Usually material that has since disappeared from the web

Appendix to House of Commons submission

by Stewart Fist

4.0 Appendix - Royal Adelaide Hospital research

The Royal Adelaide Hospital in South Australia (which has a top-class animal house for research), conducted two parallel studies on EMF exposure between 1993 and 1995. The research design was checked by a committee of the National Health and Medical Research Council (NHMRC) of Australia (the supreme medical research authority) and the hospital had a special committee supposedly oversighting the day to day activities.

The promoter of these two research projects, Dr Michael Repacholi (now in charge of WHO's EMF project in Geneva) sold the idea to the electricity supply organisation and cellphone industry as a way to solve their problems once and for all.

Repacholi is not so much a scientist (he has no research credentials before this), but is well-known as a spokesman and science administrator. He has long been one of the world's best known and most vocal "No Possible Effects" promoters for both low-frequency mains power and cellphones and therefore had the confidence of both the ESAA and Telstra.

The main research project was funded to the tune of $A1 million by the Electricity Supply Association of Australia (ESAA) to look at possible adverse health effects of mains-power exposure, and the second side of the project (run in parallel) was funded (probably $250,000) by Telstra (Australia's dominant carrier) to look specifically at possible effects of GSM digital cellphone exposures.

These studies were conducted in parallel during the 19 month period from August 1993 to Feb 1995. The studies both had control groups of 100 animals, which were treated identically (down to the use of "sham" exposures), and both were double-blind trials where no one knew which autopsied mice had been exposed and which had not until after the diagnosis of cancer had been determined.

4.2 The GSM R/F study:

Six months into the formal phase of the study using GSM radio-frequency exposures of one-hour a day (probably in January and February 1994) the NHMRC's statistician Val Gebski noted a sudden and obvious rise in lymphomas in the exposed mice. Autopsies by Alan Harris at the Walter and Eliza Institute began to show an unexpected increase in B-cell lymphomas—and since this was unknown, and totally unexpected, the slides were sent off to a US research institution in Maryland (USA) for confirmation.

From this point on the scientists had no doubt that this was to be an important and highly significant piece of research. The exposed mice were showing:

  • lymphomas at an earlier stage. The exposed mice had higher rates early (at about nine months), but the rate continued to rise throughout the 18 month study period;
  • initially three to four times the lymphoma rate of the unexposed (later this rate dropped slightly to 2.4 times after some suspect ones were removed);
  • the B-cell lymphomas linked to cancer, rather than the expected T-cell type found in the unexposed.
    [Note that this became evident to all researchers (including Michael Repacholi) in this early part of 1994.]

The results for both studies were due to be published in March or April 1995 (See Appendix Z/2). However the GSM radio exposure study was released on 29 April 1997 after a never-satisfactorily-explained two-year delay. It took another year for the ESAA's electricity power study to be released.

The researchers claim that a year of the GSM-report time was occupied by checking the B-cell cancer question (they say the Maryland institute took a long while to do the work), and that another year was lost because the first three major publications (Nature, Science and one other) passed it for peer-review—but then decided not to publish "for political reasons" (the quote of one scientist) and because it was too hot to handle (according to one of the others).

Michael Repacholi also told me at the London conference in November 1996 that the study had passed peer review a couple of times, but hadn't been accepted for publication. This was the only bit of information that turned out to be true. Three publications rejected it before Radiation Research published the report—and according to one of my sources, only one of the publications had any significant comments about the quality of the paper, but they all rejected it on (admitted) "political" grounds as being too hot to handle.

Telstra maintains that it played no part in the official press conference at Adelaide on 29 April. However, Telstra was the major funder of the study and is also Australia's provider of video conferencing services. Telstra was known to have had the main results of the study at least a year beforehand, although this is denied by Dr Hugh Bradlow (See Appendix X/5). The Adelaide Hospital study scientists also claim that the confidentiality clause was insisted on by Telstra, not by the scientists (see Bradlow letter also).

It is important to realise that media control is a well developed art in Australia. Holding an important Australian medical/technical press conference in Adelaide is a bit like holding a similar, rather esoteric British press conference in Kiev — it's about the same distance from the media centres and the journalistic expertise. This ensured that the major press wouldn't be able to cover it. They will try to pick the key parts of the story up from Reuters, or perhaps send an Adelaide-based cadet journalist out to cover it. "The Australian" did neither.

