Health effects of low-frequency electric and magnetic fields

Power substation magnetic field measurement using digital signal processing techniques ... Do Power Frequency Magnetic Fields Cause Leukemia in Childr...
0 downloads 0 Views 4MB Size
66 HEALTH EFFECTS OF LOW-FREQU ENCY ELECTRIC AND MAGNETIC FIELDS 33 BY DAVID A. SAVITZ

s in many evolving research areas, the only consensus among scientists regarding potential health effects from power-frequency electric and magnetic fields seems to be that the evidence is insufficient to draw firm conclusions. Thus, a meaningful evaluation of the state of knowledge requires a tolerance for uncertainty and a willingness to delve into the weight of evidence pushing opinion in one direction or another. Each new thread of positive evidence or theory must be seen as contributing some measure of support for the possibility of adverse health effects and each negative report or theoretical argument must be seen as weighing against that possibility. Unfortunately, many of those responsible for policy pertaining to electric and magnetic fields are impatient with the shades of gray that truly exist and seek the black or white they believe to be lurking in the present evidence. The inability to state an unequivocal conclusion about the health effects of electric and magnetic fields does not reflect a deficiency of the evaluator but is an accurate portrayal of the evidence. When the question is posed, Is there theoretical or empirical evidence that exposure to electric and magnetic fields at commonly encountered levels poses a threat to health?, the answer must be a firm yes. Epidemiologic evidence link-

A

52 Environ. Sci. Technol., Vol. 27, No. 1. 1993

ing power lines near residences and elevated magnetic fields to childhood cancer continues to accrue: employment in selected electrical occupations seems to confer an increased risk of leukemia, brain cancer, and perhaps breast cancer ( 1 ) : and there have been numerous laboratory studies indicative of influences on circ a d i a n rhythms ( Z ) , calcium efflux from nerve cells (31, a n d a hypothesized pathway linking such exposures to cancer ( 4 ) . It appears that the panel appointed by the Committee on Interagency Radiation Research and Policy Coordination (CIRRPC) was given a different charge. Regardless of what they were asked, the response appears to address the multifaceted question, Is there persuasive evidence that electric and magnetic fields are a major cause of clinically adverse health effects, with a biological understanding of the processes involved, a demonstration of quantitative dose-response gradients, and a firm basis for risk assessment? The answer they provided is the correct, obvious, and even obligatory one: “No.” A separate but related question is, Given the present state of knowledge and hypotheses concerning health effects of electric and magnetic fields, what priority should be assigned to further research on this issue? Most disappointingly, the panel comes to the conclusion that research on this topic is not a high priority. Presumably this was based on the inconsistencies and limitations of present theory and evi-

dence, but fails to incorporate the potential for future work in this area. The ubiquity of electric and magnetic fields surely deserves prominent consideration in assessing the importance of continued research relative to other more biologically DlaUSible but less urevalent agents. There are blatant gaps in the scientific literature on h e a l t h effects of these f i e l d s for which constructive avenues can readily be suggested. The examin a t i o n of magnetic fields in studies of rodent carcinogenicity is one example, with empirical evaluation of postulated biases from control selection in epidemiologic studies of childhood cancer another. Clearly, research o n this topic has not reached a dead end either by coming to firm conclusions or by exhausting productive approaches to challenging or corroborating the suggestions that have been put forth. Even for the most skeptical, research to assess where the positive epidemiologic and laboratory studies went astray would be of substantial value. Recently, Florig (5) undertook an analysis of the economic costs to society of uncertainty regarding health effects of electric and magnetic fields and found a sizable burden to a variety of decision makers who must act on the basis of implicit assumptions regarding such concerns. Even if public health motivation is insufficient, the need for public policies that explicitly or implicitly address these agents calls for sizable expenditures for research.

0013-936W93/0927-52$04.00/00 1992 American Chemical Sociev

c

All-or-none approach

In its impatience to find an irrefutable body of research to demonstrate and explain health hazards from electric and magnetic fields, the panel appears to accept nothing less than instant, comprehensive evidence. In this area, as in most others, isolated and tenuous bits of information that may or may not ultimately coalesce into a full picture appear from different disciplines. If

we ask, Is it a fully formed picture?, the answer is negative, but the failure of the whole does not negate the individual parts. The report consistently sidesteps discussion of the evidence that does exist by focusing on the aspects that are missing. Confronted with a series of epidemiologic studies linking childhood cancer to wiring configuration codes, rather than an evenhanded evaluation of whether that association is valid or spurious

or what factors might account for the association, the report dismisses the evidence because cancer rates have not risen in parallel with electric power consumption. A discussion of the research suggesting that calcium metabolism or subtle aspects of behavior may be influenced by electric or magnetic fields is distracted by questions of its clinical significance and ultimate public health importance. Virtually any research supportive of a Envimn. Sci. Technol.. Vol. 27, No. 1, 1993

