| Statement by the external BfS panel of experts on the KiKK study |
|
Following an internal discussion of the final report of the KiKK study on December 9 and 10, 2007, the external panel of experts arrives at the following conclusions:A. Analyses within the scope of the study1) Main result of the study is that for children under the age of 5 the risk to contract cancer and leukaemia continuously increases the nearer they live to a nuclear power plant. 2) The study-design complies with the state-of-the-art of epidemiological science. 3) The methodology of modelling the continuous distance variables is adequate. Both models applied in the study show good adaptation to the collected data. The models permit an assessment of the incidence risks in dependence of the distance of the home to the nearest nuclear power plant site. 4) The major part of the external panel of experts´ suggestions regarding analyses of the quality of data and results was realised. These analyses did not give any indications that the results were significantly distorted. 5) The authors rightly stated that the risk to contract childhood cancer and leukaemia significantly and continuously increases with increasing vicinity of the home to a nuclear power plant. The study is the methodically most elaborate and comprehensive investigation of this interrelation world-wide. The interrelation between vicinity of the home and incidence risk has thus been sufficiently proved for Germany. 6) Currently it is not possible to finally evaluate the quality of data collection, processing and analysis, since the required information was not available to the external panel of experts. It would have been possible to clarify these questions through the audit proposed by the external panel of experts. The panel of experts recommends to further pursue the issue of the audit. B. Further analyses carried out by the contractors7) The evaluation plan did not provide for the calculations on the attributive risk. Stating the risk that can be attributed to the distance of the home to a reactor and the population-related risk is indispensable for the communication of the results to politicians and the general public. 8) In the case under consideration the calculations were not carried out correctly. - Only the 0-5-km area around the nuclear power plant sites was taken into account, while the remaining parts of the area to be investigated were disregarded, although significantly increased risks were calculated there, too. - The reference population for the calculation of all cancer and leukaemia cases in children under the age of 5 was not determined correctly. In consequence, the number of cancer incidences that can actually be attributed to the vicinity of the home to nuclear power plant sites was underestimated. - Instead of additional 29 cancer incidences in children under the age of 5 for the 0-5-km area alone stated by the authors, at least 121-275 additional incidences in the area within a radius of 50 km around all West German nuclear power plant sites must be assumed for the period from 1980 to 2003. This corresponds to 8 – 18 % of all cancer incidences in children under the age of 5 having occurred within a radius of 50 km around nuclear power facilities. - Relating to all incidences in the same period registered in the German Childhood Cancer Registry, this corresponds to 1,03 – 2.35 %. This figure has to be seen as an underestimation, since due to the design of the study not all affected children could be registered. This risk is considerably above the 0.22 % reported by the authors. 9) The authors write that „… due to the current radio-biological and radio-epidemiological knowledge the ionising radiation emitted by German nuclear power plants in normal operation can basically not be interpreted as the cause.“ In contrast to the authors, all members of the external panel of experts are convinced that due to the particularly high radiation risk for small children and due to the insufficient data on emissions of power reactors, this interrelation can on no account be excluded. Furthermore, there are several epidemiological causality criteria in favour of such an interrelation. The task of science now is to find an explanatory approach for the difference between epidemiological and radio-biological evidence. 10) To explain the risk around nuclear power plants proved by them, the authors refer to so-called confounders, selection mechanisms not described in detail, or statistical chance. In view of the study results, the external panel considers all three explanatory approaches to be improbable. Frankfurt/Main, 10 December 2007 For the panel of experts: E. Greiser K-H. Jöckel W. Hoffmann More information
|