Unwarranted Fear About The Effects Of Radiation Leads To Bad Science Policy

Few issues have caused more fear and confusion than the question of the hazards of low-level radiation. There has been a remarkable failure to examine closely the evidence when discussing the issue and planning future studies. As a result, the public’s radiation phobia has been needlessly reinforced, and public money is being used on studies that are bound to be inconclusive. The problem arises, in part, because the general public—and even most scientists—are not aware that

Jun 13, 1988
Rosalyn Yalow
Few issues have caused more fear and confusion than the question of the hazards of low-level radiation. There has been a remarkable failure to examine closely the evidence when discussing the issue and planning future studies. As a result, the public’s radiation phobia has been needlessly reinforced, and public money is being used on studies that are bound to be inconclusive.

The problem arises, in part, because the general public—and even most scientists—are not aware that exposure to ionizing radiation is only weakly carcinogenic. The evidence comes from a number of different studies. Among the 82,000 survivors of the Hiroshima-Nagasaki bombing, for example, an average exposure of 27 rem only raised the number of cancer deaths by 6% over the number expected in the absence of such exposure. Radiation-induced leukemia among the survivors began to appear two to three years after exposure, reached a peak at five to six years, and declined thereafter.

Studies in experimental animals have shown that the duration of the exposure, not just overall dose, is important. Acute exposure generally results in a much higher incidence of tumors than does the same dose delivered over a period of hours to days. It is possible that the Chernobyl accident may provide a direct answer about dose-rate effects in humans. According to an August 1986 Soviet report, there were about 25,000 people living between three and 15 kilometers from the damaged reactor who accumulated a radiation dose of about 50 rem. Assuming that the Soviet scientists will report accurately on the medical follow-up of these people, we should learn within a year or two whether this group will show an increase in leukemia comparable to that observed in the Hiroshima-Nagasaki survivors. Because of its short latent period, leukemia is a better marker of radiation injury than is the incidence of solid tumors.

More evidence that low-level radiation is only a weak carcinogen comes from follow-up studies of patients receiving radionuclide therapy. By 1968, an estimated 200,000 people in the United States alone had been treated with radioactive iodine for hyperthyroidism, and the number has probably doubled since then. Yet a study of 36,000 hyperthyroid patients from 26 medical centers, of whom 22,000 were treated with 131 I and the remainder with surgery revealed no difference in the incidence of leukemia between the two groups. The overwhelming conclusion of these studies—and others that could be cited—is that doses under 10 rem, particularly when given at a low dose-rate, are not demonstrably carcinogenic.

All of this evidence, unfortunately, seems to have been widely ignored. There have been much-publicized reports claiming an increase in radiation-related malignancies at dose levels too low to account for such increases, based on the Hiroshima-Nagasaki experience. A report from the Centers for Disease Control in 1980, for example, claimed that a follow-up of about 3,000 participants in Operation Smoky, an atmospheric-nuclear test in Nevada, revealed 10 leukemia deaths when only 3.97 would have been expected, a standard mortality ratio (SMR) of 2.5. The considerable publicity associated with this report necessitated a very expensive follow-up of other participants in the nuclear weapons testing program.

Yet in another similarly sized group of people who had been exposed in Operation Greenhouse, a Pacific test, and whose radiation doses were about twofold larger, the results were different. Instead of 4.43 leukemia deaths, as would have been expected by the time of the follow-up, only one was observed (SMR = 0.23). Should one claim that the excess radiation was protective in this group? In fact, the differences from the expected leukemia rates in both groups were simply a consequence of the small sample number. When a larger cohort of 48,000 participants was studied, the actual leukemia rates closely matched the expected rates (SMR = 0.99). Low-level radiation exposure is not nearly as deadly as the general public tends to think it is.

Still, science policy-makers persist in reexamining the question. On February 6, 1988, The New York Times reported that the “National Institutes of Health have quietly initiated a study of cancer deaths among nuclear power plants; according to a letter released by Senator Edward Kennedy. This study is obviously designed in response to political pressure and cannot be justified on the basis of scientific merit.

There have already been a number of studies of cancer near nuclear installations. The latest is a British report, summarized last year in Nature (329:499, 1987). The summary includes tabulations of those malignancies that were more frequent near nuclear installations than in control areas (positive results), and those that were less frequent (negative results). The tables reveal that there are substantially more negative results than there are positive results. One table does suggest an increase in leukemia in areas near the nuclear plants. But more than half the regions where two-thirds of the population lived within six miles of a nuclear plant had fewer leukemia deaths than the regions where only one-third of the population lived within 10 miles of the plant. This would be consistent either with a hypothesis that low-level radiation is protective, or that all these studies are confounded by small sample sizes. This report, therefore, offers no compelling reason to restudy the problem. Exposures to residents around functioning nuclear installations are trivial compared to variability of natural background radiation throughout the United States.

Three years ago, in response to a mandate from Congress, radio-epidemiologic tables were published under the imprimatur of NIH. These tables presented the probabilities of radiogenic causation of a variety of malignancies according to the best scientific information then available. I assume that the NIH continues to agree with the data in this report, which would predict no measurable ef fects due to radiation levels in the vicinity of nuclear power plants. is regrettable that the NIH has yielded to political pressure by beginning a new study rather than affirming the scientific conclusions reached in its own report. By this action, a disservice is done, not only to science, but to the U.S. populace as well. It contributes to and reinforces the radiation phobia that interferes not only with our ability to employ the safest means for generating electricity but even with our ability to make optimal use of radiation in medicine.

Rosalyn S. Yalow is senior medical investigator at the Veterans Administration Medical Center in Bronx, N.Y She received a Nobel Prize in 1977 for the development of the radioimmunoassay technique.