The aim of this work is to develop a critical analysis of quantitative risk assessment in the field of ionizing radiation and to provide new estimates of attributable risks for particular situations of environmental exposure to ionizing radiation. This work is based on knowledge about dose-response relationships and ionizing radiation exposure of the general population. The work focuses on two different situations that both present an important interest for public health: lung cancer associated with domestic radon exposures (natural situation) and thyroid cancer associated with the Chernobyl accident fallout (accidental situation).
The assessment of lung cancer risk associated with domestic radon exposure considers 10 dose-response relationships resulting from miner cohorts and case-control studies in the general population. A critical review of available data on smoking habits has been performed and allowed to consider the interactions between radon and tobacco. The exposure data come from measurements campaigns carried out since the beginning of the 1980's by the Institute for Radiation protection and Nuclear Safety and the Health General Directory in France. The French lung cancer mortality data are provided by the INSERM. Estimates of the number of attributable cancers are carried out for the whole country, stratified by 8 large regions (ZEAT) and by 96 departments for the year 1999 allowing to perform a sensibility analysis according to the geographical level of calculation. Uncertainties associed to risk coefficients and exposures have been quantified and it's impact on risk estimates is calculated. The estimated number of deaths attributable to domestic radon exposure ranges from 543 (90% uncertainty interval (UI): 75-1,097) to 3,108 (90% UI: 2,996-3,221). The corresponding risk fractions range from 2.2% (90% UI: 0.3%-4.4%) to 12.4% (90% UI: 11.9%-12.8%).
The assessment of thyroid cancer risk in the most exposed area of France due to the Chernobyl nuclear power plant accident fallout, is based on the excess relative risk model published by E. Ron et al. in 1995. The target population is composed of all children younger than 15 years and living in the eastern part of France at the time of the accident (26 April 1986). The incidence rates between 1978 and 1997 are analysed and forecasted up to 2007 using age-period-cohort models. Thyroid doses are estimated from all available data about the contamination in France from the Chernobyl fallout/ The risks characterization is carried out by considering various scenarios on the projection of spontaneous incidence rates, the effectiveness of internal exposures compared to exteral ones and the levels of radioactive deposits. Uncetainties to the risk coefficient, the expected spontaneous incidence rates and the thyroid dose are considered. The estimated number of thyroid cancer cases in excess between 1991 and 2007 for the target population ranges from 5 (90% UI: 1-15) to 63 (90% UI 12-180). In comparison, depending on the scenario of spontaneous rate projection, the number of spontaneous cancer cases should range from 894 (90% UI: 869-920) to 1,716 (90% UI:1,691-1,741) between 1991 and 2007. The results show that the thyroid cancer incident rate increase observed in France during last 20 years cannot be explained by the Chernobyl fallout.
The work provides an adaptation of the classical risk assessment method integrating each of its step such as a discussion about the choice of the dose-response relationship. The data analysis considers the interaction between ionizing radiation and other risk factors, a spontaneous incidence rate projection and a quantification of uncertainties. This work provides new results showing the importance of the choice of the dose-response relationship, of the consideration of the spontaneous incidence trend and of the quantification of uncertainties in risk assessment. In conclusion, this work provides new knowledge for public health by the analysis of enlarged epidemiological data.