I HAVE neighbors who worry because they don't have healthy cockroaches in their home; the only one they've seen in 15 years was on its back, kicking. They suspect the previous owner contaminated the house with a powerful pesticide, so they sleep with the windows wide open, even in mid-winter. Why not? they shrug; it also dissipates the radon seeping up from the basement.
A year ago, watching Dan Rather on the evening news, they learned of a report prepared by the Environmental Protection Agency that warned that magnetic fields of 60 hertz are a "probable, but not proven," cause of cancer in humans. These fields emanate from the ordinary electric wires that supply power to American homes. My neighbors moved their daughters' bed to the middle of the room so it wouldn't be near any wall outlets.
A few weeks later, the report, "Evaluation of the Potential Carcinogenicity of Electromagnetic Fields," was revised to downgrade the hazard from a "probable" to a "possible" cause of cancer. That would seem to reduce 60-hertz magnetic fields to the same risk category as artichokes or shoe polish. Some saw the downgrading as evidence of a cover-up; others objected that even the revised report arouses concerns that are not justified by the evidence.
D. Allan Bromley, the White House science adviser, intervened, insisting that experts from the EPA review the report.
The review panel, composed of medical researchers, epidemiologists, engineers and physicists, met in marathon public sessions and listened to wide-ranging views. The panel found the EPA report had "serious deficiencies" and recommended that it be completely rewritten and then re-reviewed.
The review panel stopped short of exonerating 60-hertz magnetic fields; it wasn't persuaded by arguments that biological effects from these fields are impossible.
But among the deficiencies in the report, the panel found the statistical connection between power-line fields and cancer to be shaky and the laboratory evidence of biological effects to be unclear and incomplete. The panel was particularly critical of the failure to find a "dose-response relationship"; that is, more does not seem to be worse when it comes to magnetic fields.
Confirming a dose-response relationship is generally required in classifying something as a hazard. Most scientists regard the failure to prove this as a classic symptom of bad science.
The panel's cautious downgrading of the evidence linking electromagnetic fields and cancer is not unusual; in recent months, the threat from dioxin has been downgraded from cataclysmic to slight or even nonexistent.
Cyclamates, banned from diet soda 21 years ago, have been given a clean bill of health by new studies.
We are now told that more health problems may have been created by efforts to clean up asbestos than would have resulted had it been left alone.
An international team of nuclear experts is even suggesting that the Soviets greatly over-estimated the long-term health effects of Chernobyl.
Research on weak effects always involves uncertainties and creates a troubling ethical dilemma for scientists.
Should suspicions of an environmental hazard be announced at once to allow people to adopt a strategy of "prudent avoidance," or should any announcement be delayed until all the scientific checks and reviews are completed? In fact, there is little choice. Some of the coverage will be excessively sensational, but the media will not and should not wait for scientists to give the all-clear. What should be given greater emphasis is the scientist's obligation to try to put the risk in proper perspective for the public.
My neighbors should know, for example, that the risk from electromagnetic fields, if any, is very slight. In the 50 years from 1930 to 1980 the per-capita consumption of electric power in the U.S. increased tenfold -- yet the incidence of childhood leukemia, the disease that was first associated with power-line fields, showed no change.
They should also know that the National Cancer Institute has undertaken a four-year study of the problem. Before they switch to candles, it would be a good idea to wait until the real risks are known.
Robert L. Park is a professor of physics at the University of Maryland College Park.