[ EMF-Link Home
| Online Store
| Comments ]
This report reviews recently published articles and ongoing research on the potential adverse health effects from exposure to power frequency electric and magnetic fields. Over 300 studies, published between June 1992 and June 1993, are reviewed. They cover the areas of epidemiology, general human electromagnetic field (EMF) health effects and relevant research findings in the areas of animal tumor studies, experimental studies of biological, physiological, neurobehavioral and immunological effects, cell-level research, and engineering studies of exposure assessment and steps to be taken for mitigation of exposure. Where possible, conclusions are drawn regarding the significance of these research findings for human health issues. This concluding section provides an overall summary of the most significant findings assembled in the earlier sections.
While considerably more than half of the reported studies describe positive findings regarding effects of EMF, these effects are not completely consistent, and results obtained by one group cannot always be replicated by other researchers. A number of the studies outlined have in fact looked at the possible reasons for such inconsistency. In many cases variability may be ascribed to the limitations imposed by the exposure systems, inadequate definition of the exposure, stray EMFs entering the system, and "window effects" reflecting the apparent dependence of an effect on precise combinations of frequency and intensity. In other cases the variability is of a biological origin, as with chicken eggs, or due to quite unsuspected factors which may operate, as with the recent discovery of an interaction with fluorescent lighting in an animal skin cancer induction experiment. There is much ongoing research aimed at developing well-controlled EMF exposure systems, standardizing and characterizing exposures, and identifying potential interfering factors; it is to be hoped that this effort will help reduce the degree of inconsistency and allow firm conclusions to be reached in the future.
Epidemiologic Studies
The two major topics of recent epidemiologic research have been childhood leukemia in relation to residential exposure to the EMF fields from outside power lines, and adult leukemia and brain cancer in relation to occupational EMF exposure. The other areas of interest are the effects of EMF on breast cancer and on reproductive outcomes. Five major residential and ten occupational epidemiologic studies were reported this year from Denmark, Finland, France, the Netherlands, Norway, Sweden, Taiwan and the United States. More than half of them reported at least one positive association between EMF exposure and malignant disease, principally leukemia. Several of these studies attempted to overcome one of the major defects of earlier studies, namely the uncertainty regarding EMF exposure. In the Swedish residential study, historical records of power line load were used in an attempt to better estimate exposures during the critical period of tumor formation, five to fifteen years prior to diagnosis. However, despite this effort, the small number of cases in many of these studies, resulting from the relative rarity of the diseases of concern (childhood and adult leukemias, childhood brain cancer), prevented statistically strong conclusions from being drawn. Some of the major findings are highlighted below.
The studies covered in this Status Report (taken singly or together) still do not provide evidence of an association between EMF and health outcomes that is more conclusive than that presented in previous Reports. However, the results of several of the better epidemiologic studies are definitely suggestive of some effect and support earlier studies showing similar trends. The major strength of these studies is that they highlight the current limitations of using epidemiology to provide answers in this area. Research examining brain cancer and leukemia in workers, leukemia in children, and the risk for retarded fetal growth and spontaneous abortion is currently in progress or in press. Several groups are also examining measures of exposure to improve this aspect of study design for future epidemiologic studies. The latter will need to be larger, and more detailed knowledge of the exposure used will be necessary. These studies will also be improved as more knowledge of the basic biologic mechanisms involved is accumulated, allowing the use of intermediate end-points between the time of exposure and the occurrence of disease to shorten the time needed to accumulate enough cases for a valid study.
Residential studies
A Swedish study by Feychting and Ahlbom detected an association between childhood leukemia and residential exposure to power line EMF, with higher exposure over 2-3 mG as defined by historical records of power utilization. This was based on a population of approximately half a million individuals, consisting of all persons in Sweden who had lived within 300 meters of a high-voltage power line during the period 1960- 1985. One strength of this study is the quality of the data available to the researchers. Records of residence, cancer occurrence, locations of power lines and their electrical load over a long time period were linked together. A consistent measure of exposure, and better selection of cases and controls, are attributes of this study which have been lacking to one degree or another in previous studies. The fact that the positive association seen in this study is consistent with previous studies, is suggestive of an actual relationship between EMF and childhood cancer. However, the small number of cases studied (adults: 325 leukemias and 223 brain tumors; children: 39 leukemias, 33 brain tumors and 70 other cancers) precludes the possibility of any conclusive results. Interestingly, there was no correlation between childhood leukemia and spot measurements of local fields greater than 2 mG, supporting earlier studies and the value of wiring code as a reflection of actual power usage and EMF exposure.
