Dose-criterion based verification of radiation risks of death in the cohort of Russian participants of clean-up operations following the Chernobyl accident

As more data on health effects of radiation are accumulated, the more groups of diseases are investigated with the use of radiation epidemiology approaches and quantitative indicators. Radiation associated risk of the same disease estimated by different investigators may vary considerably. For this reason it has become necessary to develop simple criteria that could allow identification of sufficient conditions for radiation risk existence and its assessment with the use of available follow-up data. The purpose of the study was to develop the sufficiency criterion to identify radiation associated risk existence with data of the cohort study within frames of the linear (on radiation dose) multiplicative risk model. With the use of radiation risk estimates, personal dosimetry and health history data of the Chernobyl liquidators included in the cohort study and followed up at the National Radiation Epidemiological Registry (NRER) for the period from 1986 over 2019, we identified the diseases that caused death statistically associated with radiation dose. The developed criterion of exceedance of average doses related to death cases over averaged by the cohort doses with the follow-up time weights is the sufficient condition of radiation risk of death existence within the linear multiplicative risk model. The use of developed threshold criterion allowed formulating two new hypotheses on models for radiation risks of death. Hypothesis 1: For radiation risk of death from solid cancers. It is the first demonstration of the existence of two time-dependent phases of stochastic effect manifestation: 24 years after exposure to radiation. Further testing of the significance of dependence of risk from time is necessary. Hypothesis 2: It is the first time demonstrated the existence of radiation risk of death from the diseases of the digestive system, as well as the existence of the latency period for the risk manifestation, the length of the period is about 9 years. Quantitative estimate of radiation risk and conclusion on its statistical significance will be obtained in further studies. As a result, the developed threshold criterion of sufficiency for existence of radiation risk estimated within the linear multiplicative model with the use of the cohort study data has demonstrated its usefulness in the study of cohort of the NRER Chernobyl liquidators. Additional researches of probabilistic distribution of dose difference between cases of death and whole cohort are required for development of the corresponding statistics when checking a zero hypothesis of lack of radiation risks.