Hyatt Regency in Phoenix, AZ, April 6–9, 1998
The seating is a square arrangement of tables, with the committee members assigned to specific locations. The chairman sits on one side, flanked by two committee members who have been assigned to write a contemporaneous record of the committee’s deliberations. They are called rapporteurs, with the word given its French pronunciation. The chairman and the rapporteurs occupy one entire edge of the table arrangement, which creates a teacher–student spatial arrangement.
Each committee member’s seat is marked with a large sign that gives his name and affiliation. Both pieces of information are written on each side of a placard that is folded in the middle so that it sits well on the table and can be read by the other committee members regardless of where they sit. The placards carry the logo of the NIEHS but do not indicate the background or professional status of the committee members, most of whom are strangers to me. They could be Ph.D.’s, M.D.’s, D.V.M.’s, or in the case of the foreigners, could have other kinds of academic degrees. The placards don’t specify what the committee members do at the place where they work. It would have been nice to know where they stood at home.
The attention to detail is impressive. Little is left to chance. To ensure that scientists assigned to particular committees go to the right room, little colored dots have been placed on each participant’s name badge and the corresponding colors have been posted on the walls outside the meeting rooms. The rooms all have names, but in case we didn’t read them we could follow the colored dots.
The NIEHS provided an unlimited supply of coffee, cokes, tea, and bottled water, and a limited supply of muffins and cookies, and a $39 a day food allowance. The chairman and the rapporteurs get an additional $150.
The big dog is Christopher Portier, who was appointed by Dr. Olden to organize and implement the RAPID program. Portier is a statistician who specializes in analyzing data to ascertain any risks from various pollutants in the environment. He projects an aura of keen interest in his task, someone who will listen attentively to the various points of view, but someone who will not act imprudently or allow himself to be seen as a champion of one or other viewpoint.
Portier is a small thin man, with a ring of thick brown curly hair that girds his head like a halo. He looks like a balding Leonardo de Caprio. The default position on his face is a mild smirk, mediated by a slight curve of his mouth on the left side. The expression is most pronounced when he is confronted with an issue that he had not previously considered. In that situation he smirks but says nothing.
He seems able to appreciate the important points in science articles in fields outside his experience, and he seems politically aware. As a federal bureaucrat, for purposes of comity he must cooperate with an alphabet soup of other federal agencies that have some connection with the EMF area. He might think that the bureaucrats from the Department of Energy are in bed with the power industry, but nevertheless he must treat them like colleagues and equals. He does this job admirably, rarely betraying his true feelings about the scientific sophistication of the other federal agencies. I think he thinks they’re stupid, and that he has an abiding contempt for the research program of the Electric Power Research Institute. Even so, the law that authorized the RAPID program required the NIEHS to cooperate with the power industry in seeking the truth about health hazards of EMFs. Unfortunately, this meant that the power company position and its particular spin on the scientific evidence were invited inside the tent.
Portier is a committee guy. He operates by appointing advisory groups whose chief role is to inform him and to act as his sounding board. Within the NIEHS, he has quite a small staff to orchestrate his inquiry. Several persons from outside NIEHS but inside the federal government are listed as part of Portier’s staff. But taken together, the group is largely for logistical purposes, and has no responsibility to directly evaluate the question posed by Congress to Olden and delegated to Portier.
Portier divided the EMF studies into three main areas, each of which was the subject of a meeting. Here in Phoenix the reports were divided into ten areas, some broad (brain cancer), some probably important but hard to define (immunotoxicity), and some obscure and probably irrelevant (cell calcium). Portier personally chose each invited participant to the meeting. The meeting was open to the public—O.J. Simpson or Monica Lewinsky could have come had they chosen to do so—but only the invitees were permitted to opine regarding EMF hazards.
Why would the invitees come to Phoenix to attend the meeting? Well, it was an all-expense-paid trip to a nice location just to talk about science. Second, like me, many of the invitees were NIEHS grantees. Having received from several hundred thousand to more than a million dollars to perform research, it would seem downright ungrateful and I think just plain wrong to refuse to attend the meeting. Third, particularly for the foreign scientists, one can imagine how honored they must have felt when asked to come to the United States for the purpose of giving their opinions.