Similar studies conducted in Denmark by Olsen, in the Netherlands by Schreiber et al., and in Finland by Verkasalo showed no strong associations between EMF exposure and cancers. These studies also suffered from the problem of there being too few cases of the diseases of concern at specific exposure levels to provide definitive answers.
Occupational studies
Like residential studies, previous occupational studies have suffered both from a lack of direct assessment of exposure and from the small number of cases of the diseases of interest available for study. Exposure estimates have typically been based upon job classification, which recent work has shown to frequently provide a misleading indication of actual exposures to EMF. Recent studies covered in this report have attempted to deal with these problems, with greater and lesser degrees of success.
The studies by Sahl et al. and Matanoski et al., carried out in California and New York City, respectively, used a double design (a case-control study within the context of a retrospective follow-up study), employed actual measurement of on-the-job magnetic field exposures to supplement employment and industrial hygiene records, and examined large populations (power company and telephone company employees). Sahl et al. found no strong associations between cancer (leukemia, lymphoma and brain cancer) and EMF, but Matanoski et al. found consistent associations between level and duration of exposure to EMF and leukemia. However, the authors of both studies cautioned against drawing strong conclusions from their results, since although the base populations from which their subjects were drawn are large, their cases were drawn from mortality records, and there were only a small number of deaths occurring due to the illnesses of interest. In Norway, Tynes et al. found no association between EMF exposure for electric railway workers and either leukemia or brain cancer. Bowman et al. found an increased risk for leukemia among workers in occupations classified as "electrical," but could find no associations or trends when estimates of EMF exposure were introduced into the evaluation. A study by Floderus in Sweden found an association between chronic lymphocytic leukemia and cancer, but no other leukemias or cancers; and a study in Denmark by Guenel found no strong associations. The Danish study was the only study reported here that examined breast cancer; no association was found.
Only one study covered in this report examined parental exposure to EMF and the occurrence of astrocytomas in their children (Kuijten et al.). Although significantly more fathers of cases were electrical or electronic equipment repairmen during the preconception period, no other significant associations were found. This lack of strong associations may have resulted from the low statistical power for some job groupings; the numbers of parents falling into these categories were too small. Since this study focused only on astrocytoma, the results cannot be applied to other types of brain cancer, although astrocytomas do account for about 70% of childhood brain tumors.
Animal Tumor Studies
Two major reviews of the EMF literature during the last year concluded that EMF is unlikely to have genotoxic effects and therefore unlikely to act as an initiator of carcinogenesis, although not all combinations of electric and magnetic field intensity have been evaluated for possible genotoxic activity. However, there is still much uncertainty regarding the ability of EMF to act as a tumor promoter by enhancing the effects of mutagenic chemicals or ionizing radiation. The data reported during the past year have been equally as contradictory as in previous years.
Even within particular research groups, the findings have been inconsistent. Scientists at the Bureau of Radiation and Medical Devices, Health and Welfare Canada, whose earlier work suggested that 60-Hz magnetic fields have a transient co-promoter activity in the chemical induction of skin tumors in mice, were unable to replicate their previous results, finding that exposure to florescent light may be critical to any promoter activity involving magnetic fields. A report from the School of Veterinary Medicine at Hannover, Germany, described a significant 50% increase in the incidence of mammary tumors in groups of 100 female rats treated with the initiator DMBA and exposed to a 50-Hz, 1-G magnetic field, whereas previous results using approximately 30 animals per group were not statistically significant. They believe that the critical factor in their experimental results is the use of a larger exposure facility which permits simultaneous exposures of 100 rats per experimental group.
