Methodological considerations on the estimation of non-initial radiation effects among Hiroshima atomic bomb...

A recent study by Ohtaki et al. (Frontiers in Public Health, 2025) proposed that non-initial (residual) radiation contributed substantially to solid cancer mortality among Hiroshima atomic bomb survivors, estimating residual radiation doses exceeding 2 Sv in areas west of the hypocenter. This commentary examines four methodological considerations relevant to interpreting these findings. First, the study infers residual radiation dose from the same excess relative risk (ERR) used to assess cancer risk-a form of circular reasoning that precludes independent validation of the dose-response relationship. Second, the attribution of geographically asymmetric cancer mortality to a specific exposure mechanism is difficult to substantiate without individual-level data on potential confounders; a simple sensitivity analysis suggests that modest non-zero ERR in the assumed control region could substantially reduce the reported excess. Third, the cohort entry criterion of confirmed Hiroshima residency in 1970, approximately 25 years post-bombing, may introduce survivor and migration bias. Fourth, sex-differential behavioral factors following the bombing may confound the interpretation of sex-specific ERR estimates. These considerations do not negate the importance of the research question but suggest that the magnitude of residual radiation effects proposed by Ohtaki et al. warrants careful evaluation. Independent dosimetry and individual-level data are needed to clarify the role of residual radiation in long-term health outcomes among atomic bomb survivors.