Members of the public and scientists have been concerned about environmental contamination from nuclear weapons and nuclear power generation for a long time. The National Academy of Sciences is currently working on a request from the Nuclear Regulatory Commission to design an epidemiologic study of cancer around nuclear facilities in the USA.
People living near nuclear facilities may see an epidemiologic study as a way to shed light on their health concerns. An epidemiologic study could do that. However, if epidemiologic studies are not well-designed, they can be used to dismiss the public’s concerns and avoid implementation of public health protections.
There are many perils of epidemiologic studies, especially ones focused on low-level exposures. It’s easier to detect the effect of larger exposures, for example of nuclear workers, than the effects of smaller exposures, for example of people living near nuclear facilities. Furthermore, radiation exposures of most nuclear workers are monitored, whereas exposures of residents are not. This presents a big challenge, because an epidemiologic study that cannot sort people correctly into exposed and unexposed groups cannot detect an effect of exposure.
Several epidemiologic studies in Europe have found excess childhood leukemia among children living near nuclear power plants. These studies compared children living close to nuclear plants – within 5 km (3 miles) – to children living further away. No similar studies have been conducted in the USA, in part because we don’t have a national medical program that counts cancer cases, and in part because most of our health data are only reported for large geographic areas like counties.
The National Academy of Sciences study could be designed to improve on the European studies. However, the Nuclear Regulatory Commission has asked for a study that includes adults, who are less sensitive to radiation exposure than children. Furthermore, adult cancers may appear decades after exposure, increasing the opportunity for people to move between exposed and unexposed areas. Studies of adults, of large areas like counties, and of cancer death instead of cancer diagnosis, would not advance scientific knowledge about health effects of living near nuclear facilities, but such a study could become grounds for dismissing concerns about radiation releases. Another problem is that epidemiologic studies may be conducted under the assumption that radiation exposure is too low to affect cancer. Then, if an excess is found among people living near nuclear facilities, scientists must attribute it to some other unknown cause. This circular logic – evidence of the effect is dismissed because it is already believed there can be no effect – is unscientific but is dressed in the trappings of science to make it appear reasonable.
Members of the public concerned about radiation exposures from nuclear facilities should critically consider any proposed study to decide whether to give it their trust and support.
– Steve Wing, University of North Carolina
For further reading: Wing S, Richardson DB, Hoffmann W. Cancer risks near nuclear facilities: The importance of research design and explicit study hypotheses.Environmental Health Perspectives, 119:417-21, 2011.