Leukaemia | Radiation protection

Exploring the leukaemia link

8 December 2011

The latest report of the UK committee on medical aspects of radiation in the environment (COMARE) in May 2011 again finds no link between nuclear power plants and leukaemia in Great Britain; it also analyses results from the German KiKK study, which did find a link. An excerpt is reproduced here, together with responses to queries made since publication.

COMARE has reviewed the evidence from a variety of studies undertaken in a number of countries, using different methodologies, to determine whether there is an increased risk of childhood leukaemia in the vicinity of nuclear power plants (NPPs) as well as presenting a new geographical data analysis for Great Britain. It has also considered additional factors, which may have influenced the study results, including the status of cancer registries, the types of reactors used in the various countries, and the associated radionuclide discharges and doses to the general public from these discharges and other sources of exposure.

When considering the different methods of epidemiological analysis it should be recognised that each kind of study has strengths and weaknesses, which depend substantially on the details of the design and the area of application. As childhood leukaemia is a rare disease, the sample numbers in epidemiological studies are frequently small. For example, the KiKK case–control study in Germany included only 37 cases of leukaemia in children under 5 years of age, living within 5 km of an NPP over the 23 year period of the study [1, 2], and the new analysis for Great Britain presented in this report had 20 observed cases (under 5 years of age, living within 5 km of an NPP) over the 35 years of the study.

Previous geographical studies in Great Britain, including that described in COMARE’s 10th report, showed no significantly increased risks of childhood cancer, or in particular childhood leukaemia and non-Hodgkin lymphoma (NHL), within 25 km of an NPP, or any significantly increasing trend in incidence with proximity to an NPP. A further analysis of British data, specific to leukaemia and NHL incidence rates among children aged 0–4 years living within a 5 km radius of an NPP between 1969 and 2004, showed no significant increase in risk [3, 4]. Studies in other countries, such as in France and Finland, also reported no general increase in childhood leukaemia or childhood cancer incidence near NPPs.

For the new British geographical study presented in this report, the primary analysis shows no statistically significant evidence of an association between leukaemia risk and proximity to an NPP in Great Britain in children under 5 years of age. It is therefore possible to conclude that, in spite of its limitations, the geographical analysis of British data is suggestive of a risk estimate for childhood leukaemia associated with proximity to an NPP that is extremely small, if not actually zero.

A large number of analyses could be carried out in geographical studies like those reported in the report, varying the diagnostic groups included, the ages, the distance from the nearest plant, the choice of confounding variables and the statistical methodology. It would be highly misleading and unethical to carry out analyses and choose results that appeared to support a particular standpoint. COMARE is well aware of the danger of ‘cherry-picking’ or ‘boundary tightening’ (as the practice has been described). To avoid any question of doing so—even inadvertently —the working group charged with preparing the 14th report convened a special meeting. The characteristics of the analysis to be performed were systematically considered without any knowledge of how particular choices would affect the results. The member of the group involved in the analyses drew up the list of characteristics but gave no intimation of what might have been known from handling the data, and did not take part in any decision.


The KiKK case–control study in Germany [5] concluded that there is evidence of a raised risk of leukaemia in children under 5 years of age living within 5 km of an NPP during 1980–2003, but not for distances greater than 5 km. The results were confirmed by an independent analysis of the data [6]. Cases identified in earlier German investigations heavily influence the findings of the KiKK study (for the time periods 1980–1990 and 1991–1995). The more recent data (for the period 1996–2003) provide less evidence of an increased risk.

A marked excess incidence of childhood leukaemia has been found in the neighbourhood of the Krümmel NPP in northern Germany. The Krümmel cluster started in 1990 and continued until at least 2005, so its influence on the KiKK study results must be taken into account. For 1991–1995 and 1996–2003, the evidence for an increased risk of leukaemia in young children living within 5 km of German NPPs excluding Krümmel is only weak. The Northern Germany Leukaemia and Lymphoma (NLL) study [7] could not explain the Krümmel cluster in terms of routine radioactive discharges.

There is a disparity in the leukaemia risk levels for the period 1980–1990 assessed by geographical studies and the KiKK case–control study (see attachment). Possibilities for this difference include the distance measurements used and control selection in the KiKK study. Further investigation is required to understand this peculiarity – in particular because in the absence of the Krümmel data, the data for 1980–1990 are influential in generating the KiKK study findings.

