Shock journalism.

The junk reporting behind the power line-cancer connection

Wind shear is one of the major banes of air travel, a killer held responsible for causing 18 crashes and 575 deaths since 1970. Aviation authorities have recently found a way to combat this menace through Terminal Doppler Weather Radar, which can provide precious minutes of advanced warning. This radar could have prevented the deaths of 37 passengers in Charlotte, North Carolina, in July. It will be installed now at Charlotte and some 150 other sites around the country.

But two of the nation's largest airports, JFK and LaGuardia on Long Island, won't be enjoying its life-saving benefits any time soon. Politicians and activists have stopped its installation there, since Long Island residents--wrongly convinced that they are already suffering an epidemic of breast cancer--have been terrified into believing that electromagnetic fields (EMF) from the radar would afflict the area with even more cancer.

This terror--indeed, most of the nation's terror of EMF in general--can be readily traced to one person. If and when a plane crashes into Long Island because of wind shear, you should know who is responsible. Meet Paul Brodeur.

Brodeur is a reporter who has made his career by digging up massive conspiracies. He wrote a book on asbestos, Outrageous Misconduct, that contributed greatly to spreading the national asbestos scare. But he has devoted the latest years of his life to trying to prove that EMF from radar stations and especially power lines are "the most pervasive--and covered up--public health hazard Americans face." Like A Nightmare on Elm Street's Freddie Krueger, Paul Brodeur keeps coming back, most recently in his 1993 book The Great Power-Line Cover-Up.

Clearly, many think highly of Brodeur. His articles have won many awards, including the Sidney Hillman Foundation Award, Columbia University's National Magazine Award, an American Association for the Advancement of Science/Westinghouse Science Writing Award, and an American Bar Association Award. His last installment on EMF in The New Yorker, "Calamity on Meadow Street," was a finalist for the National Magazine Award and the influential, albeit politically correct, Publisher's Weekly could barely contain its excitement over The Great Power-Line Cover-Up, calling it an "important, riveting expose" which "should be on Al Gore's desk as he tackles environmental health problems, and on Hillary Rodham Clinton's bookshelf as she reorders national health priorities." It probably was.

But it's The New Yorker and its stodgy reputation that gives Brodeur his true credibility and allows him to rise above the tabloid level. In a periodical that humbly calls itself "the best magazine in the world," Brodeur has published several huge articles on EMF that formed the basis not only for The Great Power-Line Cover-Up but also for his 1989 book, Currents of Death.

Brodeur might do as well as he does simply because fear sells. When my book Science Under Siege was published last year, the producers of Larry King Live, fresh from having launched the cellular phone/brain cancer scare, said they thought it might be good to highlight the anti-hysteria side of things by building a show around a chapter in my book on EMF. They asked Brodeur if he would appear on such a show, but Brodeur refused to appear with me. King's producers bumped me, added Brodeur and two other proponents of Brodeur's thesis, and put them all against a single physicist with little or no television debating background.

Brodeur seems to share certain traits with horror author Stephen King, who says that when he feels he's having trouble scaring the reader he just disgusts them with gore. Thus, in the case of one tumor which Brodeur attributes to power line exposure, he treats us to the niceties of brain surgery. "He removed my forehead bone in two pieces," he quotes the victim, "and took out a meningioma--a generally nonmalignant but often fatal tumor--which was the size of a small grapefruit and had got entangled in my optic nerves and stretched them out as fine as ribbon. In order to take out the tumor, he had to sculpt out a small piece of my brain." Brodeur's description goes on and on, literally ad nauseam.

Those who warn of the dangers of EMF exposure want to essentially eliminate that exposure, either through burying all lines in residential areas or by buying up all exposed houses. H. Keith Florig of Resources for the Future estimates that power companies are already spending $1 billion a year to reduce EMF exposure, but that is a tiny figure compared to what they will have to spend if the recommendations of anti-EMF activists become law.

