of Medicine, Tel-Aviv University, Israel, 20National Centre for Epidemiology Surveillance and Health Promotion, National Institute of Health, Rome, Italy, 21Department of Environment and Primary Prevention, National Institute of Health, Rome, Italy, 22Department of Preventive Medicine and Public Health, Keio University School of Medicine, Tokyo, Japan, 23Department of Public Health, Tokyo Women’s Medical University School of Medicine, Tokyo, Japan, 24School of Population Health, University of Auckland, Auckland, New Zealand,25School of Population Health, The University of Western Australia, Perth, Australia, 26Centre for Public Health Research, Massey University, Palmerston North, New Zealand, 27Norwegian Radiation Protection Authority, Østerås(Oesteraas), Norway (T.T. previously, L.K. currently), 28Now at National Institute of Occupational Health, Oslo, Norway, 29The Cancer Registry of Norway, Oslo, Norway, 30Formerly at Norwegian Armed Forces, Sessvollmoen, Norway, 31Now at The Norwegian Financial Services Association (FNH), Oslo, Norway, 32Institute of Environmental Medicine, Karolinska Institutet, Stockholm, Sweden, 33Department of Medical Epidemiology and Biostatistics,Karolinska Institutet, Stockholm, Sweden, 34Centre for Epidemiology and Biostatistics, University of Leeds, Leeds, UK,35The Health Sciences Research Institute, University of Warwick, Coventry, UK and 36Institute of Cancer Research, Sutton, UK.

    Acknowledgements
 Top Abstract Introduction Methods Results Discussion Conclusion Supplementary data Funding Appendix Acknowledgements
 References

The authors are grateful to Dr Christopher P. Wild (Director, IARC, Lyon, France) for guidance provided to the INTERPHONE Study Group in coming to closure on the final draft of this article. Special thanks are also due to Peter Inskip (NCI, Bethesda, MD, USA) and Jorn Olsen (Danish Epidemiology Science Centre, Aarhus, Denmark and UCLA, Los Angeles, CA, USA) for their advice to the Study Group in revising working drafts of the article. The authors would like to thank Lesley Richardson (Montreal, Canada—formerly at IARC) for her major role in the coordination of the study and assistance in drafting the present article and Dr Baruch Modan (Israel—deceased) for his assistance and enthusiasm in the design and setting up of this study. James Doughty performed miracles implementing the CAPI in several languages and several versions, assisted by Roger Parslow. We would also like to thank Jan Ivar Martinsen for additional programming work. Liz Findlay contributed a great deal to the development of materials and training of interviewers. We would also like to thank all the research assistants and interviewers in the different study centres for their efforts to ensure that the study was carried out with care and with due consideration for the participants.
The Australian team would like to acknowledge the support given to study design and implementation by Associate Prof. Michael Besser and Prof. Andrew Kaye and the substantial contributions neurosurgery, neuropathology and other clinical staff made to conduct of the study; and to thank the fieldwork staff in Melbourne—Monique Kilkenny, Georgina Marr, Tracey McPhail, Fiona Phillips, Hayley Shaw, Yvonne Torn-Broers; and Sydney—Matthew Carroll, Sally Dunlop, Virginia MacDonald and Elizabeth Willows—and the many interviewers for their hard work, and the NSW and VictorianCancer Registries for aiding case