Portier not only invited foreign scientists, he placed them in control of the meeting. It was not unusual, for example, for someone from Finland, Switzerland, or Quebec to be afforded a prominent role in shaping the discussion of a particular scientific point. Portier could be certain that the foreign scientists would not complicate his life by stirring the pot of competing interests regarding EMFs that exist in the United States. In most other countries in the world, I think that the question posed by Congress must look amusing. No other country is concerned about the health hazards of EMFs to the extent that the United States is concerned. No other country has spent as much money, litigated and politicked as much, published as many scientific reports, or generally speaking, gotten as exercised by powerline EMFs. There seems to be something peculiarly American in the idea that it’s wrong for companies to save money by involuntarily exposing people to electromagnetic fields on the basis of a claim that the practice is safe, in the opinion of the scientists who work for the industry. There is no significant opposition to this idea any place outside the United States, except perhaps Sweden.
The foreign scientists seem to view the EMF question more like an academic exercise, than as a serious attempt to ascertain what is or is not safe for the American public. The foreign scientists laugh a lot and take things in stride, as if there were nothing serious going on. They already made their bones just by getting here.
Portier designed a clever organizational scheme to control the meeting and manage the scientists, which generally is like trying to herd cats. He created ten committees which he called BOGs, one for each area he had identified. Each invitee was appointed to three BOGs.
Prior to each BOG meeting, each BOG member was given a list of scientific articles that Portier considered suitable for discussion. No one actually said that articles other than those chosen by Portier couldn’t be considered. It was simply that, from a practical viewpoint, it would have been mighty tough to do that.
What is a BOG meeting like? We sit around the table like a bunch of children at a back yard barbecue for our grandfather, Portier. When he speaks, we all listen. When he wants something written down, it is written. When he wants to discuss something in detail, we talk about it endlessly. When he is finished, we are finished. Nothing about the discussion would knock your socks off from a scientific perspective. “Have you measured such-and-such?” one member would ask another member. “Yes, we measured it but there was no effect.” “Okay, write down ‘no effect,’” Portier says, and we move on.
The process had no discipline. Most members who spoke, were obviously reading the reports at the meeting for the first time. There was no proportion between the seriousness of our endeavor and the process by which it was carried out.
Portier himself didn’t chair any of the BOGs. But he attended many of them, and when he was present, he dominated, like a President or Pope. The other scientists would address questions to him about particular reports or about scientific procedure. If he wanted something noted, it appeared in the report. If he wanted something omitted, it was omitted.
It is easy to understand why a sense of obsequiousness pervaded the room when Portier was present. Who were there? People from his staff. People from his advisory committee. People from other federal agencies. Foreign scientists. Industry scientists, and a handful of other scientists who more-or-less admitted they knew little about the subject area of the meeting. Portier was in a position to get exactly what he wanted from that meeting. If he had wanted us to take a stand against beer in cans, no problem, we would have done it. Portier said that he “wants to capture diversity of opinion,” but how do you do that when you choose who is attending the meeting, tell them what to consider, and arrange for the people who will write the history? What you get from that process, I think, is what you want.
He said he didn’t want the BOGs to be consensus-seeking committees, but rather committees that would provide him with a full range of opinion on a particular point. Nevertheless, every signal he sent was to the effect that we were to create a consensus of the scientists present.
In response to a question, Portier said that most of the chemicals presently recognized by his agency as harmful to the immune system do not have a well described mechanism of action. Thus, something can be harmful without being understood mechanistically. Nevertheless, Portier organized the meeting so that there was a tight link between the question of mechanisms of EMFs (of which there are none known) and the question of whether EMFs affect human health. The gist of his strategy was to suggest that knowledge of mechanisms was somehow important in judging health risks, and that a firm conclusion regarding the latter couldn’t be made in the absence of knowledge of mechanisms. The term mechanism was the second most frequently used word at the meeting (after inconsistent as in “inconsistent results”). I was particularly offended by this aspect of Portier’s agenda because it was hypocritical.