Testifying before Congress last year, Florig calculated that were we to bury all the power lines that expose people to levels of EMF that worry activists, it would cost about $200 billion. That would show up as a nearly $4,000 one-time surcharge on the electric bill of a family of four. Buying out houses near those power lines as an alternative to burying the lines would cost merely $90 billion.

All this might be justified if we were facing one of the greatest health threats known. But the evidence, such as it is, appears to be largely based on fear, conspiracy-mongering, ignoring contrary evidence, and statistical gerrymandering.

A Conspiracy So Immense

The anti-EMF thesis is built around studies showing its biological effects on simple organisms and small animals, along with some studies done on humans. Many researchers think these effects mean nothing. "The solid weight of biological research" argues that electric and magnetic fields should not be able to cause cancer, said British epidemiologist Richard Doll in a New York Times interview. Doll, the man given credit for providing epidemiological proof that cigarette smoking causes cancer, said that "for the sort of fields we're talking about, we know they don't damage DNA. The idea that they should cause cancer is bizarre."

Still, as Brodeur repeatedly states in Currents of Death, there are "32 published studies demonstrating ELF [extremely low frequency, another term for EMF] effects." What Brodeur doesn't say is that hundreds of studies have been performed in which EMF shows no biological effects. Further, "biological effects" aren't necessarily harmful. As the authors of a report on EMF for the Congressional Office of Technology Assessment noted, "A biological effect is not necessarily a significant health consequence."

However, even a distant possibility of biological effects from EMF makes it hard to dismiss out of hand the possibility that they could cause harmful ones. Thus, numerous epidemiologists have sought out evidence of EMF's harm or lack thereof in real-world situations. The major studies have looked at power-line exposure, since threats from radar microwave emissions are taken much less seriously even among those who fear EMF.

According to Brodeur, the evidence of these power line studies is utterly conclusive. His books would be meaningless if they weren't. As their titles make clear, it's not merely that the weight of evidence says power lines cause illness, but that the evidence is so overwhelming that to deny it is to be engaged in a coverup.

But not even the scientists whose work Brodeur cites are as certain as he is. Epidemiology is a crude instrument, rather like the first radar units. As the occasional Cuban defector who flies into U.S. airspace undetected shows, even today identifying a single plane can be tough. Both radar and epidemiology are much better suited to detecting major problems. Epidemiology can detect strong cancer correlations like smoking and lung cancer fairly readily. But the risk ratios from the childhood leukemia-EMF studies, if they show a correlation at all, are in the area of two to three, meaning a two or three times higher risk of contracting the disease, compared to a 10- to 30-fold difference in cancer risk between cigarette smokers and nonsmokers.

Those small risk ratios indicate that the risk of EMF, assuming there is one, is much smaller than such clear carcinogens as tobacco smoke. And such a small blip on the risk radar screen may prove to mean nothing at all. No one study is definitive. Rather, researchers look at each study the way a professor would look at a quiz. Any student may have a particularly good or bad day, and quizzes may be flawed, but averaged over the course of a semester, enough quizzes provide a fairly good assessment of where a student stands.

In the case of EMF and cancer, there have been so many quizzes that it is best to look at the broad overviews in various government reports rather than focus on each one individually.

In May 1989, the OTA released a report, one line from which appeared in a multitude of articles and in Brodeur's book Currents of Death: "As recently as a few years ago, scientists were making categorical statements that on the basis of all available evidence there are no health risks from human exposure to power-frequency fields. In our view, the emerging evidence no longer allows one to categorically assert that there are no risks."

Unfortunately, most of these articles and Brodeur's book neglect to complete the sentence: "...but it does not provide a basis for asserting that there is a significant risk." For those whose jobs expose them to EMF, it concluded, "the evidence to suggest that exposure to extremely low frequency electromagnetic fields presents an occupational hazard is extremely weak."

With childhood leukemia, the EMF danger that has gotten the most attention, the OTA report found that "collectively the studies do not provide good evidence that ELF exposure increases the risk of leukemia." Nowhere in Brodeur's work does this line appear. But, they added, carefully hedging their conclusion, "at the same time the evidence precludes categorical statements that no such risk exists."