Also, via a satellite video conference hook-up, Dr Michael Repacholi was brought into Australia from Geneva (he had headed off there to his new job with the WHO EMF Project before the study really got underway). Telstra maintains that the hospital paid for this extraordinary cost—but if so, it is the first time that any Australian hospital has ever had this sort of money to throw around (it would have cost $5,000 at least).

What's more, this claim that the hospital was paying for it, allowed Telstra to block expert journalists like myself from participating in the video conference. Normally with a satellite link of this kind, Sydney and Melbourne technical journalists would have been invited to participate via Telstra's Sydney and Melbourne video conferencing rooms (they have a number, and one is always available).

I tried to get access to Telstra's facilities in Sydney, but was not permitted. In fact, although I write more about this subject than all other Australian journalists combined, I wasn't invited to Adelaide either, or advised by anyone officially that it was taking place!

The day before the 29 April announcement, Australia also celebrated the first anniversary of the infamous Port Arthur massacre in Tasmania where over 20 people died at the hands of a deranged gunman. This had triggered a nation-wide determination to recall and destroy all unnecessary weapons, which resulted in a fierce battle between pro-gun and anti-gun lobbies.

Consequently it was well-known beforehand that all Australian newspapers had plans to devote large sections of their news pages to multiple images of the formal ceremony (attended by political leaders and families of the victims) and discussion on gun control. Television coverage of the ceremonies and discussions on gun control filled all news and current affairs programs also. Radio had gun control on its mind, totally.

There would not have been a better day in the last few years in which to release a scientific story that the carriers and their spin-doctors wanted to cover up and keep quiet. This absolutely ensured that the official press-conference on April 29 was submerged in news coverage by the Port Arthur material.

After three years in almost total security, the story of the research finding was also leaked on the Sunday to the Hobart Mercury—in a way that has never been explained—meaning that "the scoop" was buried in a small provincial Tasmanian newspaper at a time of state and national mourning.

I had a contact in Adelaide who kept me informed, and I wrote the story for the national broadsheet, The Australian—thinking it could perhaps get front page coverage on the day after—but the news editor did not publish it at all, for reasons, she said, "of space", and later because she claimed it to be "old news" since it had been previously published in Hobart.

The story would not have got out to the wider world if I hadn't subsequently written full details in my columns on May 6 and May 15 (See both Appendix Z/3 and Z/4) explaining the significance, and I also circulated this material and some other notes to journalists around the world via e-mail and Internet forums.

The Adelaide Hospital study is now seen as probably the most significant research finding ever released which directly implicates radio frequency exposure as a potential cancer-promoter. It proves, almost beyond doubt, that radio exposures can have serious adverse effects on biological tissue—something long denied by the industry.

It does not link mice cancers to humans—but nor does any animal research. The importance also lies in the fact that findings were statistically highly significant (OR=0.999) and the research controls and protocols were excellent. What's more, neither the scientists or the funders could remotely have been said to have wanted these results to emerge.

In my opinion, they did everything they could (legally) to stop them being heard or understood. See the Government's response in Appendix Z/1.

4.3 ESAA mains-power study

This side of the study used 625 mice overall, with 97 being "sentinel wild mice" (just to check for random diseases).

The 528 transgenic mice were divided into five groups (111, 105, 103, 105, 104) and the first became the unexposed control (actually "sham-exposed" or pretend-exposed) group. The others were exposed to constantly levels of 1, 100 and 1,000 microTeslas of magnetic field which was carefully filtered to remove the normal "transients" — leaving just pure 50Hz waves.

Why this was done has never been explained — although it is probably understandable if the ESAA had feared that random radio frequency transients may have been causing or promoting cancer. It created results which could not really then relate to normal human exposure.

One other 1,000 microTesla group received exposure on a 15 minute on, 15 minute off cycle.