53

causal relation between exposure and health is dismissed based on the “signal-to-noise” problem (which has been challenged by biophysicists [SI).the lack of support from other disciplines, and indirectly by the admittedly irrational extent of public anxiety. Presumably the high level of public attention to this issue frequently leads the authors to interrupt reviews of scientific evidence with the conclusion that there is “no cause for concern.” A more informative approach might objectively examine each thread of evidence, consider how firm it is, and determine what research would be likely to support or refute it. A separate discussion of the coherence of the body of evidence would then be valuable. and yet another section might address the community psychology and public policy aspects. Secular trends in exposure and disease A specific theme that is invoked for both cancer and adverse reproductive outcomes is that whatever the research might suggest, electric or magnetic fields could not adversely affect health because the rise in electric power use over time has not produced epidemics of cancer, birth defects, or miscarriage. In fact, the two observations are true but unrelated. In order to produce parallel trends in exposure and disease, as is readily observed for smoking and lung cancer, for example, the secular trend data would require: (1) markedly rising exposures, (2) accurate data on the rate of the disease over time, (3) an absence of major concomitant changes in other risk factors for the disease, and ( 4 ) strong association between exposure and disease. In regard to electric and magnetic fields. all are absent. The most critical problem is that the rising use of electric power has probably not been accompanied by any increase in exposure. In moving from a preindustrial to industrial society, exposures surely increased, but that was over a time period that is not amenable to study given the absence of accurate health records. In the past 40 years it seems unlikely that exposures have risen. One hit of empirical evidence comes from Rhode Island (71.where homes occupied in the 1960s tended to have higher exposures than those occupied in the 1970s (81, presumably 54 Environ. Sci. Technol.. Vol. 27. No. 1, 1993

reflecting the movement from central cities to the suburbs, with larger yards and greater distances from homes to power lines. This demog r a p h i c s h i f t was w i d e s p r e a d throughout the United States. In-home wiring practices also changed over that period from a pattern that generated large magnetic fields by spatial separation of the phases to a method in which the phases are closer in space and largely cancel one another. Increasing the voltage of power lines over time leads to lower currents, compensating in part for the greater use ofelectricity. Ultimately, we cannot assess how the population exposure has changed from the 1950s to the 199Os,but there is good reason to believe that it does not parallel the striking increase in electricity consumption during that interval. The report invokes lack of secular trends in childhood cancer to counter results of case-control studies, as though the calendar time in which children lived were a superior exposure marker to characteristics of individual children’s residences. The health events of interest, cancer and reproductive outcomes, are not available over long enough periods to address the pre-electrical era for which a contrast in exposure can be made. Only during the past 20 years or so have reliable registries for cancer incidence and birth defects become available. with mortality data for cancer markedly influenced by improvements in survival (particularly for childhood cancers) during that period. Thus, the era in which reliable health data are available does not coincide with the period in which exposures were likely to have changed dramatically. Finally, although temporal trends in exposure and disease can be informative, the marked changes in other relevant factors over the same period might spuriously create or mask an association with the agent of interest. Changes in dietary habits, control of infectious diseases, and improvements in medical technology for diagnosing and classifying disease would likely obliterate any change related to an exposure that has a moderate or small relation to disease. Thus, distinguishing between the presence of a moderate effect of electric and magnetic fields on health and no effect cannot possibly be done on the basis of secular trends. One might speculate on how the reviewers would (and should) react to an assertion that the rise in brain cancer in the elderly in the

past 20 years (9, 10) is attributable to increased use of electric power. Conclusions In spite of the numerous evaluations of the literature that have been conducted, expert panels such as the one established by CIRRPC offer the potential for important insights. The most constructive criticism of existing research is that which leads to testable, refutable assertions. The burden should be placed on the critics to postulate errors in the positive reports and to suggest studies that would either demonstrate that such errors occurred or enhance the previous report by demonstrating that it was free from such errors. It is unfortunate that more effort and enthusiasm were not devoted to channeling the panel’s wisdom and apparent skepticism into suggestions for productive research that would ultimately reveal the truth regarding health effects of electric and magnetic fields. David A . Sovitz i s a n epidemiologist a t the University of North Carolina School of Puhlic Health. He received his Ph.D. in epidemiology from t h e University of Pitfsburgh in 1982. His resenrcti interests me focused on environmeniol influences on reproductive health a n d cnncer.

References (1) Savilz. D.A.: Ahlbom, A. In Biologic

Effects of Electric and Magnetic Fields: Carpenler, D. 0.. Ed.: Academic Press: New York, in press. (21 Groh, K . R.: Readey. M. A , : Ehret. C.F. In Extremely Low Frequency Electromagnetic Fields: The Question of Cancer: Wilson, B. W.: Stevens. R. G.: Anderson. L. E., Eds.: Battelle Press: Columbus. OH, 1990. (31 Blackman, C. F. In Extremely Low Frequency Electromognetic Fields: The Question of Cancer: Wilson. B. W.: Stevens. R. G.: Anderson. L. E., Eds.: Battelle Press: Columbus. OH, 1990. (41 Stevens. R. G. el

al. FASEE]. 1992. 6,

853-60. (51

Florig. H. K. Science 1992. 257, 46892.

(SI Weaver, I. C.: Astumian. R. D.Science 1990.247.459-62. (71

Fulton.

I. P. et al.

Am. I. Epidemiol.

1980. 111.292-96. (81 Wertheimer, N.: Leeper. E. Am. 1. Epidrrniol. 1980. 3.461-62. (91 Polednak. A.’P.’ 1. Notl. Cancer Inst. 1991.83.1679-82. (101 Desrneules, M.: Mikkelsen. T.: Mao. Y.I. Notl. Cancer Inst. 1992. 84.44245.