Public concern has grown in recent years over the possibility that subtle or delayed adverse health effects might result from exposure to power-frequency magnetic fields surrounding transmission and distribution lines and the electrical devices that have become common in residences and workplaces. Concern over the possible health effects of exposure to low-intensity, 60-hertz (60-Hz) power-frequency magnetic fields was a driving force in setting research agendas for government and private organizations and led to a series of workshops held in 1990–1992 with participation by representatives of the Department of Energy (DOE), the National Institute for Occupational Safety and Health (NIOSH), the Environmental Protection Agency (EPA), the Electric Power Research Institute (EPRI), public utilities, state governments, and the scientific community. The workshops were designed to elicit strategies for research in the biologic effects of magnetic field (MF)1 exposure, and various methods of disseminating research findings to the public were considered. The workshops provided the basic framework for establishment of a national program in electric and magnetic field research that was ultimately authorized by Congress in the Energy Policy Act of 1992 (Public Law 102–486). This program is commonly called EMF-RAPID (Electric and Magnetic Fields Research and Public Information Dissemination). An effort was made to ensure that the EMF-RAPID research activity was coordinated and targeted to produce answers to a number of important questions raised in the Energy Policy Act:
Determine whether or not exposures to electric and magnetic fields produced by the generation, transmission, and use of electrical energy affect human health;
carry out research, development, and demonstration with respect to technologies to mitigate any adverse human health effects; and
provide for dissemination of information . . . to the public".
Charge to the committee
In response to a request from the DOE, following the directives of the Energy Policy Act of 1992, the National Research Council established a committee of scientists and engineers to review the activities conducted under the EMF-RAPID program. The Research Council committee, the Committee to Review the Research Activities Completed Under the Energy Policy Act of 1992 (known as the EPACT Committee), also was asked to review the research agenda and strategies adopted by DOE and the National Institute of Environmental Health Sciences (NIEHS), and those suggested by other federal and nonfederal groups. Specifically, the EPACT Committee was asked to:
Review and evaluate the scientific and technical content of projects completed under the EMF-RAPID program. The results of this review, with implications of the effect and significance of the work in addressing the issues defined in the national EMF research plan, are to be reported to DOE, the EMF Interagency Committee (IAC), and the National EMF Advisory Committee (NEMFAC).
Review continuing research activities for scientific content and suitability to meet the goals of the EMF-RAPID plan. Continuing projects will be given only a brief review with regard to content and direction, based on a brief statement of scope, goals, and progress submitted to the committee using a format developed specifically for that purpose to assess their potential to fulfill the goals of the national program.
Recommend modifications to the EMF-RAPID program, as appropriate, based on information the committee acquires concerning new research findings not available when the national research agenda was developed, from briefings the committee might request, and from analysis of research strategies developed by other groups.
Assess the scientific and technical content, and make recommendations as deemed necessary, for activities initiated under the EMF-RAPID agenda to promote the transfer of information derived from research projects.
This report, submitted at the conclusion of the EMF-RAPID program, reviews and evaluates the scientific and technical content of projects completed under the EMF-RAPID program and makes recommendations regarding the transfer of information derived from research projects. The focus of this report is on the research conducted within the EMF-RAPID program. Other work is discussed only in the context of important findings that were made in projects conducted as part of the EMF-RAPID program. An interim report, EMF Research Activities Completed Under the Energy Policy Act of 1992, was issued by this committee in 1995.
The EMF-RAPID program faced many organizational and administrative obstacles. For example, there was a delay of about 2 yrs in funding implementation, and total funding (about $41 million) fell substantially short of the original planned level ($65 million). Without the delay additional peer-reviewed publications might have been available as EMF-RAPID came to a close for appropriate agencies to evaluate potential health effects. If the additional monies had been available, it would have been possible to include additional important activities such as focused interlaboratory replication projects. Furthermore, the program used the National Institutes of Health (NIH) grant process for the biology studies administered by NIEHS. That process, as implemented in this program, did not lead to complete reports at project completion, because the goal was to produce papers in the peer-reviewed literature and this goal was not always met by the individual projects. Project summaries available at program's end were of uneven quality, and the majority of the summaries did not provide complete reports of the research effort. The lack of complete reports and limited availability of published journal articles at the time the program ended made it difficult to judge the quality, completeness, or significance of the biologic studies funded by the EMF-RAPID program. The NIH-style approach also provided considerable flexibility to the principal investigators in
implementation of their research plans. Those complications reduced the information available in late 1998 and made it difficult to assess the usefulness and quality of the NIEHS portion of the EMF-RAPID program.
