High doses of ionizing radiation are an established cause of childhood leukemia. However, substantial uncertainty remains about the effect of low doses of radiation, including background radiation and potential differences between genetic subgroups of leukemia have rarely been explored. We investigated the effect of the background gamma radiation on childhood leukemia using a nationwide register-based case-control study. For each of the 1,093 cases, three age- and gender matched controls were selected (N = 3,279). Conditional logistic regression analyses were adjusted for confounding by Down syndrome, birth weight (large for gestational age), and maternal smoking. Complete residential histories and previously collected survey data of the background gamma radiation in Finland were used to assess the exposure of the study subjects to indoor and outdoor gamma radiation. Overall, background gamma radiation showed a non-significant association with the OR of childhood leukemia (OR 1.01, 95% CI 0.97, 1.05 for 10 nSv/h increase in average equivalent dose rate to red bone marrow). In subgroup analyses, age group 2-<7 years displayed a larger effect (OR 1.27, 95% CI 1.01, 1.60 for 1 mSv increase in equivalent cumulative dose to red bone marrow). Suggestive difference in OR by genetic subtype was found. Our results provide further support to the notion that low doses of ionizing radiation increase the risk for childhood leukemia, particularly at age 2-<7 years. Our findings suggest a larger effect of radiation on leukemia with high hyperpdiploidy than other subgroups, but this result requires further confirmation.

Our results do not indicate that higher residential mobility or moving to municipalities with more inhabitants is associated with risk of childhood leukemia.