The Magical Malignancy Multiplying Machine

Radiation and Human Health, by John Gofman, San Francisco: Sierra Club Books, 1981, 928 pp., $29.95.

John Gofman is the "brain trust" of the antinuclear movement, but can his brain be trusted? Radiation and Human Health, his "magnum opus," is presented as the definitive book in the field, but errors of logic and interpretation appear in almost every section. Gofman claims that virtually every other expert in the field of medical effects of low-dose radiation is wrong and that only his interpretation and extrapolation of the basic data can be correct.

For example, Gofman "guarantees" (his expression) one cancer death for every 268 person-rads of radiation exposure to the general public, while the most widely accepted study of this question roughly estimates that there might be one fatal cancer for every 2,000–5,000 person-rads. (The person-rad is a measure of radiation absorbed dose, equivalent to one rad given to one person, or 1/1000 rad given to each 1,000 people, etc.) Thus, Gofman predicts 10 to 20 times more fatalities from a given dose of radiation than does the National Research Council in its widely accepted 1980 BEIR Committee report.

He has claimed in media presentations that diagnostic x-rays are causing 90,000 deaths per year (nearly one-quarter of all cancer deaths), while the standard estimate is about 4,000. Note that an estimate of lives saved by diagnostic radiology would be greater than either of these numbers.

His alarming predictions of carcinogenic and genetic disaster from the small amounts of radiation exposure due to nuclear power have galvanized many uninformed people into blind fear of nuclear power generation. In a recent survey, college students ranked nuclear power as the most important health risk in the United States, whereas in fact the best actuarial estimate ranks it only 20th in importance.

Admittedly, this is a difficult field. Since experts do disagree, how are Gofman's assertions to be assessed? First, one must look carefully at his assumptions, his use of data, and the internal consistency of his arguments. Second, his predictions should be compared with facts. I have done both and consider his conclusions to be worthless.

Item: Gofman's comment on the 45 deaths observed in the study of an area of China with high natural background radiation is that "it is obvious that the number of cancer deaths is much too small to provide statistically significant results." But he himself uses less than 10 excess cancer cases found from the zero-to-nine-year age group at Hiroshima and Nagasaki as the basis for his entire edifice of calculations (the magical malignancy multiplying machine), without even a nod to the question of statistical significance or the possibility that there might be, in the lives of children who were bomb victims, other factors (stress, depression, loss of parents, chemical carcinogens, etc.) that increased the incidence of cancers.

Item: Gofman adopts a "relative risk" model that assumes the number of radiation-caused cancers is derived from the product (that is, multiplication) of the radiation dose and the spontaneous cancer rate. He uses "magic" to go from the observation that many different cancers have been found, with various frequencies, in association with moderate radiation doses, to the grand hypothesis that all types of cancer are induced, at the same rate of increase per rad dose, by low-dose irradiation, regardless of dose rate.

He ridicules the alternate model, claiming that "the absolute risk method has no theory at all." In fact, the absolute risk method is equivalent to a perfectly reasonable theory—that the number of radiation-induced cancers is independent of the spontaneous cancer rate but does depend on the dose. This theory has just as much evidence as the first and is gradually becoming more accepted. It gives results that are not too dissimilar to those of the relative risk model when calculations are done by more responsible hands than Gofman's.

The radiation-health community, including myself, has learned much from Dr. Gofman in the past. He showed us that the combination of regulatory and promotional functions of the Atomic Energy Commission in the early '70s represented conflict of interest, resulting in its being split into the Energy Research and Development Administration and the Nuclear Regulatory Commission. He reinforced by endless repetition (and now exaggeration) the idea that fetuses and children are more susceptible to radiation effects than adults are. He helped turn our attention to the apparent violation of individual rights entailed in release of radioactive material (or any other pollutant) into the environment, where it can do potential harm to humans whose permission was not given.

Unfortunately, there is no easy solution to this last problem. For the problem is not one of nuclear power or even of industrial civilization: every time a farmer turns a furrow, more naturally occurring radon escapes from the earth to the atmosphere, increasing minutely the risk of lung cancer for everyone. Quite simply, "There ain't no such thing as a risk-free life."

Gofman's new book, however, is worse than worthless. It is potentially dangerous to the health of the public, disastrous to the nuclear defense of the United States, and could make world war more likely. If enough people were frightened by its (false) prophecy of damage due to medical x-rays, they would obtain needed tests less often, resulting in increased illness and death due to undiagnosed and untreated disease. If enough people were deluded into thinking, like Jonathan Schell, that The Fate of the Earth is the extinction of all life in case of nuclear war, they could be convinced (fooled) into advocating complete unilateral nuclear disarmament. If enough people around the world accepted its (false) message that nuclear power presents an unacceptable health risk, there would be more dependence on oil, with increased likelihood of war over Middle East petroleum supplies.

Luckily, the book itself will probably not have much impact. It is so grandiose and mock-encyclopedic that less than 10 percent of those who buy it will read it. It is so ill-conceived and badly written that less than 10 percent of those who read it will understand it. If reason prevails, less than 10 percent of those who understand it will believe it. If common sense prevails, the "magical malignancy multiplying machine," like any other pile of radioactive junk, will just decay away.

Howard Maccabee is both a physicist and a medical doctor who currently works as a radiation oncologist.