identification. The Canada – Montreal team acknowledges the diligent work of fieldwork staff including Marie-Claire Goulet, Sylvie Plante, Sally Campbell and the interviewer team. We are grateful to Dr Rafael Glikstein and Dr Geneviève Matte who contributed to the tumour localization efforts. The following hospitals and physicians in Montreal permitted access to their patients: Hôpital Charles-Lemoyne (Dr C. Chaâlala, Dr J. Demers, Dr N. Gauthier, Dr A. Roux). Centre hospitalier de l’Université de Montréal (Dr M.W. Bojanowski, Dr A. Bouthillier, DrJ.-J. Dufour, Dr R.A. Moumdjian). Hôpital Maisonneuve-Rosemont (Dr L.N. Poirier, Dr M. Séguin). Hôpital du Sacré-Coeur (Dr J.-F. Giguère, Dr M.F. Giroux). Jewish General Hospital (Dr M.J. Black, Dr E. Marmor, Dr G. Mohr), Montreal Neurological Institute (Dr R. Del Maestro, Dr A. Olivier, Dr A. Sadikot). The Canada – Ottawa centre gratefully acknowledges the work of the interview team, particularly Lynn Pratt and Daniel Bédard for their leading roles in study coordination; participating clinicians at the Ottawa Hospital included Drs Charles B. Agbi, Brien Benoit, Martin J. Corsten, Vasco F. DaSilva, André Lamothe, Howard J. Lesiuk, William Miller, Paul F. Odell, and David Schramm. The Canada – Vancouver centre wishes to acknowledge the work carried out by Dr Alison Pope, Patricia Nelson, Nelson Ha, Dr Kaushik Bhagat and the interviewer team. The Finnish centre thank Dr J.J. Jääskeläinen (Helsinki University Hospital), Dr S. Valtonen (Turku University Hospital), Prof. J. Koivukangas (Oulu University Hospital), Prof. M. Vapalahti (Kuopio University Hospital), Dr T. Kuurne (Tampere University Hospital) and Prof. R. Sankila (FinnishCancer Registry). We would like to thank the French fieldwork team, Mary-Pierre Herrscher, Fatima Lamri, Agnès Boidart, Hélène Gire, Juliette Krassilchik, Judith Lenti, Delphine Maillac, Frédérique Sonnet, Flore Taguiev, Julie Frantz, France Castay, Florian Gay, for their excellent work; Prof. Doyon (Paris) and Dr Marc Hermier (Lyon) who wereactively involved in the both the development of the methodology for tumour localization and the review of all cases in France; all the hospital services who assisted us in the ascertainment of cases: Lyon – Centre Hospitalier Lyon – Sud (Prof. Dubreuil), Hôpital Neurologique Pierre Wertheimer (Prof. Trouillas, Dr Honnorat, Prof. Confavreux, Dr Achiti, Prof. Fisher, Prof. Vallée, Drs Farsi and Mahla, Prof. Bret, Dr Ricci, Prof. Sindou, Prof. Deruty), Hôpital d’instructiondes Armées Desgenettes (Dr Felten), Centre Léon Bérard (Dr Frappaz), Clinique du Tonkin (Dr de Garassus, Dr Brudon); Paris - Hôpital de La Pitié Salpétrière (Profs Fohanno and Cornu, Dr Lopes, Dr Bloch, Dr Capelle, Dr Duffau, Prof. Delattre, Dr Sanson, Prof. Hauw, Prof. Poirier, Dr Marsault), Hôpital Foch (Prof. Visot, Dr Gaillard, Dr Dupuy, Prof. Chabolle), Hôpital Beaujon (Prof. Sterkers, Dr Bouccara), Hôpital Lariboisière (Prof. Georges, Dr Blanquet, Dr Koot, Prof. Tran Ba Huy), Hôpital Ste Anne (Prof. Roux, Dr Turak), Fondation Rothschild (Dr Mouder, Dr Daguet, Dr Piekarski), Hôpital d’Instruction des Armées du Val de Grâce (Prof. Bequet, Prof. Renard, Prof. Desgeorges) Hôpital St Joseph (Dr Gauthier), Centre Hospitalier intercommunal de Poissy-St Germain en Laye (Dr Cambon), Centre Hospitalier Sud-Francilien (Dr Serre), Centre Hospitalier de Meaux (Dr Améri); Marseille - Hôpital de la Timone (Prof. Peragut, Dr Regis), as well