US radiation study sparks debate

Nature : According to the US Nuclear Regulatory Commission, most of a person’s background-radiation exposure comes from natural sources and from some medical exams, with less than 1% coming from nearby nuclear reactors. The agency has asked the National Academy of Sciences to examine the potential cancer risk involved in living near a nuclear power plant, and the NAS is consulting experts on the best way to go about this. The feasibility and usefulness of the study are already a matter of debate. The last US-wide study, published by the National Cancer Institute in 1990, found no evidence of significantly increased risk. However, that study grouped people by county regardless of their actual distance from a nuclear plant, and it only considered cancer deaths. Two decades later, GPS devices can now pinpoint where people live in relation to a reactor and researchers can now look for patterns in cancer diagnosis, not just cancer deaths. Nonetheless, various potential problems with the new study have been cited: It can’t control for the presence of confounding factors, as laboratory research can; low-dose effects of radiation might not be picked up; epidemiological studies measure correlation and do not prove causation. Steve Wing, an epidemiologist at the University of North Carolina at Chapel Hill, says that if there is an effect, it will be easiest to see in children and fetuses, as their rapidly dividing cells make them more sensitive to radiation. Wing and his colleages wrote an article on how best to design the NAS study in the 1 April issue of Environmental Health Perspectives.