Reproductive and Developmental Studies
In view of evidence for the role played by weak electrical gradients in controlling critical cell migration during embryonic development, it is not surprising that potential effects of EMF remain an active area of study. Except for one report describing a higher incidence of spontaneous abortion for those exposed to ELF magnetic fields above 9 mG, most recent studies do not support earlier reports of an increased incidence of miscarriages among women working with video display terminals. The general conclusion is that job stress is the major factor involved. A report on the medical evaluation of high-voltage station staff from the Ukraine also described increased incidence of abortion in women and of sexual dysfunction in men, accompanied by cardiovascular effects, but there was little definitive information provided on exposures either to EMF or to potentially toxic chemicals.
Recent studies with the developing chick embryo system, which has been an important experimental model since the early work of Delgado, have emphasized such critical factors as alterations in the frequency of the EMF, the need for an exposure at constant frequency for at least 10 seconds, and stray environmental fields, as well as the time at which the embryos are most sensitive (during gastrulation). There has been more emphasis on the use of mammalian embryos in recent research, but most experiments have not disclosed significant effects of EMF on their development.
Immunologic Studies
The early Soviet reports described an immunosuppressive action of EMF, but later work has not consistently confirmed this. It has been suggested that EMFs may interfere with the function of the immune system by some action at the cell membrane. However, it is difficult to come to any firm conclusions regarding the actions of EMFs on the immune system and their potential health implications. This reflects the many different measures of immunity that have been used in various studies, and the very inconsistent results obtained. The latter represents a major problem. Some recent studies have described a stimulatory effect on the immune system, as in a report in which exposure of rats to 60-Hz sinusoidal magnetic fields increased the ability of the natural killer (NK) cells to destroy foreign cells, and another in which implantation of magnets into rats restored the immune response to sheep red cells that had been diminished by removal of the pineal gland. On the other hand, sheep penned for prolonged periods (six months or more) underneath a 60-Hz power line showed evidence of decreased immune function. Results from experiments carried out in vitro, in which isolated lymphocytes were exposed to EMFs of various types, include both stimulation and inhibition of lymphocyte proliferation (a necessary early phase of immune response) by different laboratories, enhanced synthesis of interleukin-2 and immunoglobulin, and either inhibition or no effect on the activity of natural killer cells. Some of the variation may result from inconsistency in the underlying effects of EMF on calcium flow across the cell membrane. This flow is a necessary element in lymphocyte activation, and effects on the process increasingly appear to depend on specific windows of EMF frequency and amplitude. The stage of maturation of the lymphocyte preparations may also be a key factor in the variability.
Physiological and General Health Studies
Physiological and general health studies reported during the past year include general physiological effects in humans and a wide range of studies on circadian rhythms in both humans and animals.
A recent occupational study from Sweden of workers using video display terminals, reported the development of skin symptoms, and a rise in blood levels of the stress hormones adrenalin, thyroxine and prolactin, and a fall in testosterone, in those claiming hypersensitivity to the display terminals and electricity. Analysis of the data suggested that in fact skin symptoms resulted from non-specific job stress, for which the display terminals, so intrinsic a part of the work environment, served as a conditioned stimulus for the skin problem.
Another study involving intermittent exposure of healthy human volunteers to combined electric and magnetic fields in a sophisticated environmental exposure system at the Midwest Research Institute, produced statistically significant slowing of the heart rate, and psychological effects such as alterations in evoked response, which seem to depend primarily on the magnetic component of the EMF. Further, there was significant variation in the response to intermittent magnetic fields in different individuals, depending in part on the individual's heart rate prior to exposure and the change in systolic blood pressure. These effects on the heart fell within the range of normal physiological variations.
One of the major EMF bioeffects to be discovered in the past decade, the result of work by Wilson, Reiter, Lerchl, Semm, and others, is that high-level 60-Hz electric or magnetic fields and weak static magnetic fields can suppress the normal nocturnal rise in melatonin in animals and humans. This rhythm of melatonin production appears to have important regulatory functions in the body's natural circadian cycle. Disruption of normal circadian cycles is thought to have negative effects on behavior, including producing depression and disruption of work performance, and on basic biological functions, including the operation of the immune system and the body's ability to control cancer, so that a confirmed effect of EMF could have serious health repercussions.
