Noncancer mortality among the Japanese atomic bomb survivors

Abstract

Yukiko Shimizu and her colleagues from the Radiation Effects Research Foundation in Hiroshima have recently published the results of the updated analysis of deaths from causes other than cancer among the Japanese survivors of the atomic bomb explosions, covering the period 1950-1990 (Shimizu Y, Pierce D A, Preston D L and Mabuchi K 1999 Studies of the mortality of atomic bomb survivors. Report 12, Part II. Noncancer mortality: 1950-1990 Radiat. Res. 152 374-389). The primary analyses are based on 27 000 deaths, 30% more than in the previous report on this subject. These latest findings confirm a statistically significant trend of an increasing rate of noncancer mortality with increasing dose, due to trends for diseases of circulatory, digestive and respiratory systems. At 1 Sv, the proportional increase is about 10%, much smaller than for cancer (at around 50%), but the numbers of excess deaths are more comparable. The all-important question of the dose-response relationship remains unresolved: linearity is possible, but the data are also consistent with a threshold at 0.5 Sv. The authors conclude that misclassification, confounding or selection effects are unlikely to fully explain the findings. Enhanced knowledge of the biological mechanisms underlying these epidemiological observations would enable the nature of the dose-response to be better understood. Also reported is a statistically significant trend of an increasing rate of noncancer blood diseases with dose, which cannot be accounted for by misclassification. This paper also confirms that suicide rates tend to decrease with increasing dose. Clearly, the impact of these latest results for noncancer mortality upon risk estimates for low dose exposures is going to be much discussed by the radiological protection community.