as all those in the Departments of Medical Information and all the hospital personnel, particularly the secretaries and the staff in the medical archives, whose assistance proved essential to the success of the project. The German group wish to thank their team members Dr Eva Münster, Marianne Brömmel, Stephanie Estel, Iris Hettinger, Melanie Kaiser, Katharina Kunna-Grass, Klaus Schlaefer, Dr Jürgen Wahrendorf and Anna Wilms and all the interviewers for their skilful work. They thank the clinical Interphone team for their support and collaboration [Bielefeld:Prof Falk Oppel (Neurosurgical clinic), Dr Uwe Dietrich (Neuroradiology), Dr Volkmar Hans (Neuropathology), Heidelberg: Prof. Andreas Unterberg, Prof. Stefan Kunze, Dr Karsten Geletneky (Neurosurgical clinic), Prof. Klaus Sator, Dr Jochen Fiebach (Neuroradiology), Prof. Marika Kiessling (Neuropathology), Mannheim: Prof. PeterSchmiedek, Dr Jochen Tüttenberg (Neurosurgical clinic), Prof. Christoph Groden, Dino Podlesek (Neuroradiology), Prof. Uwe Bleyl, Dr Rainer Grobholz (Neuropathology), Mainz: Prof. Nico Hopf, Dr Dorothee Koch (Neurosurgical clinic), Prof. Wolf Mann, Prof. Nickalaos Marangos (ENT clinic), Dr Wibke Müller-Forell (Neuroradiology), Prof. Hans Hilmar Göbel (Neuropathology)]; dedicated to the memory of Prof. Axel Perneczky (Neurosurgical clinic). The Israeli centre wishes to acknowledge the following neurosurgeons for the help they provided in patients recruitment and ascertainment : Dr Avi Cohen (Soroka University Medical Center), Prof. Moshe Hadani (Chaim Sheba Medical Center), Prof. Zvi Ram (Tel-Aviv Medical Center), Prof. Zvi Harry Rappaport (Rabin Medical Center), Dr Sigmund Rothman (Assaf Harofeh Medical Center), Prof. Felix Umansky (Hadassah Hebrew University Medical Center), late Prof. George Vaaknin (Tel-Aviv Medical Center), Dr Uriel Wald (Assuta Hospital) and Prof. Menashe Zaaroor (Rambam Health Care Campus). We are grateful to Dr Chen Hoffmann and Dr Dvora Nass (Chaim Sheba Medical Center) who contributed totumour localization and the review of cases. We acknowledge the diligent work of the fieldwork and office staff includingEtti Aviezer, Tehila Ben-Tal, Meirav Dolev, Yonit Deutch, Tamara Rodkin, Ahuva Zultan and the interviewer team. The Italian team (including Prof. Bruno Jandolo, Prof. Paolo Vecchia, Dr Stefano Martini, Dr Emanuela Rastelli, Dr Antonello Vidiri, Dr Rita Basili, Dr Caterina Carnovale Scalzo, Dr Edvina Galiè, Eng. Lucia Ardoino, Eng. Enrica Barbieri, Dr Cristiano Tesei, Dr Rossella Rossi and Massimo Lucibello) dedicates this article to the memory of Prof. Emanuele Occhipinti, and wishes to thank all the neurosurgeons, ENT-surgeons, neuroradiologists, pathologists and health managers contributing to the study: Prof. Umberto Agrillo, Dr Amalia Allocca, Dr Mostafà Amini, Dr Cinzia Bernardi, Dr M. Bonamini, Dr Loredana Bove, Prof. Luigi Bozzao, Dr Alessandro Bozzao, Dr Mario Braga, Dr Fabrizio Breccia, DrVelia Bruno, Dr Andrea Brunori, Dr Antonella Buffoni, Prof. Arnaldo Capelli, Prof. Giampaolo Cantore, Prof. Natale Cantucci, Dr Emanuela Caroli, Prof. Cosimo Cassano, Dr Alessandra Castelnuovo, Dr Costanza Cavuto, Prof. Lucia Cecconi, Dr Franco Cerquetani, Dr Carla Colacecchi, Dr Antonio Comberiati, Dr Valeria D’Alfonso, Dr Giovanni De Angelis, Dr Luca de Campora, Prof. Roberto Delfini, Dr Carlo Della Rocca, Prof. Marco De Vincentiis, Dr Domenica Di Stefano, Prof. Stefano Esposito, Prof. Alfredo Fabiano, Dr