EMF-RAPID engineering research program
The EMF-RAPID engineering research program, administered by DOE, used a contract-funding mechanism; therefore, 11 complete and final reports were available to the EPACT Committee at the program's end. Many of the engineering studies were of little use, principally because they were initiated without a hypothesis related to what effect was to be studied. Furthermore, most of the reports were completed as the EMF-RAPID program was ending and so had little effect on the program as it developed. Projects that focused on developing methods and protocols for power-frequency magnetic field measurements were conducted simultaneously with projects designed to characterize human exposure, so the recommendations of protocol-development studies were of little value to the projects designed to characterize human exposure. Of the five projects concerned with characterizing human exposures in the home, workplace, and public areas, only the Enertech study of 1,000 persons (Enertech Consultants 1998b) provided a robust set of data that one could use to analyze the typical range of human exposure to MFs in the United States. The EMF database project is of no practical use at the time of this review, because no major data sets have been entered from potential users. Of the 11 projects completed at the time of the review and available to the committee, only two are regarded as noteworthy: the Enertech study (Enertech Consultants 1998b) and the Evaluation of Field-Reduction Technologies project (IIT 1997), concerned with mitigation of MF exposures, which yielded useful results on field-reduction technologies for a variety of common sources of power-frequency MF exposures. The other engineering studies are of value primarily as a compilation of state-of-the-art MF measurement practices.
The engineering studies contributed nothing directly to the question of health effects but did show that temporally-averaged MF exposures of subjects in a broad range of occupations and residential environments were remarkably similar. A major conclusion highlighted by the engineering effort is that the range of time-averaged exposures is very small, that most people are exposed to 0.1–0.2 microtesla (0.1–0.2 µT), 60 Hz MF, and very few to more than 0.4 µT. That finding demonstrates that it is extremely difficult to obtain large numbers of subjects for epidemiologic studies with substantially different, temporally-averaged MF exposures. The inability to identify heavily-exposed and minimally-exposed populations is a severe limitation in epidemiologic efforts to assess possible risks associated with MF exposure. In addition, the results of the engineering studies demonstrated that any MF-induced biologic effects would have to exhibit remarkably low response thresholds (about 0.1–0.2 µT) in order to have important implications for adverse human health effects, or alternatively, that such health effects would have to be shown to be caused by high but infrequent exposures.
EMF-RAPID scientific research program (biology)
Quality control at program initiation
The decision to provide biologists with engineering help for generation and characterization of MFs at the beginning of the program was important, as was the effort to encourage investigators to make greater use of conventional quality-control procedures, such as blinding, replication, and positive controls. This programmatic objective was a direct result of the practices developed by DOE as it conducted its bioelectromagnetics program in 1975–1995.
Categories of results at program completion
Biologic studies in the EMF-RAPID program can be grouped in five categories:
No MF results reported in project summary. These are projects that at the time of reporting had not produced either positive or negative findings.
No EF or MF effects reported. These are studies in which the investigators sought but were unable to find effects of EF or MF exposure.
Effects reported but data insufficient to calculate magnitude of effect.
EF or MF effects reported and data sufficient to calculate magnitude of effect.
Replications of earlier positive reports. These studies attempted to reproduce earlier reports of MF effects.
Value of studies for evaluation of possible power-frequency magnetic field health effects
Project reports in the first and third categories have limited value for evaluation of MF health effects. The second category contains projects which showed no EF or MF effects at the time of review. Many of the EMF-RAPID projects, instead of being well-defined studies to replicate earlier positive claims, investigated new biologic end points. Some of the reports available to the committee do not give the rationale for the research, and most of these studies had negative results. In the case of the fourth category, in which results were reported with supporting data sufficient to estimate magnitude of effect, the committee is unaware of replications of these studies in other laboratories. Because the existence of any biologic effects of low-level, power-frequency MFs is still in question, the fifth category, consisting of attempts to replicate previously reported positive effects, has been the most important part of the EMF-RAPID program.
It should be noted that normal exposures in the domestic setting are of the order of 0.1–0.2 µT. "High" exposures in epidemiology are > 0.4 µT. On that basis, biologic effect studies in the EMF-RAPID program used "high" exposures. Much of the program
employed fields 3 to 4 orders of magnitude greater than those which the general human population normally experiences.
The NIEHS biologic-research program made two important contributions that reduce somewhat the concern about whether the use of electric power might have adverse health effects. The first contribution was the effort to replicate previous reports of extremely-low-frequency (ELF) MF biologic effects, mostly through in vitro studies. All the attempted replications in the EMF-RAPID program have had negative or equivocal results. Because these replications were conducted in an environment of increased concern for field characterization and protocol development, their results are persuasive.
