July 2004
Industry Rules RF
Controlling Research, Setting Standards and Spinning History
f you had any doubts that the wireless industry is in total control of the RF health debate, you need only to have gone to the workshop held at the FCC’s Washington headquarters on June 28. By the end of the day, the fog would have lifted.
Motorola’s Joe Elder told the assembled delegates from the U.S., the EU, Japan and Korea that the health issue is just about settled. There is no credible evidence that casts doubt on the current 4 W/Kg threshold for ill effects from mobile phone radiation, he said.
Elder has changed his tune since joining Motorola a few years ago. He spent most of his professional career at the EPA where he worked on RF radiation and health. Back then, Elder had a radically different outlook. In the early 1980s, he was in charge of EPA’s RF health review. His 268-page report, Biological Effects of Radiofrequency Radiation, issued in 1984 after a rigorous external peer review, concluded, “[B]iological effects occur at an SAR of about 1 W/Kg; some of them may be significant under certain environmental conditions.”
The review document was designed to be the basis of a national RF exposure standard. But the EPA quietly shelved the plan after the TV and radio industry complained that it would be too expensive.
At the next coffee break, we asked Elder what had changed. What had we missed over the years? Which studies had prompted a quadrupling of the RF threshold for thermal effects?
Elder replied that what he had really meant back then was that the threshold was somewhere between 1 and 4 W/Kg. We walked away unswayed. If anything, the data that have accumulated over the last 20 years argue for a tougher, not a more lenient, standard.
In his talk, Elder, said that many of the low-level biological effects that have been reported could not be repeated or confirmed. Bernard Veyret, a member of the EU delegation, made a similar point in his own talk. Each, for instance, cited a 1994 study by Henry Lai and Bill Guy of the University of Washington in Seattle, which showed that exposure to pulsed 2450 MHz microwaves at an SAR of 0.6 W/Kg impaired the learning behavior of rats.
Numerous attempts to replicate this study have failed, according to both Elder and Veyret. Veyret went further, proclaiming that any question that RF radiation can affect an animal’s memory, performance or learning is now “closed.” Veyret and Elder each said that Lai had gone wrong by using a maze made of wood. As a rat traveled through Lai’s maze in search of pellet rewards, they asserted, it left chemical markers (urine, etc.) that were absorbed in the wood. These sensory clues would then guide subsequent rats in their journey through the maze. The message was clear: Lai he didn’t know what he was doing.
It was striking that two supposedly independent critics had come up with the same deceptive argument to attack one of their colleagues. When we got back to the office the next day, we called Lai and asked for his side of the story. On hearing what had been said behind his back, he got angry. “They don’t know what they are talking about. It is quite sickening that they make such comments on other people’s work without first checking the facts,” Lai said. “The wood in my maze had a plastic coating and after every run I washed the whole maze with vinegar to get rid of any smells. This is standard operating procedure, and anybody who does this kind of work would know that.”
“What about the failed attempts to repeat your experiment?” we asked. Lai went over them, one by one:
“The 2002 study by Diane Dubreuil of France’s Paris-Sud University was quite different from mine,” Lai said. “I’m not sure why they did not see anything, but my first guess is that the RF effect depends on the complexity of the task that must be learned. At this point, I can’t say why our results do not agree.”
With respect to the paper by Brenda Cobb and Eleanor Adair at Brooks Air Force Base in Texas, which had also failed to see an effect on learning, Lai said that Bioelectromagnetics had asked him to peer review this study. “I recommended that it not be published without resolving some serious methodological problems —for instance, the animals appeared to be over-trained and the data were so variable that it would be difficult to detect any effect. But the editor accepted it anyway. Even the minor changes I suggested were ignored.” Lai noted that C.K. Chou of Motorola was the editor in charge of the Cobb-Adair manuscript at Bioelectromagnetics.
The most recent failure to see an RF effect on learning was that of Jean-Christophe Cassel of the University Louis Pasteur in Strasbourg, France. Cassel and Zenon Sienkiewicz of the U.K.’s NRPB have each tried to repeat Lai’s experiment as part of the European Perform B project (Cassel with rats, Sienkiewicz with mice). A few years ago, Sienkiewicz had run the experiment at a low SAR —0.05 W/Kg— more than ten times below the level used by Lai. “It would have been very surprising if he had seen anything with that kind of exposure,” Lai said. Cassel’s paper has cleared peer review and will be published in a future issue of Behavioural Brain Research. An advance copy was posted on the Internet on May 12 and few people, other than Veyret, who ran Perform B, have yet seen or digested it. (Cassel’s abstract may be accessed here after inserting the following doi: “10.1016/j.bbr.2004.03.031”.) Lai commented that it appears to be a fair piece of work, though he did point to some differences with his original experiment.
In the end, Lai’s learning effect may not stand up and may have to be withdrawn. But even Cassel does not believe that we’re there yet. He closes his new paper with a plea for more research to resolve the conflicting data so that the controversy is not allowed to fester.
Nor is Lai ready to call it quits. “I’m not going to close the book quite yet,” he said.
There are good reasons not to rush to judgment. This is the not the first time that industry has tried to impugn Lai’s research. Ten years ago —at about the same time that he published the RF-learning paper— Lai announced that he and N.P. Singh had found that RF radiation can increase the frequency of DNA breaks in the brains of live rats. Motorola signed up Joe Roti Roti of Washington University in St. Louis to repeat the work. When he failed, Motorola had no interest in exploring why the two labs had different results. Rather, industry’s position was that Roti Roti was right and Lai was wrong: The issue was closed.