Portier had a deeply inconsistent attitude concerning how to make scientific decisions. At the beginning of the meeting he gave a long detailed talk about his views of risk assessment, accompanied by a handout that contained hard copies of his slides. They depicted an orderly, objective, rigorous decisional process. As if to temper this hard edge, on one of his slides he wrote “Experiments don’t speak for themselves, we have to interpret them.” Great. How true. But how do we interpret them? At one point, during the discussion of a particular paper in which an investigator had reported a statistically significant biological effect due to EMFs, Portier said “It’s not enough to find effects, the effects have to be something you believe.”
During my second BOG I asked Portier, “Suppose we had two studies, one of which showed that the measured parameter was statistically significantly increased due to EMF exposure, and a second independent replicate that showed the opposite result (statistically significant decrease). Are those results inconsistent?” “Yes, to me those results are inconsistent,” he said. “Well,” I said, “suppose my hypothesis was that EMFs affected the parameter, and that I had no hypothesis whatsoever regarding the direction of the effect? My idea is that EMFs will be transduced and, because the system is nonlinear, the dependent variable may be increased or decreased. Wouldn’t you agree that, with this model and this hypothesis, if the postulated results were observed, then the results should properly be labeled ‘consistent?’” Portier thought for a few moments and then said “Let’s put Andy’s concern aside and go on.” Portier said he was doing it because my question was “too theoretical.” But it wasn’t theoretical, it happens all the time. I think that was the moment I saw that the NIEHS effort would fail.
On the Flight Home
Throughout the sessions, confusion was obvious regarding both scientific reasoning within the context of particular studies, and how the results of groups of studies were to be generalized for the purpose of drawing an overall conclusion. One major problem involved the supposed importance of a dose-response relationship as a criteria of validity of studies. The relationship between the establishment of a mechanism and the establishment of validity of empirical data was another major problem. Some BOG members were sensitive thoughtful persons, but most held views regarding scientific reasoning that could have been shot down flat in an open debate. But real dialogue rarely took place because disagreeing with colleagues was generally regarded as rude. Pointing out specifically why you thought a colleague was dead wrong was unthinkable.
The meeting contained a cross-section of scientists, most of whom were narrowly educated in a particular specialty, and almost none of whom were in a position to see the big picture. Each of them was like someone in Plato’s cave, chained in such a way that they could see only two-dimensional shadows, and not the three-dimensional reality that gave rise to the shadows.
When confronted with the problem of how to find scientific truth about public-health hazards of EMFs, Portier’s first step should have been to analyze all previous attempts to accomplish this task, with an eye to discern why those attempts failed. One needs to talk to Portier for only five minutes to recognize that he has only the dimmest idea of the history of the problem he is attempting to solve. Being unaware of all of the previous mistakes, he is fated to repeat them.
Reflection a Month Later
One day, the Director called Portier and told him that he was to carry the water on the EMF project. It must have been an intoxicating day for Portier, but when he sobered up he must have realized that Congress is a political animal, that the laws passed are in response to political pressures, and therefore that there was a history regarding the issue with which he was charged. If he had paused and asked why, for the first time in history, Congress had addressed a specific health-related question to the National Institutes of Health and appropriated funds for the research necessary to answer the question, he might have evolved a more prudent and reasonable decisional scheme. If he had looked at the reasons for the failures of the other blue-ribbon committees, he could have remedied them in the process he was charged to design and implement.
What could have been done? First, there needed to be a recognition that, at the end of the day, there will be winners and losers when the question whether EMFs affect human health is answered affirmatively or negatively. An affirmative answer would cost the power companies money. Not only would they be required to widen some rights-of-way or underground the powerline, they would probably be sued by people who they previously exposed to EMFs and who got sick. On the other hand, if the question were answered negatively, then the people who live beside powerlines would simply have to live with that situation. Some would get sick because of it. They would not like the situation, but they would be stuck with it. A basic sense of justice tells you that wherever there will be winners and losers, there needs to be a fair procedure whereby people on each side of an issue can challenge the reasoning and values of those on the other side.