Francesco Federico, Prof. Luigi Ferrante, Dr Anna Rita Fetoni, Dr Letizia Feudi, Prof. Roberto Filipo, Prof. Roberto Floris, Prof. Felice Giangaspero, Dr Renato Gigli, Dr Marco Giordano, Prof. Gianfranco Gualdi, Prof. G. Guglielmi, Dr Massimo Iachetti, Prof. Giorgio Iannetti, Dr Maria Rosaria Limiti, Prof. Giulio Maira, Dr Valentina Manciocco, Dr Annunziato Mangiola, Dr Ferdinando Marandino, Dr Luisa Marangoni, Prof. Pasquale Marano, Prof. Maria Enrica Martini Neri, Dr Luciano Mastronardi, Dr Arianna Mattioni, Prof. Maurizio Maurizi, Dr Maria Concetta Mazzeo, Dr Giuseppe Natali, Dr Gaetano Nostro, Prof. Antonio Orlacchio, Prof. Augusto Orlandi, Prof. Fabrizio Ottaviani, Dr Salvatore Passafaro, Dr Francesco Saverio Pastore, Dr Laura Pennesi, DrClaudio Maria Pianura, Prof. Roberto Pisa, Dr Chimene Pistolesi, Prof. Giuseppe Poladas, Dr Siavash Rahimi, Prof. Antonio Ricci, Dr Giovanna Ricci, Dr P. Rigotti, Dr Massimo Rimatori, Dr Rossana Romani, Prof. Giuseppe Santeusanio, Dr Sergio Santilli, Dr Marco Scarpinati, Dr Lauro Sciannamea, Prof. Luigi Sinibaldi, Prof. Giuseppe Spriano, Dr Maurizio Giovanni Vigili, Dr Massimo Volpe. Moreover, the Italian team is grateful to Dr Francesco Forastiere, Daniela D’Ippoliti and Stefania Palange (Epidemiologic Unit ASL RME) for their support in case ascertainment from secondary sources and control selection. The collaboration of the Italian mobile phone network operators in providing traffic data for the exposure validation studies is acknowledged. The New Zealand study team acknowledge the assistance and support of the following: the neurosurgeons and support staff at the neurosurgical units at Auckland Hospital (headed by Mr Edward Mee), Wellington Hospital (headed by Mr Martin Hunn) and Christchurch Hospital (headed by Mr Martin MacFarlane); the staff at the medical record departments at Auckland Hospital, Wellington Hospital and Christchurch Hospital; the staff at the New Zealand Health Information Service and the New Zealand Cancer Registry; Mr Martin Gledhill at the NationalRadiation Laboratory; and, the regional coordinators for the study, Ms Cara Marshall, Ms Sue Hawkins and Ms Janfrey Doak. The Swedish centre thanks the Swedish Regional Cancer Registries and the hospital staff; especially the following key persons at the hospitals: Dr J. Boethius, Dr O. Flodmark, Prof. I. Langmoen, Dr A. Lilja, Dr T. Mathiesen, Dr I. Olsson Lindblom and Dr H. Stibler (Karolinska University Hospital), Dr J. Lycke, Dr A. Michanek and Prof. L. Pellettieri (Sahlgrenska University Hospital), Prof. T. Möller and Prof. L. Salford (Lund University Hospital), Dr T. Bergenheim, Dr L. Damber, Prof. R. Henriksson and Dr B. Malmer (Umeå University Hospital). Professor Swerdlow’s team in the UK South included D. Hogben, A. Butlin, J. Owens, A. Hart, R. Knight, C. Parsley, M. Pelerin, K. Sampson, M. Snigorskaand M. Swanwick. The UK South centre thanks Prof. H. Møller, Mr B. Plewa and Mr S. Richards from the Thames Cancer Registry and the following neuropathologists, neurosurgeons, neuro-oncologists, clinical oncologists, neurologists, other health care staff, administrators and secretaries for the help they provided: Mr D.G. Hardy, Mr P.J. Kilpatrick, Mr R. Macfarlane (Addenbrooke’s Hospital); Ms M. Cronin, Ms T. Foster, Ms S. Furey, Dr M.G. Glaser, Ms F. Jones, Mr N.D. Mendoza, Prof. E.S. Newlands, Mr K.S. O'Neill, Mr D. Peterson, Ms F. Taylor, Prof. J. van Dellon (Charing Cross Hospital); Dr