That might have been the end of the story but for another European research project, called REFLEX, in which ten labs in a half-dozen countries investigated possible RF health risks. Two of the participating labs reported that they too see RF-induced DNA breaks, albeit in cellular (in vitro), rather than in animal, studies.
“Cells responded to RF-EMF exposure between SAR level[s of] 0.3 and 2 W/Kg with a significant increase in single- and double-strand DNA breaks, ” according to the summary of the just-completed four-year REFLEX project. The complete final report is slated to be released at the end of July. Franz Adlkofer of the VERUM Foundation in Munich, who led the REFLEX project, is convinced that RF radiation can indeed be toxic to DNA. “As far as I am concerned, the genotox work is fully confirmed,” he told us in late-June.
To be fair, we should add that one of the REFLEX labs failed to see an increase in DNA breaks following RF exposure: Veyret’s team at the University of Bordeaux. Veyret’s notion that the Lai’s learning effect is a closed issue was one of his many characterizations of RF bioeffects. The oddest one of all was that last year’s TNO study from the Netherlands linking health complaints to extremely weak GSM and 3G radiation, is only “half-open,” that is to say, it is already half-closed. Even before the first attempt at repeating the TNO study has gotten under way —in fact, no one has yet won funding to repeat it— Veyret would have us believe that it’s already half-settled. Apparently, we no longer need a failed replication of a study that challenges the orthodoxy. Rather, inconvenient findings may now be smeared right away and those who did the work put on the defensive without waiting to see if anyone can repeat it.
As it happens, on the very same day that Veyret was casting doubt on the TNO study, the Health Council of the Netherlands, a group that is well-known for its conservative outlook and often cited with approval by industry, issued its own evaluation. The “design and execution of the TNO study are of good quality,” the council stated, and “there are good reasons to replicate the TNO study” because of the possible “implications for public health.”
Like Elder, Veyret has changed over the years. He used to believe in nonthermal mechanisms and was dedicated to finding ways to use EMFs for fighting cancer. Back in 1991, for instance, he showed that modulated microwaves as weak as 0.015 W/Kg could alter the immune system of mice. The work was promising, but Veyret, like others who reported low-level effects, had a hard time finding support to move his research forward.
Today, with a pro-business attitude, Veyret has fewer money problems. His group at the University of Bordeaux has become industry’s go-to lab in Europe. It’s the counterpart of the Battelle Pacific Northwest labs and Roti Roti’s group at Washington University, where Motorola and its partners can be confident about what will be reported and published. Veyret is also now very popular among EMF governing councils in France and elsewhere. He is a member of ICNIRP and sits on innumerable other EMF committees. He is invited to speak at conferences and workshops around the world. Membership in industry’s No Effects Club certainly has its privileges.
The industry’s ultimate objective is to relax the U.S. cell phone health standard, which is one of the strictest in the world. The plan is to double the present exposure limit. Motorola’s Chou was slated to brief the delegates on the proposed revision of the IEEE exposure standard on the second day of the FCC workshop. Chou was one of the U.S. delegates at the FCC meeting. Motorola had two of the four non-government slots on the delegation. (Who picks the delegates anyway?) Half of the EU delegation was also from industry: Joe Wiart of France Telecom and Gerd Friedrich of the FGF, the German wireless research group. We cannot tell you what Chou said because the press and other observers were specifically not invited to this session. As the cochair of the subcommittee that is revising the IEEE standard, Chou is in an excellent position to deliver this prize for the industry. The FCC is already on the record as promising to consider adopting what Chou comes up with.
Industry spin is nothing new. The wireless industry controls much of what is reported about RF research. Nowhere is this more apparent than at the Bioelectromagnetics Society where Motorola’s Mays Swicord exercises editorial control over the newsletter. (Swicord, who used to work at the FDA, is yet another example of a public health official whose views made a U-turn after joining industry and who is now dismissive of low-level risks.) The most recent example of Swicord’s handiwork is a one-sided report on the heat shock protein (HSP) workshop held in Helsinki in April. Much of the data pointing to an HSP effect was somehow omitted from the summary that appeared in the BEMS newsletter’s May/June issue. Notably absent was any mention of the work by Dariusz Leszczynski of STUK, the Finnish Radiation and Nuclear Safety Authority. The oversight was doubly hard to explain: Leszczynski had hosted the meeting and his provocative HSP findings had been one of the major reasons for the workshop in the first place. A small dissident group within the society is now calling for an editorial board to rein in Swicord’s excesses.
Manipulating the written record of what is said and presented at scientific meetings is an often-used control tactic. Another example is last fall’s meeting on RF and the blood-brain barrier, held in Reisensburg, Germany. Its summary was also watered down to make it more palatable to industry interests. This was not hard to do since FGF (the German mobile industry) helped organize the workshop and its director, Gerd Friedrich, also serves as the secretary of COST281, the European research coordinating committee which cosponsored the meeting.
In RF land, industry is the boss. It has long decided much of what research gets funded and what gets published. It also sets the safety standards. And now it dictates what part of the story you get to hear. George Orwell said it best, “Who controls the past controls the future. Who controls the present controls the past.”