The British equivalent of the OTA report is considerably more helpful, since it appeared three years later. It looked at 80 surveys of cancer in electrical workers, eight of childhood leukemia, and several of adults exposed in non-occupational settings. This report found that "while there is suggestive evidence of an association between childhood cancer and residential electromagnetic field exposure, the methodological shortcomings of the studies are such that the evidence is insufficient to allow conclusions to be drawn."

Where the OTA report said there was "not enough evidence to judge the possibility of an association" between residential exposure and adult cancers, the British panel took a firmer position: "The data on adult cancers and electromagnetic field exposure are sparse, and available evidence does not suggest electromagnetic field exposure is a risk factor."

At the request of the Department of Labor, which was responding to the Brodeur articles, the Committee on Interagency Radiation Research and Policy Coordination (part of the White House Office of Science and Technology Policy) ordered a review of the literature. Conducted by a panel of independent scientists convened by Oak Ridge Associated Universities, it concluded that "there is no convincing evidence in the published literature to support the contention that exposures to extremely low frequency electric and magnetic fields generated by sources such as household appliances, video display terminals, and low power lines are demonstrable health hazards." It went on to apply relative risk standards in saying that "in the broad scope of research needs in basic science and health research, any health concerns over exposures to ELF-EMF should not receive high priority."

What does Brodeur make of these reports? He ignores the British one in his writings. The OTA one is cited briefly in support of his position in both Currents of Death and The Great Power-Line Cover-Up. This is notwithstanding that the chief author of the OTA report, Carnegie-Mellon Professor M. Granger Morgan, blasted Currents of Death in a review in Scientific American.

The Oak Ridge report merits but a few sentences from Brodeur, mostly an ad hominem attack on one of its authors, Dimitrios Trichopoulos. Trichopoulos, suggests Brodeur, has compromised his position because he "had testified as a consultant for Crowell & Moring, the electric utility industry's lead attorneys, at the Scientific Advisory Board meetings in the winter of 1991."

Trichopoulos is the chairman of the department of epidemiology at the Harvard University School of Public Health. His articles in major medical journals have covered the range of epidemiological subjects from hepatitis C to tobacco smoke to breast cancer. But to Brodeur he's just a utility industry sell-out. Brodeur strangely thinks that people form their opinions based on who hires them as consultants, rather than being hired as consultants based on their opinions. If he applied this standard fairly, Brodeur would point out that Dr. Samuel Milham Jr. of the Washington State Department of Health and Social Services, whose work Brodeur heavily relies on in both of his last two books, was an expert witness for the plaintiffs in a San Diego trial alleging that EMF had caused a serious illness. Brodeur discusses the trial but neglects to mention Milham's role.

The only governmental overview that Brodeur dwells on at all isn't one. That is, it was never released except in draft form. The report composed a large part of Currents and showed up on no fewer than 67 pages of The Great Power-Line Cover-Up, out of 308 in the whole book. Yet this document was clearly marked "Workshop Review Draft" and had on each page the words "DRAFT--DO NOT QUOTE OR CITE." None of which stopped Brodeur from persistently quoting and citing it. As James Jauchem noted in a critical article in the Journal of Clinical Epidemiology, "Brodeur seemed to think that being asked not to cite a draft report was synonymous with a 'cover-up.'"

But that's not the only conspiracy involved, according to Brodeur. An earlier EPA draft had used the term "probable" carcinogen but after internal review downgraded this to "possible," which has no official meaning. (Essentially, anything that hasn't been absolutely studied to death is considered a possible carcinogen.) That, too, said Brodeur, could only have been the result of conspiracy. But according to EPA official Dennis O'Connor, "If we had determined that it was a proven cause of cancer in humans, trust me there would have been pressure in succeeding years to deal with that." The agency sent the report on to an advisory board which turned it back for a number of technical reasons such as the need to look only at peer-reviewed material, according to officials. A completely new draft is scheduled for release next March.