J.J. Bending (Eastbourne District Hospital); Mr P.R. Bullock, Mr C. Chandler, Mr B. Chitnavis, Mr L. Doey, Mr R.W. Gullan, Prof. C.E. Polkey, Mr R. Selway, Mr M.M. Sharr, Ms L. Smith, Prof. A.J. Strong, Mr N. Thomas (King’s College Hospital); Dr G.M. Sadler (Maidstone Hospital); Dr S. Short (Mount Vernon Hospital); Prof. S. Brandner, Mr G. Brookes, Mr A.D. Cheesman, Prof. M.J. Gleeson, Ms J.P. Grieve, Mr W.J. Harkness, Dr R. Kapoor, Mr N.D. Kitchen, Mrs T. Pearce, Mr M.P. Powell, Dr J. Rees, Prof. F. Scaravilli, Prof. D.T. Thomas, Mr L.D. Watkins (National Hospital for Neurology and Neurosurgery); Mr A.R. Aspoas, Mr S. Bavetta, Mr J.C. Benjamin, Mr K.M. David, Mr J.R. Pollock, Dr E. Sims (Oldchurch Hospital); Mrs J. Armstrong, Mr J. Akinwunmi, Mr G. Critchley, Mr L. Gunasekera, Mr C. Hardwidge, Mr J.S. Norris, Dr P.E. Rose, Mr P.H. Walter, Mr P.J. Ward, Dr M. Wilkins (Princess Royal Hospital); Prof. T.Z. Aziz, Prof. D. Kerr, Mr P.J. Teddy (Radcliffe Infirmary); Ms M. Allen, Ms T. Dale, Mr R. Bradford, Dr C. Collis, Prof. A.P. Dhillon, Mr N.L. Dorward, Ms D. Farraday-Browne, Dr D.J. McLaughlin, Mr R.S. Maurice-Williams, Dr K. Pigott, Ms B. Reynolds, Ms C. Shah, Mr C. Shieff, Dr E.M. Wilson (Royal Free Hospital); Mr F. Afshar, Mr H.E. Ellamushi, Prof. P.M. Richardson, Mr H.I. Sabin, Mr J. Wadley (Royal London Hospital); Prof. M. Brada, Dr F.H. Saran, Mrs D. Traish, Mr D. Guerrero (Royal Marsden Hospital); Dr S. Whitaker (Royal Surrey County Hospital); Dr P.N. Plowman (St Bartholomew’s Hospital); Mrs Carole Bramwell, Prof. A. Bell, Mr F. Johnston, Mr H. Marsh, Mr A. Martin, Mr P.S. Minhas, Miss A. Moore, Mr S. Stapleton, Dr S. Wilson (St George’s Hospital); Dr R.P. Beaney (St Thomas’ Hospital). The UK North centre wishes to acknowledge the support of the following neuropathologists, neuroradiologists, neurosurgeons, neuro-oncologists, clinical oncologists, neurologists, specialist nurses and administrators based in hospitals located in Scotland (Mr Barlow, Prof. I. Bone, Ms J. Brown, Mr J. Crowther, MissR. Dolan, Mr Dunn, Mr M.O. Fitzpatrick, Mrs M. Fraser, Dr R. Grant, Dr A. Gregor, Mr Johnstone, Mr Lyndsay, Mrs S. Macnamara, Miss J. Mair, Mr R. Mills, Miss Myles, Mr B. O’Reilly, Mr V. Papanastassiou, Prof. R. Rampling, Mr Russell, Mr D. Sim, Mr P. Statham, Mr Steers, Mr Taylor, Prof. Teasdale, Prof. I. Whittle), west Midlands (Dr J.M. Anderson, Dr Barbour, Dr C.R. Barraclough, Dr P. Bennett, Dr H.G. Boddie, Mr Brind, Dr Carey, Mr M. Choksey, Mr M. Christie, Dr R.N. Corston, Prof. G.S. Cruickshank, Dr A. Detta, Mr P. Dias, Dr S.J. Ellis, Mr G. Flint, Dr D.A.Francis, Mr A.H. Grubneac, Mr S.P. Harland, Dr C. Hawkins, Dr T. Heafield, Dr R.C. Hughes, Dr D.G. Jamieson, Dr A. Logan, Mr C.H.A. Meyer, Mrs R. Mitchell, Prof. K. Morrison, Dr P. Newman, Dr D. Nicholl, Dr S. Nightingale, Dr H.S. Pall, Mr J.R. Ponsford, Dr A. Shehu, Mr Singh, Dr J.A. Spillane, Mr P. Stanworth, Dr B. Summers, Mr A.R. Walsh, Mr J. Wasserberg, Prof. A.C. Williams, Dr J. Winer, Mr S. Zygmunt), Trent (Dr R.J. Abbott, Ms Sheila Adams, Mr Ashpole, Mr R.D.E. Battersby, Prof. L. Blumhardt, Mr P. Byrne, Miss M. Cartmil, Dr S.C. Coley, Dr P. Critchley,Dr Faraj, Dr A. Gibson, Dr P. Griffiths, Dr R. Grunwald, Dr T.J. Hodgson, Mr D.T. Hope, Dr S. Howell, Dr D. Jefferson, Mr D. Jellinek, Dr N. Jordan, Mr A. Kemeny, Dr M.C. Lawden, Prof. J. Lowe, Dr N. Messios, Ms Kirsty Pardoe, Dr S. Price, Dr I.F. Pye, Mr M. Radatz, Mr I. Robson, Dr