What they actually found (I have the full study report) is a substantial rise in lymphoma in the most exposed groups above that expected (15 per cent was expected, but the highest found was 30.5 per cent). However they also found a very substantial rise in the rate in the control group (15 per cent expected, and 28.8 per cent found). This was further confused by an unexpected form of kidney disease which also hit the exposed group harder than the unexposed, and so made the interpretation of any clear-cut results virtually impossible.

These are the facts:

  • There were substantially higher levels of lymphoma in the exposed mice;
  • There were higher levels of lymphoma in the unexposed mice;
  • However the difference between the two was not significant in statistical terms:
    • the highest exposed group had 32 cases or 30.5 per cent;
    • the second highest having 35 cases or 33.6 per cent;
    • the third highest having 27 cases or 26.2 per cent;
    • the fourth highest having 31 cases or 29.5 per cent;
    • the unexposed group having 32 cases or 28.8 per cent.
  • The expected level of lymphomas in these mice (from historical records) was only 15 per cent—so even the unexposed mice were nearly double what had been expected. No one can explain this.
  • There was also a high level of kidney disease with:
    • the highest exposed group having 20 cases or 19.2 per cent
    • the second highest having 16 cases or 15.2 per cent
    • the third highest having 12 cases or 11.6 per cent
    • the fourth highest having 9 cases or 8.5 per cent
    • the unexposed group having 10 cases or 9 per cent.

        One would normally assume the significance of this is fairly substantial. There is a clear fall-off in incidence between exposed and unexposed — regularly progressing downwards as exposure is reduced. In fact the report admits this is statistically significant.

    However this was outside the parameters of the research design (to look at cancer-rate incidence) and appears just to have been ignored.

    One valid assumption here, could be that the EMF exposure was effecting the immune response in general, which explains the slightly higher lymphoma rate and the much higher kidney disease rate in the exposed animals. Such a proposition was never considered, apparently.

  • In terms of surviving the full 18 months of the research program:
    • the highest exposed group had the least survivors — 40 per cent
    • the second highest group had 42 per cent survive
    • the third highest had 50 per cent survive
    • the fourth highest had 46 per cent survive
    • the unexposed group had 45 per cent survive.

        In other words, taken overall, less of the exposed mice survived—and the exposed mice clearly were more susceptible to kidney disease. The results therefore suggested that there could be health problems with high levels of 50Hz mains power. But fairly obviously these are not statistically significant findings taken overall, and you can't come to any conclusions without understanding what caused the kidney disease.

    The only conclusion that anyone can make here with any certainty is that this research is worrying, and it certainly needs to be repeated urgently.


Especially since so many epidemiological studies also point to low level increased incidences of childhood leukaemia. (See other material.)

4.4 How the release of information was handled in both cases


The R/F cellphone study:

The GSM cellphone study was published in the May 1997 edition of Radiation Research as "Lymphomas in Emu-Pim 1 Transgenic Mice Exposed to Pulsed 900 MHz Electromagnetic Field", by Michael H Repacholi, Antony Basten, Val Gebski, Denise Noonan, John Finnie and Alan W Harris. (See abstract Appendix Z/1.)

The claim is made (to the interviewer) that a two year delay resulted from:

  • the need to confirm the finding of B-cell lymphomas
  • the fact that three of the most prestigious peer-review publications refused to handle it because they didn't want to get involved in promoting this sort of controversy — despite the fact that the paper passed through peer-review with two publications easily and quickly.
I accept both of these arguments, having constantly experienced much the same resistance by editors to enter controversial fields, myself. My guess is that this explains half of the two-year delay, and a desire to let interest cool down explains the other year.

The ELF mains-power study:

The second part of the Adelaide Hospital study was released nearly a year after the first (2 March 1998) — which was a full three years after the end of the mouse-exposure phase, and nearly three years after the analysis was first promised.

There is absolutely no excuse for this delay. The R/F study did present some problems but not the ESAA, according to the researchers themselves. They refuse to say why it took so long. But there were no B-cell implications that needed to be checked in Maryland — and the paper was published without the problems associated with the R/F study, so it could have been done in a few months.