Creating Clusters and Fomenting Fear

In The Great Power-Line Cover-Up, Brodeur recounts a press conference held by the head of the Connecticut Department of Health Services's Division of Environmental Epidemiology and Occupational Health and an epidemiologist on his staff. They told reporters that despite the panic caused by Brodeur's "Calamity on Meadow Street" in The New Yorker, there was "absolutely no clustering" in Guilford and that the state investigation found "no cancer cluster on Meadow Street."

Brodeur called this conclusion "disingenuous, to say the least." His evidence: "'The fact that three of the 29 primary brain and central-nervous-system tumors that occurred in Guilford during those 21 years developed among a handful of people who live in four of five adjacent houses on Meadow Street that are situated near a substation and very close to a pair of high-current distribution feeder lines, together with the fact that a malignant eye tumor, involving a tract of brain tissue, occurred in a woman who had lived in a sixth dwelling, next to a third feeder line, surely suggested that there was a cancer cluster of some significance on Meadow Street."

That this sentence is a grammarian's nightmare may not be coincidental. Brodeur excels at obfuscation. He also excels at what can be called "cluster gerrymandering." In political gerrymandering, the party that runs the legislature carefully draws the boundaries of an electoral district so as to squeeze as many voters from the opposite party into as few districts as possible. In Brodeur's case, he carefully draws his boundaries around cancer cases so that just as many as possible fall into his own self-defined cluster. If it serves his purpose to count only four houses from a distribution line, he will do it. But if there is a cancer in that sixth house, the area being considered will conveniently move out to that sixth house. Using such techniques, it is always possible to find a cluster, even though, as with some gerrymandered districts, you might end up with some rather bizarre shapes.

Consider the most important evidence in Brodeur's indictment of microwaves in his 1977 book, The Zapping of America. Throughout the book he refers to what he calls an extraordinary rate of cancer at the U.S. embassy in Moscow. The embassy was discovered in the early 1970s to have been bombarded by the Soviets for the previous two decades by microwave radiation for unknown reasons. After Zapping came out, a government study found that residents of the embassy were remarkably healthy and cancer-free overall. Looking at causes of death of both embassy males and females as separate groups and even breaking these further down into all causes, cancers only, and heart disease only--six different cohorts--the government study found fewer of all those than one would expect in a similar demographic group and fewer than in comparison embassies. It also looked at adult dependents and dependent children and found lower death rates than would be expected.

How did Brodeur deal with this? He declared the government study "had underestimated the health hazard of microwave radiation at the embassy by failing to point out that the death rate--particularly from breast cancer--was extraordinarily high [he doesn't say how high] among women living in apartments on the third through seventh floors of the embassy during the late 1960s and 1970s, when the irradiation of the embassy had become a source of major concern to the State Department and the CIA."

Note how Brodeur conveniently factors out the incidence (or more importantly non-incidence) of non-breast cancers, and of all non-cancerous ailments. This automatically excludes cancers in children and men, since they rarely get breast cancer. He also ignores all the workers save those on floors three through seven. He doesn't tell us that the embassy has 10 floors, meaning he ignores the first two and last three floors. Finally, he ignores the years prior to the late 1960s, even though the study covers years dating back to 1953. By taking his sharp instrument and expertly cutting out persons without cancer, he has created his cluster.

One key in identifying a legitimate cluster is that the diseases must be related. Every household is periodically struck by some sort of illness. A plausible theory of cluster causation would have to show similar illnesses, yet Brodeur does quite the opposite. He seems to connect absolutely every ailment contracted anywhere near power lines or power substations with EMF.

James Jauchem wrote in the Journal of Clinical Epidemiology of "Calamity on Meadow Street" that "in addition to brain tumors, Brodeur seemed to blame exposure to EMF for the following ailments: Osgood Schlatter syndrome [a disease affecting the tendons of the kneecaps], ganglion and keratinous cysts [small pockets of fluid resulting from inflammation], scoliosis and other spinal deformities, parotid gland inflammation, hypospadias [an abnormality of the penis in which the urethra opens underneath], heart defects, lipodystrophy [a fatty tissue disease], glomerulonephritis [a kidney disorder], strep throat, fever, dark brown urine, pain and swelling in the eyes, swelling of the face, arm numbness, severe headaches, and keratoacanthoma [a disease of the eyes]."