The data in this area display significant variation. Despite this, a recent workshop review of experimental studies on rats, hamsters, sheep, baboons, and humans, concluded "that when EMF/melatonin effects are observed, they are always in the direction of melatonin reduction and that all levels of the biosynthetic pathway including precursors, enzymes and end products are consistently depressed." Experiments reported during the past year are generally supportive of an effect of EMF on pineal melatonin production, as for example in the dramatic suppression or delay of the nocturnal melatonin peak level in Djungarian hamsters given a single acute exposure for 15 minutes to a 1-G, 60-Hz magnetic field 2 hours before the onset of nighttime. These studies have also drawn attention to the many factors that can influence the results obtained, including length of the photoperiod, diet, climate, exercise level, environmental noise and disturbance, daily variations in melatonin levels and times during which the animals are exposed relative to light and dark periods. This means that the baseline hormone levels must be well established; in humans the full circadian cycle of melatonin levels should be determined with and without exposure to generate internal controls, while a cycle should be established in animals by killing them for analysis at appropriate intervals.
Neurological and Behavioral Studies
In contrast to the dramatic but poorly documented reports of psychological responses in the early Soviet literature, recent studies represent well-documented human responses to ELF magnetic fields. Such responses could possibly have subtle negative effects on human performance. Results from the Midwest Research Institute indicated alterations in evoked response induced by intermittent exposure of healthy human volunteers to combined electric and magnetic fields. Results reported from Canada indicated an ability of a 10-mG magnetic field, applied to the temporal region of the brain, to alter the choice of words (reflecting features such as activation, pleasantness, or evaluation) used by volunteers assigned to make up narratives spontaneously. This response seemed to depend on the waveform of the applied field. Reports from England have described a higher incidence of depressive symptoms in those residing near high-voltage lines, while another study from Sweden reported a higher frequency of neuromuscular complaints and "neurasthenic" symptoms among workers in the electric power industry.
In a series of studies, conducted in the baboon colony at the Southwest Foundation for Biomedical Research in San Antonio, Texas, a social group of male baboons was exposed to a 30-kV/m, 60-Hz electric field over three, six-week periods (12 hr/day, 7 days/wk). There was an increase in behavioral features categorized as passive affinity, tension and stereotypy (persistent repetition of senseless attitudes or movements), which are types of effects indicative of a stress response. Furthermore, electric field exposure disrupted the learned behavior patterns required to obtain food pellets for a period of 1-3 days (the authors referred to this as "work stoppage"), after which the learned behavior patterns were resumed. Surprisingly, when a 500-mG magnetic component was added to the electric field, the resultant combined EMF did not produce any of the changes seen with the electric component alone.
A number of scientists have reported that weak electric or electric and magnetic fields in combination can adversely affect performance on learned tasks or may increase the state of arousal of rats. This has been extended to possible long-term effects of EMF on behavior in studies in which rats were exposed to a combined 60-Hz, electric and a circularly polarized 60-Hz magnetic field from the day of conception to eight days after birth. Such rats, when conditioned to press a lever to obtain food in response to a light, responded at consistently lower rates than sham- exposed animals when tested as adults. The effect survived through 30% of the life span of the average rat.
Cell-Level Research
A wide range of changes has been observed in experiments involving exposure of isolated cells and tissues to EMF in vitro. Although many of the reports have described alterations in cell morphology, behavior, and biochemical properties which suggest that EMF can modulate the expression (that is transcription and translation) of various important control genes, including proto-oncogenes, there is no universal agreement on this among different laboratories. Experimental results have been inconsistent depending on the growth status of the cells, the type of EMF exposure used, the molecular biology methods and assays employed, and probably other as yet unidentified factors as well. Hopefully this situation will improve with the use of new exposure systems developed recently with the specific aim of standardizing exposures and minimizing or eliminating stray interferences. Should the improved assays confirm many of the earlier reports, it would appear that EMF acts to modulate normal signal mechanisms at the cell membrane and within the cell through, in some way, mimicking natural electrical activity in the cell. This could lead to functional changes, and perhaps other alterations resembling those induced by promoters of carcinogenesis. At present, no firm conclusions can be reached with regard to these questions.