The challenge process doesn’t have to be formal cross-examination, but a winners-and-losers situation requires some recognition of the inherent adversarial nature of the situation, and an opportunity for one side to attack the spin on the evidence produced by the other side. The alternative to an adversarial process is a consensus process, and we already know that consensus processes don’t work in the EMF area—that’s why Congress authorized the RAPID.
Portier did the opposite of what was best for the general public. He brought together a handful of scientists, many of whom were out of place in the BOGs to which they were assigned, and he forced the process toward a pre-ordained consensus. As I sat there I thought, “Why the hell should I talk? This guy doesn’t want to hear what I’ve got to say.” All of the papers he assigned me to read had severe inferential limitations. I kept thinking that I wouldn’t want some government committee making a decision on scientific evidence of the kind that might affect my family on the basis of a loosey-goosey generalized discussion by people who manifested various degrees of preparation, who were never asked to explain why they had a particular opinion, and who regarded it as impolite to ask someone to explain the basis for their opinion.
Past blue-ribbon committees usually summarized their work using terms that sounded definitive and clear to the layman, but which, on analysis, were quite the opposite. For example, there is no “convincing” evidence that EMFs are health risks, or EMF health risks have not been “proven,” or EMF effects are not “robust” or “cause-and-effect relationships” have not been demonstrated. These simple-sounding terms are profoundly complex, and highly subjective. It’s quite possible that I could be “convinced” that EMFs were health risks by certain items of evidence that are not sufficient to convince a power company scientist or stockholder. Portier repeated all of these mistakes in the documents and guidance that he provided to the symposium participants, and he even added some additional terms that had not appeared prominently in the EMF dispute. One example is “immunotoxicity.” Is something immunotoxic if it causes any change in immune systems of animals? Only if it reduces the endpoint? Is it necessary to go further and show effects of EMFs on host-resistance endpoints? If the EMF caused a change in an immune endpoint, would that justify concluding that it “affects human health?” If not, what would be the state of the immunotoxicology literature that would justify that conclusion?
Another frequently used word that seems at first glance to have a specific meaning but ultimately turns out to be subjective is abnormal. No one could define the term in any meaningful or objective manner, and there was never any consensus regarding what it meant in the context of laboratory studies. Even the significance of the distinction normal-abnormal was obscure. Should investigators search for EMF bioeffects under the assumption that if such bioeffects occur and can be imputed to human beings, then it will be assumed that they are adverse for the subjects? Alternatively, as the power-company spokesmen argue, should the committee members be looking for biological effects in animals? Then, if they are found, determine which of those effects are abnormal? Then, determine which of those abnormal bioeffects can be imputed to exposed human subjects? Nobody knows.
There was much confusion regarding the argument that the observed effects were “small.” The term had different connotations in different BOGs. Sometimes it referred to the difference between an exposed and a control group in relationship to the difference between control groups from different laboratories where the experiment was replicated. In other cases, it meant that the measured parameter was within the range of measurements that is ordinarily considered to be normal for the parameter in the species. In still other cases the term meant that the difference between the exposed and control group in the EMF experiment was small in comparison with the effect produced by the investigator’s favorite chemical. If that chemical produced a difference in means of 1000%, and the EMF produced a difference of 50%, then the effect of EMFs was “small.” The point is that there desperately needs to be a determination of what the reasoning rules will be prior to evaluating the evidence. Are we to regard EMFs as affecting human health if they produce any change, or do we require that the change be bigger than X%, or outside the normal range, or greater than inter-laboratory variation in control groups, or what?