K. Robinson, Dr C. Romanowski, Dr G. Sawle, Dr B. Sharrock, Prof. P. Shaw, Dr C. Smith, Dr W. Temperley, Dr G. Venables, Mr B. White, Mr A.M. Whiteley, Dr Wills) and West Yorkshire (Dr Al-Din, Dr D. Ash, Dr J. Bamford, Dr M. Bond, Dr G. Bonsor, Dr L. Bridges, Dr B. Carey, Dr Chakrabarty, Mr P. Chumas, Dr D. Dafalla, Dr H. Ford, Dr Gerrard, Dr Goulding, Dr J. Howe, Dr S. Jamieson, Dr Johnson, Dr Louizou, Mr P. Marks, Dr M. Nelson, Dr S. Omer, Mr N. Phillips, Mr S. Ross, Dr I.Rothwell, Dr H. Spokes, Dr J. Straiton, Mr G. Towns, Mr A. Tyagi, Mr P. Vanhille, Dr M. Busby). The views expressed in the publication are those of the authors and not necessarily of the funders.
Conflicts of interest: The following potential conflict of interests have been declared. The Canadian Wireless Telecommunications Association provide technical support in the Interphone study in Canada by providing access to cellular telephone billing records from Interphone subjects, data on power output levels from base stations, and equipment used by the Ottawa-based research team in measuring power output levels from cellular telephones. CTWA had no involvement with the design or conduct of the Interphone study itself.
Professor Armstrong’s travel expenses to give an invited lecture were paid by Australian Centre for Radiofrequency Bioeffects Research, which identifies Telstra Australia as a participating institution. Ms Brown currently owns 426 Telstra shares worth $1452.66 and her husband owns 852 shares worth $2905.32 as of the 10th December 2009.

KEY MESSAGE
  • INTERPHONE is the largest case–control study of mobile phone use and brain tumours yet and includes the largest numbers of users with at least 10 years of exposure. A reduced OR for glioma and meningioma related to ever having been a regular mobile phone user possibly reflects participation bias or other methodological limitations. No elevated OR for glioma or meningioma was observed ≥10 years after first phone use. There were suggestions of an increased risk of glioma, and much less so meningioma, in the highest decile of cumulative call time, in subjects who reported usual phone use on the same side of the head as their tumour and, for glioma, for tumours in the temporal lobe. Biases and errors limit the strength of the conclusions that can be drawn from these analyses and prevent a causal interpretation.



    Notes

*List of members of this study group is available in the Appendix. Back

    References
 Top Abstract Introduction Methods Results Discussion Conclusion Supplementary data Funding Appendix Acknowledgements References

1 Bernhardt JH, Matthes R, Repacholi MH, eds. Non-thermal effects of RF electromagnetic fields. In: Proceedings of the International Seminar on Biological Effects of RF Electromagnetic Fields and Related Health Risks; 1996 Nov 20 (1997) Munich, Germany: International Commission on Non-Ionizing Radiation Protection.
2 McKinlay A. Possible health effects related to the use of radiotelephones - recommendations of a European Commission Expert Group. Radiol Protect Bull (1997) 187:9–16.
3 Repacholi MH. Low-level exposure to radiofrequency electromagnetic fields: health effects and research needs (Review article). Bioelectromagnetics (1998) 19:1–19.[CrossRef][Web of Science][Medline]
4 Royal Society of Canada. A Review of the Potential Health Effects of Radiofrequency Fields from Wireless Telecommunications Devices (1999) Ottawa: Royal Society of Canada.