EMF Reduction and Mitigation Research
Options for reducing exposures to EMF are being explored so that utilities and regulatory agencies may act quickly to implement mitigation techniques if scientific research indicates a need to limit public exposures to EMF. Researchers are exploring methods for reducing fields from sources such as transmission lines, distribution lines, consumer appliances, building wiring, and water pipes used as ground connections in many buildings.
There exist many planning and design techniques for reducing EMF exposures from transmission and distribution lines. Magnetic fields can be reduced by:
Researchers have also been investigating ways to mitigate fields from residential low-- voltage grounding systems. In many residences, multiple electrical ground connections are made to water pipes to reduce fire and shock hazards. However, these grounding practices create a path for current from household appliances and other customer loads to return through the ground connection to the water line. Investigations into the effectiveness, cost, and feasibility of options for reducing magnetic fields due to grounding practices are only now under way. Among the concepts for reducing ground currents are improving the neutral return path back to the transformer so that its impedance is less than the ground path, isolating the grounding of water, telephone, and cable TV systems, and inserting an imped- ance between the neutral and the ground at the residential electrical panel. Electric Power Research Institute (EPRI) is supporting several projects that will contribute to an understanding of the costs, safety implications, and degree of exposure reduction that may be achieved from these and other possible field reduction techniques.
Summary
Research on the potential health effects from exposure to power frequency EMF has expanded considerably over the past year and recently published studies have made important contributions to elucidating the nature of biological effects and to determining the possible implications of EMF exposure for human health. However, it is still not possible to arrive at definitive conclusions regarding the health effects from EMF exposure, based on the existing body of scientific evidence. There is no definitive indication that EMF exposure does or does not cause adverse health effects.
Epidemiology studies have examined the health of populations exposed residentially and occupationally to EMF. Several of these studies have reported an association between EMF exposure and the increased risk of certain rare types of cancer. One study found an association between childhood leukemia and historical records of electric current flow in power lines in the vicinity of homes, but not with short-term measures of EMF exposure. Other studies have not found a link between EMF and health effects, and even those that have demonstrated such a link are weak statistically because of the small numbers of cases. Overall, the epidemiological results are suggestive of a link but remain inconclusive.
Biological effects from EMF exposure have been observed in cells, tissues, whole animals, and humans. Observed biological effects include altered pineal gland function following exposure to ELF EMF or to static fields, co-promotion of cancer in certain cell types, selective enhancement and reduction in the production of some gene products associated with growth and development, changes in the development of embryos in some species but not others, temporary alterations in behavior in nonhuman primates and in rodents, inconsistent changes in the activity of various types of immune system cells, and altered human-heart activity following exposure. While these results continue to provide evidence of biological effects, the health implications of EMF exposure remain unclear. The existence of a number of biological effects from EMF exposure is of concern; however, it is premature to conclude that these biological effects imply adverse health effects for exposed humans. In addition, the biophysical or biological mechanisms that would explain how EMF interacts with living systems to produce these biological effects are not understood.
Despite the uncertainties in the biological and epidemiological evidence, researchers are continuing to assess human exposures to electric and magnetic fields and to investigate techniques for reducing exposures. There exist several techniques for reducing the amount or duration of time individuals are exposed to EMF, and the feasibility and effectiveness of additional techniques are being explored. In most cases, utilities have limited their mitigation measures to least expensive alternatives given the uncertainty of health risks and the benefits of reducing field levels. In most cases, this means giving consideration to EMF issues in the siting and design of new lines rather than rebuilding older lines. Ultimately, determining the need for implementing exposure reduction techniques will depend on the results of scientific research to determine which, if any, aspect of exposure is relevant to potential human health risks.