Actually, in The Great Power-Line Cover-Up, Brodeur adds a number of other maladies to that list, including bleeding ulcers, benign ovarian tumors, epilepsy, and a growth on the hand merely described as "disabling." In a fit of understatement, Jauchem writes, "Surely it would be difficult to find a scientific basis for linking EMF exposure to this menagerie of ailments."

The Connecticut Academy would seem to agree. In its report it stated that "the diseases reported on Meadow Street were multiple, and there was not a recognizable cluster of any given disease as defined and could be evaluated by procedures of the U.S. Centers for Disease Control. Thus, the Academy concurs with the conclusion reached by the Connecticut Department of Health Services. There is no indication of a recognizable cluster of any type of relevant cancer."

Likewise, Dr. Michael B. Bracken, professor and vice chairman of epidemiology and public health at the Yale School of Medicine, told The New York Times, "There are clusters of any kind of cancer," but "in Guilford, it's not even a cluster because the cancers aren't related. These are individual tragedies."

Brodeur dismisses Bracken by pointing out that he had received a grant from the utility-industry funded Electric Power Research Institute of Palo Alto, California, for $2.5 million. Once again, the double standard is in effect: Two of the three American studies cited most often by those arguing an EMF-cancer connection, including Brodeur, were also funded by EPRI. Utilities finance practically all the research on EMF. If they didn't, nobody else would.

This puts Brodeur in a wonderful position. When a utility-funded study finds an EMF-cancer link, he says that the industry's own study has proved his point. When an industry-funded study or person argues against such a connection, he says it or he is untrustworthy because of who funded it.

Foreign Affairs

Scientific attention continues to focus on EMF. One study came out shortly after the Oak Ridge report and received tremendous attention. As Brodeur reported in The Great Power-Line Cover-Up, "On September 30th of 1992, the wind went out of the utility industry when officials of Sweden's National Board for Industrial and Technical Development formally announced that they intended henceforth to 'act on the assumption that there is a connection between exposure to power frequency magnetic fields and cancer, in particular childhood cancer.'"

This action was prompted by two studies at Sweden's Karolinska Institute, one of childhood leukemia and the other of occupational exposure and brain cancer. Both claim to have found a connection between cancer and EMF. These studies contradicted two British studies released in 1990 and 1991 that found no such connection, yet the British studies go unmentioned in Brodeur's work.

Indeed, they are virtually unmentioned anywhere in the popular media. A Nexis computer database search a few months after the Swedes made their announcements found that while 81 media outlets mentioned the Swedish studies, only one mentioned the two British ones, even though the British studies appeared in a peer-reviewed medical journal (the British Journal of Cancer) while the Swedish study was released directly to the media with no review. Rather than go the usual route of peer review and publication in a medical or science journal, the Karolinska Institute did an end run around the usual careful if stodgy system by simply handing an abstract of their findings over to reporters at a San Diego conference on EMF. Many members of the media appear to be unaware that epidemiological studies often conflict, and simply took--or at least represented in their articles--these to be essentially the final word on the subject.

Though the popular media refer to these Swedish studies as more or less definitive, EMF authorities do not. In a letter to Science in July 1992, the Oak Ridge panel noted glaring inconsistencies between the two Karolinska studies and an earlier Swedish study from 1986.

The 1986 one noted an inverse relationship between EMF and childhood leukemia, for example. On the other hand, it did find a correlation between EMF and childhood brain cancers while the later ones found none. The Oak Ridge publics went on to find contrast after contrast. Epidemiologists generally believe conflicting positive results should be viewed as negative overall.

In the eyes of an epidemiologist, the inability of each of the other studies to confirm one specific connection lowers the possibility of that connection being real. Gerald Draper, the director of the Childhood Cancer Research Group at Oxford University, pointed this out in an editorial in the British Medical Journal. "The possibility that magnetic fields associated with electricity transmission may cause Some cases of childhood cancer cannot be dismissed," he wrote, "but the lack of consistency among published studies, and the absence of an accepted biological explanation for such a relation, means that we have to conclude that at present no causal relation has been established."