5 Cardis E, Kilkenny M. International case-control study of cancers of brain and salivary gland - Report of the feasibility study. 99/004. 1999. Lyon: International Agency for Research on Cancer (IARC), IARC Internal Reports.
6 Cardis E, Kilkenny M. International Case-Control Study of Adult Brain, head and neck tumours: results of the feasibility study. Rad Prot Dos (1999) 83:179–83.
7 Dreyer NA, Loughlin JE, Rothman KJ. Cause-specific mortality in cellular telephone users. JAMA (1999) 282:1814–16.[Free Full Text]
8 Johansen C, Boice J Jr, McLaughlin J, Olsen J. Cellular telephones and cancer–a nationwide cohort study in Denmark. J Natl Cancer Inst (2001) 93:203–07.[Abstract/Free Full Text]
9 Schuz J, Jacobsen R, Olsen JH, Boice JD Jr, McLaughlin JK, Johansen C. Cellular telephone use and cancer risk: update of a nationwide Danish cohort. J Natl Cancer Inst (2006) 98:1707–13.[Abstract/Free Full Text]
10 Auvinen A, Hietanen M, Luukkonen R, Koskela RS. Brain tumors and salivary gland cancers among cellular telephone users. Epidemiology (2002) 13:356–59.[CrossRef][Web of Science][Medline]
11 Christensen HC, Schuz J, Kosteljanetz M, et al. Cellular telephones and risk for brain tumors: a population-based, incident case-control study. Neurology (2005) 64:1189–95.[Abstract/Free Full Text]
12 Hardell L, Nasman A, Pahlson A, Hallquist A, Hansson MK. Use of cellular telephones and the risk for brain tumours: a case-control study. Int J Oncol (1999) 15:113–16.[Web of Science][Medline]
13 Hardell L, Carlberg M, Hansson MK. Pooled analysis of two case-control studies on the use of cellular and cordless telephones and the risk of benign brain tumours diagnosed during 1997–2003. Int J Oncol (2006) 28:509–18.[Web of Science][Medline]
14 Hardell L, Carlberg M, Hansson Mild K. Pooled analysis of two case-control studies on use of cellular and cordless telephones and the risk for malignant brain tumours diagnosed in 1997–2003. Int Arch Occup Environ Health (2006) 79:630–39.[CrossRef][Web of Science][Medline]
15 Hardell L, Mild KH, Carlberg M, Soderqvist F. Tumour risk associated with use of cellular telephones or cordless desktop telephones. World J Surg Oncol (2006) 4:74.[CrossRef][Medline]
16 Hepworth SJ, Schoemaker MJ, Muir KR, Swerdlow AJ, van Tongeren MJ, McKinney PA. Mobile phone use and risk of glioma in adults: case-control study. Br Med J (2006) 332:883–87.[Abstract/Free Full Text]
17 Hours M, Bernard M, Montestrucq L, et al. [Cell Phones and Risk of brain and acoustic nerve tumours: the French INTERPHONE case-control study]. Rev Epidemiol Santé Publique (2007) 55:321–32.[Web of Science][Medline]
18 Inskip PD, Tarone RE, Hatch EE, et al. Cellular-telephone use and brain tumors. N Engl J Med (2001) 344:79–86.[Abstract/Free Full Text]
19 Klaeboe L, Blaasaas KG, Tynes T. Use of mobile phones in Norway and risk of intracranial tumours. Eur J Cancer Prev (2007) 16:158–64.[CrossRef][Web of Science][Medline]
20 Lahkola A, Auvinen A, Raitanen J, et al. Mobile phone use and risk of glioma in 5 North European countries. Int J Cancer (2007) 120:1769–75.[CrossRef][Web of Science][Medline]
21 Lahkola A, Salminen T, Raitanen J, et al. Meningioma and mobile phone use – a collaborative case-control study in five North European countries. Int J Epidemiol (2008) 37:1304–13.[Abstract/Free Full Text]
22 Lonn S, Ahlbom A, Hall P, Feychting M. Long-term mobile phone use and brain tumor risk. Am J Epidemiol (2005) 161:526–35.[Abstract/Free Full Text]
23 Muscat JE, Malkin MG, Thompson S, et al. Handheld cellular telephone use and risk of brain cancer. JAMA (2000) 284:3001–07.[Abstract/Free Full Text]
24 Schuz J, Bohler E, Berg G, et al. Cellular phones, cordless phones, and the risks of glioma and meningioma (Interphone study group, Germany). Am J Epidemiol (2006) 163:512–20.[Abstract/Free Full Text]
25 Takebayashi T, Varsier N, Kikuchi Y, et al. Mobile phone use, exposure to radiofrequency electromagnetic field, and brain tumour: a case-control study. Br J Cancer (2008) 98:652–59.[CrossRef][Web of Science][Medline]
26 Cardis E, Richardson L, Deltour I, et al. The INTERPHONE Study: design, epidemiological methods, and description of the study population. Eur J Epidemiol (2007) 22:647–64.[CrossRef][Web of Science][Medline]
27 Breslow NE, Day NE. Statistical methods in cancer research. Volume I - The analysis of case-control studies. IARC Sci Publ (1980) 32:335–38.