Some criticisms of COMARE’s 14th report concern the methodology of the British study. It has been suggested that the same case-control methodology as was used in the KiKK study should have been used by COMARE. This shows a lack of understanding of the available data, for the KiKK study used population registers for the selection of controls, which are unavailable in the UK, there being no requirement (as there is in Germany) to declare officially the place of residence. On the other hand, case-control studies, while capable of capturing information at the individual level, are subject to potential biases between the cases and the controls. This is particularly true when the basis of data collection is different for the two groups. In the KiKK study, details of the cases were ascertained from records obtained from the cancer registry. In contrast, the controls were obtained by a complex method of sampling from population registers and it is known that inaccuracies exist in this selection process. These considerations necessitate caution in the interpretation of the KiKK case-control study, as well as the geographical studies.

Both British and German studies have considered the risk of leukaemia around potential nuclear sites (locations selected for an NPP but where construction was never undertaken) and found similar risk levels in some sites to those around active nuclear installations. This is suggestive of a risk associated with factors other than the operation of the plant, such as the nature of the location itself.

About leukaemia

Childhood leukaemia is not a single, homogeneous disease and both acute lymphoblastic leukaemia (ALL) and acute myeloid leukaemia (AML) have definable subtypes. Although it may not be possible currently to establish the cause of an individual child’s leukaemia, evidence suggests that factors other than radiation (including the existence of predisposing and spontaneously occurring genetic mutations and an abnormal immune response) are likely to be important in many cases.

Substantial evidence has been published in the past two decades indicating that patterns of infection are important in determining the risk of childhood leukaemia. The geographical distribution of cases throughout Great Britain found in the 11th COMARE report is consistent with this evidence [9]. It is plausible that unusual infectious processes of relevance to the risk of childhood leukaemia have occurred in the vicinities of some nuclear installations, increasing the risk there. However, the biological mechanism needs to be established before a definitive conclusion on the role of infection in the aetiology of childhood leukaemia can be drawn.

Childhood cancer is much less common than adult cancer. For epidemiological analyses investigating incidence rates for childhood cancers, there is a requirement for a comprehensive childhood cancer registration system that maintains a high level of ascertainment. Cancer registration systems vary between countries, operating on either a regional or national basis.

Considering the possibility that the radionuclide discharges from NPPs and the associated effective dose to members of the general public living in their vicinity may be a factor in the increased incidence of childhood leukaemia reported from some countries, COMARE investigated the levels of discharges and doses for the UK, France, Germany and Switzerland. Taking 1999 as an example year, the levels of discharges of specific radionuclides varied considerably between countries and between individual NPPs within each country. This variation may in part be due to discrepancies in reporting and differences in measuring practices. The requirement for reporting specific radionuclide discharges was not consistent across European countries in this selected year. The UK had the highest values of discharge levels for the radionuclides considered in this report. In each of the four countries, regulatory authorities assess the doses received by the public from the discharges of nuclear installations, and these are a small fraction of the overall radiation doses. In the UK, the measurement of radionuclides in food and the environment shows that doses to the general public from discharges from NPPs are well below the 1 mSv annual effective dose limit.

The report explicitly does not include Sellafield. The aim of the 14th COMARE report was to investigate nuclear power plants, and the Sellafield site was excluded from the analysis for a number of reasons [10]. The COMARE’s 10th report considered all nuclear installations individually, so that it is perfectly possible to identify the well-known excess of cases near the plant. The COMARE is committed to a review of the incidence of childhood cancers around both Sellafield and Dounreay, which is currently underway.

It should be noted also that a continuing nuclear power programme would be possible without a UK reprocessing facility.

This report has also considered the implied radiation doses to the general public determined from gaseous and liquid discharge data for the UK, France and Germany for 1999 [11]. These doses are assessed using the habits of the local population as well as incorporating meteorological and geographical factors and are therefore not solely related to distance from an NPP. Doses from liquid discharges also depend on the destination of the discharge. The calculations are reliant upon the availability of accurate discharge records, and the levels reported to the European Commission do not necessarily constitute the complete discharge inventory, due to variations in the reporting requirements. The implied doses for 1999 for all three EU countries were estimated to be substantially below the annual effective dose limit of 1 mSv for the general public and the average annual effective dose from natural sources of radiation. It is estimated that, in the UK, the annual effective dose from discharges from the nuclear industry accounts for around 0.0065% of the average annual effective dose from natural and medical sources of exposure. In Germany, the Commission on Radiological Protection concluded that radiation exposures to residents in the vicinity of German NPPs are lower, ‘by a factor of considerably more than 1000’, than the level that could cause the raised risk of childhood leukaemia reported from the KiKK study [12].