This was clearly an attempt to let the publicity surrounding the R/F study blow over. No one wanted the claimed findings of safety to be examined too carefully. In July to December 1997 also, there was a very strong and very vocal reaction to the New England Journal of Medicine's Editorial (July 1997 — See Appendix E/2)[45] calling for all funding of mains-power research to be curtailed. The UK New Scientist magazine also took this line in a story and an editorial. (See Appendix E/5.)[46]

This was followed, for many months after, by a spate of scientific letters and reports to editors of trade and scientific magazines pointing out that the "definitive" research (the NCI—Linet study) on which the Journal and New Scientist had based their claims, was incorrectly interpreted. However, I don't recall anyone ever printing a retraction.

In fact the NCI-Linet study confirmed numerous previous studies that showed a slight link between high mains-power exposure and childhood leukaemia. It is now used by everyone as one of the more definitive studies linking mains-power to ALL in children. (There are about 10 others with similar findings — mainly US and Swedish) and a few which are inconclusive.


    Through leaking it to a provincial newspaper in a day when Australia's interests were elsewhere, then holding the press conference in a remote part of the country and refusing to use a video-conference link already set up and paid for.

Mains Power:
    A Sydney press conference with all the trappings; press releases issued under the University of Sydney banner; one-to-one interviews for television news.

    The theme of the press conference was to play down the significance and emphasise that this only applied to mice—not humans. The principal scientists who had done the work were not made available and they stopped questions very quickly after the key announcement and speech from Geneva by Repacholi.

The discussion section of the report leads off with a typical weazle-word statement that " "this single study"" (suggesting that the evidence is unique and isolated, when it is not) " "cannot be applied directly to assessment of human cancer risk."" (Has any animal research ever been directly applicable to human risk?)

This paragraph and the study itself is most notable for its complete lack of discussion about the meaning and implications of the positive effects — and for presenting as many reasons as they can think of for discounting the finding — and for suggesting that this could somehow be viewed as a trial on the usefulness of transgenic mice in radiation studies.


And this finding is only one of hundreds which have consistently shown this, with varying degrees of validity and credibility over many years. It fits almost perfectly into the overall "assemblage" of evidence accumulated by many different independent biomedical researchers from many varied studies on animals and cell-cultures.

Mains Power:
    The findings were finally made public at a Sydney press conference, complete with press releases from Sydney University's Centenary Institute headed "No Evidence for Cancer Link with Powerlines". The report was published also in Radiation Research journal.

Professor Basten sought out the ABC and commercial television stations and did one-to-one interviews with each, where he was quite adamant that this research now proved almost beyond doubt that power-lines did not cause cancer. I recorded three TV news programs that night, and each led with this statement, and each later included a Basten interview stating that power-lines now had the all-clear.

The ESAA also issued a press release headed "Australian study reveals no evidence for EMF cancer link" which says:

"The result is reassuring because the mice used in the study were particularly sensitive to showing subtle changes of any magnetic field effects [A total distortion—there is no known link between magnetic fields and lymphoma] and the magnetic field levels used in the laboratory were much greater than those experienced by people in daily life."
[Actually some of them were at about normal exposure levels, and some were higher.]

  • Highly significant increases in lymphoma in the exposed group compared to the unexposed.
  • A different type of lymphoma in the exposed.
  • A much more dangerous type of lymphoma in the exposed.
  • Earlier onset of lymphoma in the exposed.

Mains Power:
    The result is confused by the kidney disease, and no real conclusions can be drawn. However, there were noted increases in both lymphoma (very small) and kidney disease (highly significant) in the exposed groups, and this needs urgent research to clarify.

    The claim that it is not possible for radio signals to affect biological tissue is resoundingly defeated. The indications are that cellphone type and level of radiation can act as a cancer promoter—and this needs urgent further work, and justifies the call for the application of the prudence principles.

Mains power:
    More and urgent research needs to be done (NO ONE IS DOING IT). This study has proved nothing — but it suggests that compromise of the immune system might be the mechanism.

4.5 Manipulation of the media
    What interests me here is the way in which the release of the information was manipulated—by the scientists, by the hospital, and by the ESAA and Telstra (it is often not clear which)—and sometimes by all of them together.

Remember, two and a half years after the completion of the study, not one word of results had leaked out. In the interim, Dr Repacholi had attended dozens of conferences and given dozens of interviews, and still vocally maintained his stance that there was no evidence connecting cellphone exposures to adverse health consequences—knowing all the time that his mice had shown a major, highly significant, increase in basal-cell lymphomas.