It's impossible to know just how much EMF exposure a child received, so some system must be used to estimate it. The first study to link childhood leukemia and EMF from power lines was conducted in Denver by psychologist Nancy Wertheimer and her physicist friend, Ed Leeper. Unable to actually enter people's homes, they instead measured the distance between distribution lines and the homes themselves.

Using what they called wire Code methodology, they and some of their successors, such as Dr. David Savitz of the University of North Carolina, found increased cancers with increased exposure. But Savitz, using spot measurements in the house, found that with this method there was no correlation between cancer and exposure. Dr. Stephanie London in Los Angeles added yet another method of measurement, hanging a detector around children's necks to measure their actual exposure. She found no correlation between actual exposures and EMF, nor between spot measurements and EMF, but did find a correlation using the Wertheimer-Leeper wire code. Similarly, Dr. Anders Ahlbom, chief researcher behind the Swedish studies, found that actual EMF measurements in homes, as opposed to his calculations of the exposure he thinks the individuals received over several years, correlate to a reduction in leukemia risk.

It may be that the coding system is truly best. It may be, as Savitz has suggested, that present-day spot measurements of EMF can't reflect fluctuations in fields that might have happened over time. But to say that London and Ahlbom have found an EMF-cancer correlation, as Brodeur and others have said, is to simply assume that the wire code is the best measurement tool.

No one knows if that's the case, however. This led Savitz, in his April 1993 overview of EMF studies in the American Industrial Hygiene Association Journal, to write "a number of inconsistencies and questions prevent a conclusion that magnetic fields cause cancer, including the failure of measured fields to be as strongly associated with cancer risk as are wire codes."

More recent studies in the October 1993 British Medical Journal, larger than the Swedish ones, also cast doubt on the EMF-cancer link. Just as the media ignored foreign studies prior to the Swedish ones and downplayed the negative associations in those studies, they likewise ignored these, from Sweden's neighbors, Finland and Denmark. Both were vastly larger than the Swedish one. The Danish one had 1,707 children with tumors and almost 5,000 controls, while the Finnish one followed essentially the entire child population of the country, over 150,000.

The Finnish study found no statistically significant correlation between EMF and all types of cancer combined or any individual cancer, except when it then broke the individual cancers down by gender, when it found a correlation to high EMF exposure and nervous system tumors in males. That correlation stood on the finding of three tumors in a single boy. By changing the analysis' to per-patient instead of per-tumor, even that connection disappeared.

But What If?

Ironically, one of the solutions that occur to most people concerned about EMF--placing themselves further away from the source--could theoretically make things worse. That's because some of the studies finding positive EMF-cancer correlations indicate that low levels of exposure are worse than higher ones. This seems strange, because the usual role of toxicity is the more exposure, the worse off you are. As one Carnegie-Mellon researcher, Indira Nair, explains, our conception of what causes cancer is based on our observation of chemical and ionizing radiation, such as that from X-rays or atom bombs, harm inducers. Assuming EMF are harmful, they would he inducing harm in a completely different way, by interfering with the timing of certain cell operations rather than by tearing apart molecules as carcinogenic chemicals do.

So when Paul Brodeur makes one of his television appearances with a gaussmeter (which measures EMF emissions) and urges viewers to get their own and avoid high-emission areas, even assuming his claims of the hazards of EMF are true, he would still be giving useless or even dangerous advice.

There are plenty of reasons to doubt an EMF-cancer link, as presented above. But even if it is true, what is the worst case scenario? Neighborhoods and schools exploding with cancer and other illnesses that Brodeur describes? Hardly. Sweden, with a population of 8 million, has about 70 cases of childhood leukemia a year. Dr. Ahlbom, chief author of the Swedish study that Brodeur so heavily relies on, nonetheless told The New York Times, "We're debating whether or not at most one childhood leukemia per year out of those 70 might be attributed to high voltage power lines." Sweden's population is about one 30th that of the United States, so that one Swedish leukemia would extrapolate to 30 American ones. Childhood leukemia is curable about 70 percent of the time, meaning the Ahlbom formula would attribute nine American children's deaths to power line-induced leukemia.

