Consequences of a large-scale nuclear accident and guidelines for evacuation: a cost-effectiveness analysis.

PURPOSE: We aimed for a quantitative evaluation that justifies guidelines for evacuation which take into consideration both the human and economic costs. To the best of our knowledge, such an evaluation has not been performed yet. The present guidelines published by the International Atomic Energy Agency (IAEA) are probably based on averting radiation risk only; IAEA did not cite any quantitative estimation of the human cost of evacuation. MATERIALS AND METHODS: Quantitative estimation of the human and monetary costs of evacuation and, alternatively, the human and monetary costs of radiation exposure (non-evacuation). Associating human life with monetary value is psychologically difficult and somewhat challenging ethically; however, there is no escape from such an association (cost-effectiveness analysis) when making decisions regarding public health and safety, since extraneous public expenditures lead to a statistical life shortening. Estimating worst-case health consequences of irradiation, we used the conservative linear no-threshold (LNT) model because this model is widely used in spite of its controversy. In our estimation of the human cost of evacuation, we considered three factors: (a) direct loss of life (after Fukushima, 1% of the evacuees died within 2 years due to causes directly related to their evacuation), (b) loss of quality of life, and (c) loss of wealth leading to loss of life. The connection of economic loss with loss of life was performed according to the median cost-effectiveness threshold of 50-100 thousand USD per quality-adjusted life year. RESULTS: Even according to mortality calculations based on LNT, the overall loss of life due to evacuation is higher than the loss of life due to irradiation if the population-averaged first-year radiation dose is 500 mSv or less. CONCLUSIONS: Based on the performed analysis, we suggest avoiding evacuation if the projected first-year dose is below 500 mSv. This suggested action level is about five-fold higher than the action level presently recommended by the IAEA (100 mSv per year).