The second important contribution was the completion of several investigations of the relationship between MF exposure and cancer through controlled laboratory experiments in animal models. Nearly all the animal studies relevant to the EMF-cancer issue had negative results, even at field levels that were orders of magnitude greater than the levels typical of human exposure, including two of singular importance: the IIT Research Institute life-span exposure study examining cancer initiation by ELF MFs (NTP 1998a), a large screening study funded by NIEHS outside the EMF-RAPID program; and the EMF-RAPID-funded Battelle study on breast-cancer promotion in rats by ELF MFs (NTP 1998b), a replication of work by Loescher (Loescher and others 1993) which had previously reported positive findings. The outcomes of the animal experiments completed under EMF-RAPID, like those conducted elsewhere, do not support the hypothesis that MF exposure is involved in the carcinogenic process.
In vitro studies
Evaluation of the question of possible health effects of power-frequency MFs is handicapped by the absence of any robust effect that can be reproduced consistently from one independent laboratory to another. The EMF-RAPID in vitro replication studies further illustrate this problem. Their findings demonstrate that, without independent confirmation, caution must be exercised with regard to the weight that can be given to reports of effects, even if the effects appear to be large.
The in vitro results, for the most part, do not show effects that can be demonstrated as resulting from MF exposures at the field intensities that were explored. Few studies showed effects; the few reported effects were small and their connection to disease processes are speculative at best and irrelevant at worst. The results of several in vitro replication studies on gene expression were important in that they failed to support previous indications of a MF effect.
Many of the in vitro studies described did not develop surveys that permit generalization of any positive correlation between MFs and biologic response. The cell lines or the gene-expression assays were usually few and limited in scope or sensitivity. Some papers reported to be submitted or in preparation have not appeared, raising concerns about the reproducibility or quality of their data.
Is cancer a possible MF health effect?
The EMF-RAPID biologic research contributed little evidence to support the hypothesis that a link exists between MF and cancer. The results of the in vitro studies do not support an MF effect on cancer initiation, promotion, or progression, and they should be recognized as important studies in the overall evaluation of potential carcinogenic effects of MFs. Attempts to study the relationship between MFs and cancer in rodents have comprised more than a dozen studies including the EMF-RAPID-funded project at Battelle (NIEHS 1998a). The results of this latter study were negative in spite of the exposure of the animals in these experiments to extremely high MFs for long periods. The largest fields were 5,000 µT, which is more than four orders of magnitude greater than the field levels found in typical homes.
In contrast with the laboratory research, some epidemiologic studies have reported differences in the incidence of cancer associated with MF exposures that differ by as little as 0.2 to 0.4 µT. Estimates of MF exposures in epidemiology are crude at best. Because there is no support for a MF-cancer link in animal experiments where very large, carefully monitored fields have been used, occasional suggested epidemiologic associations between MFs and cancer are not supported by the EMF-RAPID animal data. It may be that the observed epidemiologic associations either are too weak to be significant or could be attributable to one or more confounding variables that have not been identified.
When the EMF-RAPID program began, emphasis was placed on two important phenomena—cancer promotion and gene-related effects in vitro. Experiments supported by EMF-RAPID provided some evidence to support, and considerable evidence to refute the view that power-frequency MFs can have biologic effects. Evidence of any robust and replicated effects on the development of cancer is lacking. That leaves the database slightly larger and the conclusions essentially unchanged from the Research Council's most recent review (NRC 1997).
The committee concludes that the NIEHS EMF-RAPID biologic research program would have benefited from a contract-funding approach with a requirement for complete reports and/or peer-reviewed publications at program's end. The investigator-initiated grant approach is excellent for basic research, but it is not effective for dealing with a time-limited, focused research program involving an applied-research question. The grant mechanism offers investigators freedom to change the focus of their research, provides too little opportunity for program management, and does not require timely, detailed reporting of results. As previously noted, the lack of complete reports and limited availability of published journal articles at the time the program ended made it difficult to judge the quality, completeness, or significance of the biologic studies funded by the EMF-RAPID program. At a minimum, complete, detailed, and accurate project reports should have been required of the investigators at program's end.
EMF-RAPID communication program
The communication effort initiated by EMF-RAPID is reasonable. The two booklets and the telephone information line are useful, as is the EMF-RAPID Internet site. There are two limitations to the effort. First, it is largely passive, responding to inquiries and providing information, rather than being active. Second, much of the information produced is in a scientific format not readily understandable by the public.