Dr. Jorgen Olsen, author of the aforementioned Danish study, also said that even if power lines did cause cancer, EMF would be responsible for "one case every five years in Denmark, which has a population of five million." That figure, extrapointed to the United States, would be about 10 cases a year, and three deaths.

Further evidence that the number of childhood leukemia cases caused by EMF in the United States, if any, is quite tiny is the utter lack of any correspondingly large rise in leukemia cases with the 438 percent increase in per capita electricity use in this country from 1950-87. While there has also been an increase in per capita childhood leukemia cases reported nationally during this period, it is less than one 10th the rate of increase of electricity use, and most or all cases may he due to more complete reporting. Connecticut's tumor registry, the first truly comprehensive program in the country, has seen virtually no increase in childhood cancers overall or childhood leukemia specifically.

Reluctant Sources

The scientific debate about possible EMF risk is ongoing. But it is those on whom Brodeur depends for scientific backing who are most critical of his writing, presumably because they feel their work is being misappropriated.

To read Brodeur, you would think Sam Milham has provided the smoking gun linking worker exposure to EMF and cancer. Yet the Worcester (Mass.) Sunday Telegram cited Milham as saying, "My mind is still open on whether the general public has a significant risk." But, he said, "to put things in perspective, I don't think the risk is an inordinate one. In other words, a whole lot of cancer can't be related to electric fields and power lines."

David Savitz is probably the best-known EMF epidemiologist. He authored a Denver study linking power lines to childhood leukemia and is cited repeatedly in The New Yorker and Currents of Death. Brodeur writes, "Far and away the greatest blow to the effort of the utility industry to deny that 60-hertz electric and magnetic fields could pose a health hazard came on November 20, 1986, when Savitz and his colleagues announced the results of their long-awaited...study."

But Savitz panned Currents of Death in his review in the Journal of the American Medical Association, writing: "If held to the standard of a balanced, scientific review of the evidence, Currents of Death simply falls. There is little evident attempt to separate legitimate criticism of the scientific evidence suggesting health harm from irrational resistance based on the implications of that evidence. Personal or institutional bias is invoked as the only possible reason for failure to accept what the author (but few others) considers irrefutable evidence. Deficiencies in some studies are not mentioned, while contradictory evidence receives strong criticism."

Indeed, in response to the Brodeur EMF scare, Savitz felt compelled to issue an open letter to "Persons concerned about reports of electromagnetic fields and childhood cancer," In it, he stressed the inconclusive results of his study and stated that "interest or concern may be justified, but our study is not sufficiently convincing to warrant drastic action by homeowners."

That statement, like Milham's, has never made it into any of Brodeur's works.

Each child's death is a tragedy, but there are many potential tragedies parents must worry about that are both far more certain and far greater than EMF exposure. Each year, 2,000 people between the ages of 5 and 14 are killed in traffic accidents, with 120,000 injured and often crippled for life. Most of these deaths and injuries could have been prevented through use of safety restraints. Trichopoulos, the Harvard epidemiologist, notes that it doesn't take much imagination to realize it is far easier and less expensive to reduce that figure than to try to reduce a number approaching zero to an even lower number. Ahlbom himself has said, "That's the type of information that's very difficult for people to grasp. People just think, 'It's a risk, and if it's a risk it must be significant.' That's not the case." He added, "Getting across how small these numbers are is extremely difficult." Especially when a crusading reporter at an influential magazine has made it his life's ambition to do just the opposite.

Michael Fumento is the 1994 Warren Brookes Fellow in Environmental Journalism with the Competitive Enterprise Institute in Washington, D.C., and author of The Myth of Heterosexual AIDS (Regnery Gateway) and Science Under Siege: Balancing Technology and the Environment (William Morrow).