Yet Michael Repacholi told me off-the-record at a London Conference on 15 November 1997 (it is recorded in my journalist's notebook) that the research had turned up "nothing of any significance". He also revealed that the group had problems finding a publisher. I already knew this from another source in the research team. (See Appendix Z/12 e-mail from Repacholi also).

At the same London conference, he was very vocal in supporting industry claims that there were no studies linking cellphones to adverse health effects and strongly criticised a few scientists who had turned up positive results. There were dozens of people at the conference who can attest to this.

At this time Dr Repacholi was the head of WHO's EMF Project and probably the second most powerful cell-research-funding bureaucrat in the world (Dr George Carlo was the most powerful)—yet he was publicly denying and discounting his own unpublished research.

At that time Repacholi had known for over two years that the Adelaide Hospital research finding was the most significant link yet discovered. It had a "highly significant" p-value, and an Odds Ratio (OR) of 0.999 — meaning that this doubling of leukemia in the exposed mice could only have arisen by chance once in a thousand experiments. This is 10 times more significant than the normal 1 per cent "high-significance" level in a very well-conducted live animal trial.

Even more significance lies on the difference between the exposed and unexposed groups — and in the fact that normal GSM cellphone handset exposure levels produced this cancer-promotion (not initiation) result with one hour of daily exposure over nine to 18 months.

So Repacholi's claim to me that the findings were "not significant", and also the fact that he continued to maintain at scientific conferences that there was no evidence that cellphones can be harmful, was not the truth. A number of scientists have also complained bitterly about this sort of conduct by Dr Repacholi, on this and other related matters—but this was my first and only experience.


A point not made elsewhere in the literature is that, at the conclusion of the 18 month period of the study, the remaining live mice were killed—BUT NOT AUTOPSIED.

Since the level of lymphomas in the cellphone study had been rising steadily from six to 18 months, it would have been reasonable to assume that even more significant figures would have been found if the last live animals had been autopsied. This was not done—and I find this fact very disturbing.


I couldn't get any of the electronics magazines interested, or the science magazines I sometimes write for, so finally I wrote a few columns (See Appendix Z) and followed up a year later with a couple showing how the industry was manipulating both the cellphone health issue and the electricity mains-power results (also included here).

I also wrote and circulated a couple of long pieces on the cellphone research to other journalists via the Internet. New Scientist then approached me to write 750 words on the subject.

This is a very short story for such a complex subject. They then rejected it and asked me to write a longer explanatory piece — which I did — and which they then cut to about 500 words, dumping all the explanatory material, and filling the page with public-relations rubbish from the US Cellular Telecommunications Industry Association saying why the study wasn't of much importance (See Appendix Z/10).

They published this under my name after very strong objections from me — and then refused to publish a letter where I disowned the second part of the story as it appeared. I now refuse to write for this publication, or their sister publication Electronics Weekly.


A few weeks after the news of the Adelaide Hospital research spread around the world, Nokia and the European manufacturers mounted a tactical response. They dragged out two studies that they had funded to prove that cellphones were safe.

They hurriedly put together a Helsinki press-conference (May 22 1997), based on the work of Maila Heitanen (completely unknown in the R/F research field), who had subjected 19 University students to an electro-encephalogram and switched a GSM phone on in the room, while watching the brain waves.

She concluded that there was no change in the brain waves.

To build up some credibility, they also had a progress-report by Jukka Juutilainen (a highly respected researcher in this field) who was part-way through a mouse study (which appears never to have been published).

They then had Reuters issue a press release which went around the world, and was widely quoted as:

An amazing number of newspapers and magazines around the world carried this as a lead story, because if followed the "scare story" of the Adelaide mice by only a few weeks — and it kept advertisers happy — and it arrived as part of Reuter's regular news feed.

The full details are in Appendix F, including my e-mail correspondence with the two scientists, asking how they got caught up in this sham. You will find this most instructive.

It is not unusual. In fact, I would say this sort of "good-news" publication is the dominant form of cellphone-and-health news circulated by the major press agencies.