The three NIEHS review symposia (NIEHS 1997; NIEHS 1998b; NIEHS 1998c) provided an additional forum for communication among scientists and engineers interested in MF and health. A fourth review symposium presented projects completed in the EMF-RAPID engineering program (NIEHS 1998d). The committee concludes that the most important accomplishment of the three biology review symposia was to educate newcomers who were brought into the risk-assessment process being considered by NIEHS. The scientific value of the literature reviews produced by meetings is less than that of the 1997 Research Council report (NRC 1997), the NIEHS working-group report (NIEHS 1998a), or other published reviews.
The NIEHS working group2 produced an extensive, updated review of the entire literature (NIEHS 1998a) related to all aspects of research on the effects, if any, of MFs—a useful accomplishment that unfortunately was overshadowed by the use of the International Agency for Research on Cancer (IARC) method to review the status of MFs as a potential human carcinogen. Labeling power-frequency magnetic fields a class 2B human carcinogen (possible human carcinogen) conveys to the public a conclusion that our committee believes is not supported by the underlying research. The committee notes that no working-group members voted that MFs belonged in the "carcinogenic" classification (IARC 1) and no members voted for the "probably carcinogenic" classification (IARC 2A).
The Electric and Magnetic Fields Research and Public Information Dissemination program, in its relatively brief existence, has achieved a number of important objectives. There has been an increase in the research activity devoted to possible adverse effects of exposure to MFs on human health. An effort was made to ensure that this research activity was coordinated and targeted to produce answers to a number of important questions raised in the Energy Policy Act of 1992. From a technical perspective, the EMF-RAPID program has helped in an incremental manner to reduce uncertainty, strengthening the conclusion that power-frequency MFs are unlikely to have significant adverse effects on public health. The engineering studies underscore the fact that time-averaged MF exposures in a wide variety of occupations are remarkably similar, which has implications for the interpretation and design of epidemiologic investigations. Good science and good engineering make it essential to distinguish between repeatable results that correspond to testable hypotheses and empirical observations that are not predicted and not reproducible. Large amounts of data do nothing to strengthen claims of cause and effect that are fundamentally weak. The literature on laboratory studies of this
subject contains many conflicting claims. One way to deal with the situation is to subject the most promising positive reports to multiple independent replications. EMF-RAPID made a substantial start on that task. The largely negative results of the replication studies reduce the credibility of many of the original claims of MF effects.
The committee observed that several aspects of the national EMF research plan were not addressed, especially in epidemiologic studies and policy analysis. The lack of meaningful efforts in these fields (only one project was funded in each) probably was due to the lack of adequate time and to funding limitations.
An earlier Research Council assessment of the available body of information on biologic effects of power-frequency magnetic fields (NRC 1997) led to the conclusion "that the current body of evidence does not show that exposure to these fields presents a human health hazard. Specifically, no conclusive and consistent evidence shows that exposures to residential electric and magnetic fields produces cancer, adverse neurobehavioral effects, or reproductive and developmental effects". The new, largely unpublished contributions of the EMF-RAPID program are consistent with that conclusion. We conclude that no finding from the EMF-RAPID program alters the conclusions of the previous NRC review on the Possible Effects of Electromagnetic Fields on Biologic Systems (NRC 1997). In view of the negative outcomes of EMF-RAPID replication studies, it now appears even less likely that MFs in the normal domestic or occupational environment produce important health effects, including cancer.
The results of the EMF-RAPID program do not support the contention that the use of electricity poses a major unrecognized public-health danger. Basic research on the effects of power-frequency magnetic fields on cells and animals should continue, but a special research funding effort is not required. Investigators should compete for funding through traditional research-funding mechanisms. If future research on this subject is funded through such mechanisms, it should be limited to tests of well-defined mechanistic hypotheses or replications of reported positive effects. If carefully performed, such experiments will have value even if their results are negative. Special efforts should be made to communicate the conclusions of this effort to the general public effectively.
The following recommendations are made by the committee:
The committee recommends that no further special research program focused on possible health effects of power-frequency magnetic fields be funded.
If, however, Congress determines that another time-limited, focused research program on the health effects of power-frequency magnetic fields is warranted, the committee recommends that emphasis be placed on replications of studies that have yielded scientifically promising claims of effects and that have been reported in peer-reviewed journals. Such a program would benefit from the use of a contract-funding mechanism with a requirement for complete reports and/or peer-reviewed publications at program's end.
The committee recommends that no further engineering studies be funded unless a biologic effect that can be used to plan the engineering studies has been determined.
NIEHS should collect all future peer-reviewed information resulting from the EMF-RAPID biology projects and publish a summary report of such information periodically on the NIEHS Web site.
The committee recommends that further material produced to disseminate information on power-frequency magnetic fields be written for the general public in a clear fashion. The Web site should be made more user-friendly. The booklet Questions and Answers about EMF should be updated periodically and made available to the public.