Abstract
Background: Late health effects of radiation exposure to the atomic bomb have been evaluated including the Life Span Study (LSS). Individual doses ranged widely from low- to high-dose level, which allows us to examine the effects at low-doses levels from many aspects.
Methods: Initial 120,321 people identified in 1950 were followed up until the end of 2003 using population registry system and death certificates. Among them, 86,611 subjects with known individual doses by the Dosimetry System 2002 (DS02) were analyzed for dose-response relationship for all solid cancer.
Results: Risk of all solid cancer was the best fit to the linear model in the whole dose range, but to the linear-quadratic model in the 0- to 2-Gy dose range. This finding suggested that the risk of radiation per unit exposure was lower at a low-dose level than at a high-dose level. However, this concave-upward curve was determined due to the relative lower estimates than the linear model at around 0.5 Gy level. The ERR/Gy was also higher, but not significantly, than the value that was estimated based on the whole dose range when the range was limited to 0.1 Gy or lower. The lowest range with significant excess relative risk of all solid cancer was around 0.2 Gy or larger.
Conclusions: Although no significant results were obtained at low-dose levels according to the RERF survivor cohort studies, including the latest study with the longest follow-up period, they are thought to support the linear no-threshold model. Analyses considering the confounding and interaction with radiation effects such as lifestyle, and medical radiation exposure are required to clarify the radiation effects at the low-dose levels.
