Childhood leukemias associated with fallout from nuclear testing.

Continuing concern over the possible carcinogenic effects of low-level radiation prompted us to study the population of Utah because of its exposure to fallout from 26 nuclear tests between 1951 and 1958. Certain rural counties (high-fallout counties) received most of the fallout during that period. We reviewed all deaths from childhood (under 15 years of age) cancers occurring in the entire state between 1944 and 1975 and assigned them to a cohort of either high or low exposure, depending on whether 15 between 1951 and 1958. For reasons unknown, leukemia mortality among the low-exposure cohort in the high-fallout counties was about half that of the United States and the remainder of the state. Mortality increased by 2.44 times (95 per cent confidence, 1.18 to 5.02) to just slightly above that of the United States in the high-exposure cohort residing in the high-fallout counties, and was greatest in 10- to 14-year-old children. For other childhood cancers, no consistent pattern was found in relation to fallout exposure. The increase in leukemia deaths could be due to fallout or to some other unexplained factor.