28 Greenland S, Mickey RM. The impact of confounder selection criteria on effect estimation. Am J Epidemiol (1989) 130:1066.[Free Full Text]
29 Cardis E, Deltour I, Mann S, et al. Distribution of RF energy emitted by mobile phones in anatomical structures of the brain. Phys Med Biol (2008) 53:2771–83.[CrossRef][Web of Science][Medline]
30 Vrijheid M, Richardson L, Armstrong BK, et al. Quantifying the impact of selection bias caused by nonparticipation in a case-control study of mobile phone use. Ann Epidemiol (2009) 19:33–41.[CrossRef][Web of Science][Medline]
31 Chakrabarti I, Cokburn M, Cozen W, Wang YP, Preston-Martin S. A population-based description of glioblastoma multiforme in Los Angeles County, 1974–1999. Cancer (2005) 104:2798–806.
32 Schmidt LS, Nielsen H, Schmiedel S, Johansen C. Social inequality and incidence of and survival from tumours of the central nervous system in a population-based study in Denmark, 1994–2003. Eur J Cancer (2008) 44:2050–57.[CrossRef][Web of Science][Medline]
33 Saracci R, Pearce N. Commentary: observational studies may conceal a weakly elevated risk under the appearance of consistently reduced risks. Int J Epidemiol (2008) 37:1313–15.[Free Full Text]
34 Vrijheid M, Deltour I, Krewski D, Sanchez M, Cardis E. The effects of recall errors and of selection bias in epidemiologic studies of mobile phone use and cancer risk. J Exp Sci Environ Epidemiol (2006) 16:371–84.[CrossRef]
35 Kolb B, Wishaw IQ. Fundamentals of Neuropsychology (2008) New York: Worth Publishers.
36 Vrijheid M, Cardis E, Armstrong BK, et al. Validation of short-term recall of mobile phone use for the Interphone Study. Occup Environ Med (2006) 63:237–43.[Abstract/Free Full Text]
37 Vrijheid M, Armstrong BK, Bédard D, et al. Recall bias in the assessment of exposure to mobile phones. J Exp Sci Environ Epidemiol (2009) 19:369–81.[CrossRef]
38 Birkett NJ. Effect of nondifferential misclassification on estimates of odds ratios with multiple levels of exposure. Am J Epidemiol (1992) 136:356–62.[Abstract/Free Full Text]
39 Brenner H, Loomis D. Varied forms of bias due to nondifferential error in measuring exposure. Epidemiology (1994) 5:510–17.[Web of Science][Medline]
40 Dosemeci M, Wacholder S, Lubin JH. Does nondifferential misclassification of exposure always bias a true effect toward the null value? Am J Epidemiol (1990) 132:746–48.[Abstract/Free Full Text]
41 Ahlbom A, Feychting M, Green A, et al. Epidemiologic evidence on mobile phones and tumor risk: a review. Epidemiology (2009) 20:639–52.[CrossRef][Web of Science][Medline]
42 SCENIHR. Health Effects of EMF. Scientific Committee on Emerging and Newly Identified Health Risks (SCENIHR), 2009. http://ec.europa.eu/health/ph_risk/committees/04_scenihr/docs/scenihr_o_022.pdf (21 April 2010, date last accessed).