The approximate date of this letter is December, 1997.
“In my youth, I too, entertained some illusions; but I soon recovered from them. The great orators who rule the assemblies by the brilliancy of their eloquence are in general men of the most mediocre political talents: They should not be opposed in their own way; for they have always more noisy words at command than you. Their eloquence should be opposed by a serious and logical argument; their strength lies in vagueness; they should be brought back to the reality of facts; practical arguments destroy them. In the council, there were men possessed of much more eloquence that I was: I always defeated them by this simple argument – two and two make four.”
—Napoleon, in “History of the Captivity of Napoleon at St. Helena”, 1847.
I asked Helen Caldicott to respond to the several critical letters about her article in the Los Angeles Times. Like the adversaries of Napoleon, she is eloquent, but thrives on vagueness. Her claims often don’t survive a careful comparison with the facts and she seems to want to avoid a detailed discussion of the disputed points. Rather than defend the original specific points in her article that were disputed by her critics, she chose to attempt to shift attention to other topics, such as their organizational affiliation and her views on nuclear weapons. Important topics, no doubt, but I would have more respect for her tactics, if she would either admit the errors in her original article or offer a rational defense and explain why her critics are mistaken.
Her response is basically an appeal to emotions and does little to refute the specific points mentioned by her critics. The scientists who wrote critiques of her article listed several specific points to which they objected including such claims as “A Friends of the Earth study showed that a nuclear plant must operate for 18 years before realizing one net calorie of energy.” Her critics also point out that the number of cancers produced by the wastes from coal fired power plants is “thousands of times larger than the number caused by nuclear power wastes” for the same amount of power produced. Caldicott does not respond to the criticisms which suggests that she may want to avoid discussion of her mistakes. One would think that the relative amount of waste spread over the countryside by various types of power plants would be a subject that deserves some debate, so I am disappointed that Caldicott has chosen to avoid a detailed discussion of that topic, even when she is challenged on the accuracy of her original statements.
In her response, Helen Caldicott listed several references that she claims support her view that low doses of radiation are “6 to 8 times more dangerous that originally estimated”. Let’s consider her reference number 3, on the epidemiological study of Rocketdyne workers. As usual with the scientific claims of Helen Caldicott, the study does not provide strong evidence for her position. Actually the study showed that Rocketdyne radiation workers had a lower incidence of death from “all causes” and also a lower incidence of death from cancer, than the U.S. population and also in comparison to other worker groups. Let’s consider the actual percentages. Compared to the U.S. population, Rocketdyne radiation workers had a 32% lower death rate from “all causes” and a 21% lower death rate from “all cancers”. Compared to a similar worker control group, Rocketdyne radiation workers had a 38% lower death rate from “all causes”, and a 11% lower death rate from “all cancers”. The number of radiation workers from which these percentages are derived is over 4,000 which allows a high degree of confidence that the differences are real and not due to chance.
If Rocketdyne radiation workers have such better mortality statistics, are we justified in concluding that radiation is in fact beneficial? Although there are reputable nuclear scientists who believe that radiation can be beneficial in some cases, we must be careful in drawing conclusions based on these studies. The statistical evaluation of data from epidemiological studies is full of pitfalls. For example, radiation workers are not random samples of the entire population nor are they even a random sample of the population of skilled workers. Radiation workers may well be better educated, have higher incomes and differ in many important ways from the population at large and from other worker groups. Many of these factors can influence the health and mortality of selected groups. This is often called the “Healthy Worker Effect”. What is proven beyond doubt by this and other studies is that the ill effects of radiation exposure, if any, are small in comparison to other risk factors. That in itself is an important result, one that is not often mentioned by the anti-nuclear activists.
Why then does Helen Caldicott cite this study as proof of the dangers of working with radiation? As with any large body of data, the more one divides it up and looks at small parts of the overall picture, the more likely it is that some small segment of the data will show some deviation from the expected outcome. However, as the numbers become smaller and smaller, the more likely it is that a deviation from the expected result will be due merely to chance and not to any real effect. This is somewhat like going to Las Vegas and playing for a few minutes at some game of chance.
There is a large probability of winning if you play only for a few minutes. This probability is greatly reduced, almost to zero, if you play the game all week. The mildly unusual results in the Rocketdyne study involved less than 1% of the radiation workers, not enough to really draw sound conclusions. In fact I think it is fair to say that if you divide any data set into 100 parts, at least one of those parts are likely to be anomalous, purely due to random statistical fluctuations.
It is important to note that the study of Rocketdyne workers was conducted by UCLA and according to critics of the study, it was directed by a group of activists who have a long history of being critical of the nuclear industry. The study concluded that there was an increased rate of lung cancer in those workers with an external exposure above 20 rem. This was based on 2 lung cancer deaths from 34 workers with exposure above 20 rem. The authors of the study apparently made no effort to determine if these two people were smokers, which is of course a well known cause of lung cancer.
The study also concluded that there was an increased rate of leukemia/ lymphoma in those workers with external exposure above 20 rem. This was based on 1 leukemia death and 1 “Hodgkins Disease” death from 34 workers with exposure above 20 rem. However in both the case of lung cancer and for leukemia/lymphoma, the numbers are so small that there is little statistical confidence that the apparent elevation in rates is real. It is somewhat like trying to predict the outcome of a presidential election based on a poll of a few people in your living room.
The extraction of important conclusions from a very small set of numbers is a common characteristic of epidemiological studies by the anti-nuclear groups. The most important conclusion from the Rocketdyne study and others like it, is that common personal habits and cultural factors are far more important in determining risk of premature death than is exposure to ionizing radiation. The anti-nuclear activists would save more lives, thousands of times more lives, if they attacked smoking, abuse of alcohol, and careless driving habits, rather than the use of radioactive materials.