At the meeting, like good little soldiers, we voted 100% that EMFs had not been “conclusively” shown to cause skeletal abnormalities in chick eggs. We then voted 100% that EMFs had not been “conclusively” proved to cause birth defects in animals. Most of the committee members voted to say that the results were “equivocal.” But the words were never defined. Committee members always seemed to avoid defining terms that had decisional impact. Terms like “robust,” or “equivocal,” or “controversial,” or “inconsistent,” or “cause,” were never defined, despite their enormous importance in conveying the committee’s conclusions. What, for example, is the public to understand by the conclusion that the effects of EMFs on fetal development are “equivocal?” Does that mean that it is okay to buy a house beside a powerline? Does it mean that there is no likelihood that a pregnant woman living beside a powerline will have a spontaneous abortion or give birth to a malformed child, at least partially as a result of the magnetic field from the nearby powerline? The BOG did not infuse the statement with any substantive meaning, and therefore it is unreasonable to expect that someone who reviews the BOG’s work product could do so.
During one meeting, a committee member tried to help the group decide whether EMFs affected reproduction in animals. He had a copy of the report of the 1997 NAS committee which had a Table listing positive and negative reports that showed that 30% of the reports were positive. On that basis, the committee voted to say that the results were “equivocal.”
It was easy for a chairman of a committee to implement his own agenda or impose his own bias. In these cases the chairman did Portier’s dirty work instinctively, not out of a conscious intention to be obsequious. In one committee, the chairman did not accept the existence of cause-effect relationships in biological systems unless they were 100% certain. He said so clearly. Effects were “definite” or there were no effects. Nothing in between. At one point he asked the committee to vote on whether or not the Henhouse studies showed that EMFs “could probably affect skeletal development in eggs.” I asked him to clarify whether we were to vote on whether or not we thought the effects were “probably definite” or “definitely probable.” Everybody laughed, but I wasn’t trying to be funny.
That chairman is a good human being, the kind of guy you’d like to have for a neighbor. He would come over and help you move a refrigerator, he would attend your mother’s funeral even though he never knew her, simply because you were his friend, and he would share with you the tomatoes in his garden. He doesn’t kick his dog, his TV isn’t too loud, and his kids don’t have pierced tongues or pink hair. It’s not that he isn’t a nice guy. He simply isn’t qualified.
Congress charged the Director to find out whether EMFs “affect human health.” What exactly does that mean? What is the state of the evidence that would warrant an affirmative answer? In my mind, this issue must be resolved before the evidence is discussed. As Portier looks at it, the issue doesn’t have to be discussed at all.
There’s something un-American about a process in which one man controls all important events, is not subject to any meaningful checks and balances, appoints himself as both investigator and judge, and renders a decision from which there is no appeal that has a pervasive effect on society, affecting the daily lives of hundreds of millions of people.
What do I expect? The problem should be framed in unequivocal language, the standard for decision-making should be defined, all relevant evidence should be presented and tested by adverse parties, and disinterested persons should judge the evidence and regulate the fairness of the process. Instead, we were presented with Portier’s view of the world, which will surely be endorsed by the Director and sent to Congress. Decisions affecting the public interest ought not be made by one man, regardless how smart or honest he is. No one is that good.
It is clear that Congress’ attempt to resolve the question of health hazards of powerlines by assigning the question to the NIEHS is doomed to fail. You simply can’t throw $65 million at a problem and tell somebody “fix it.” You must also specify what “fix” means and how it is to be assessed. Otherwise, the money will be spent pursuing idiosyncratic notions of “fix,” and then the putative “fixer” will simply come back and ask for more money. I think Congress is unlikely to repeat this mistake again, and hence a search must commence for other mechanisms by which scientific data can be taken over into the public domain. The NIH deals in more or less certain science. It is ill-equipped to handle the inherently adversarial issues where things are not and cannot be certain. NIH has not mechanism, staff, nor tradition for resolving scientific disputes such as whether EMFs affect human health. That infirmity would probably extends to any dispute where the public health is allegedly impaired by a pollutant under the control of an economic interest.
What will be the final result of the present process? A poorly documented, diffuse, vague, wishy-washy report in which terms are not defined, procedures are not specified, the ipse dixit of scientists is presented as fact, with the conclusion that a link between cancer and EMFs has not been conclusively proven.