Extensive investigation of the uncertainties in the risk of childhood leukaemia arising from the radionuclides discharged from nuclear installations has taken place since the first report of the Seascale childhood leukaemia cluster in 1983. Much of this work has been carried out as a result of COMARE recommendations in earlier reports. No aspect of this uncertainty in the risk assessments has been found to be approaching a level that could account for the reported increases in childhood leukaemia incidence in terms of radiation exposure.

It has also been proposed that there is a substantial underestimation of the risk of childhood cancers from the intake of radionuclides and that discharges of tritium and carbon-14 may be responsible, in part, through in utero exposure of embryos and foetuses. Evidence presented to date does not support this suggestion.

COMARE appreciates that there are a number of issues associated with this review that require further research, including establishing the biological mechanism for childhood leukaemia and the biokinetics of, and tissue responses to, internal emitters. COMARE has also recognised that basic radiobiological research underlies and supports both radiation protection issues and translational radiobiology and appreciates the importance of sustaining research in this field.


This article consists of edited excerpts from the 14th COMARE report, “Further consideration of the incidence of childhood leukaemia around nuclear power plants in Great Britain”, and from the document “COMARE 14th Report: response to comments received

This article was originally published in the November 2011 issue of Nuclear Engineering International magazine



[1] Kaatsch P, Spix C, Schulze-Rath R, Schmiedel S and Blettner M (2008a). Leukaemia in young children living in the vicinity of German nuclear power plants. Int J Cancer 122, 721-726.

[2] Spix C, Schmiedel S, Kaatsch P, Schulze-Rath R and Blettner M (2008). "Case control study on childhood cancer in the vicinity of nuclear power plants in Germany 1980" 2003. Eur J Cancer 44, 275-284.

[3] Bithell J F, Keegan T J, Kroll M E, Murphy M F and Vincent T J (2008). Childhood leukaemia near British nuclear installations: methodological issues and recent results. Radiat Prot Dosim 132, 191-197.

[4] Bithell J F, Keegan T J, Kroll M E, Murphy M F G and Vincent T J (2010). Response to Letter to the Editor. Radiat Prot Dosim 138, 89-91.

[5] Kaatsch P, Spix C, Schmiedel S, Schulze-Rath R, Mergenthaler A and Blettner M (2007). Epidemiologische Studie zu Kinderkrebs in der Umgebung von Kernkraftwerken (KiKK-Studie) Abschlussbericht. Umweltforschungsplan des Bundesumweltministeriums (UFOPLAN) Reaktorsicherheit und Strahlenschutz. Bundesamt fur Strahlenschutz, Salzgitter. Available at http://www.bfs.de/de/bfs/druck/Ufoplan/4334_KIKK_Zusamm.pdf (accessed December 2010).

[6] SSK (2009). Bewertung der epidemiologischen Studie zu Kinderkrebs in der Umgebung von Kernkraftwerken (KiKK-Studie)"Epidemiological study of childhood cancer in the area of nuclear power plants (KiKK study)." Stellungnahme der Strahlenschutzkommission " Opinion of the Commission on Radiological Protection 58, Strahlenschutzkommission (SSK) des Bundesministeriums fur Umwelt, Naturschutz und Reaktorsicherheit Radiation Protection Commission (SSK) of the Federal Ministry for the Environment, Nature Conservation and Nuclear Safety.

[7] Hoffmann W, Terschueren C, Schill W, Pohlabeln H and Greiser E (2003). The Northern Germany Leukaemia and Lymphoma Study (NLL). Report (Part I): Radioactive Nuclides of Emissions from Nuclear Installations (Main Hypothesis I) [in German]. Institute for Community Medicine, Ernst-Moritz-Arndt University of Greifswald, on behalf of the Ministry of Environment, Nature Conservation and Agriculture of Schleswig-Holstein and the Ministry of Social Affairs, Women, Family Affairs and Health of Lower Saxony, Greifswald, Germany.

[8] Kaatsch P, Spix C, Jung I and Blettner M (2008b). Childhood leukemia in the vicinity of nuclear power plants in Germany. Dtsch Arztebl Int 105, 725-732.

[9] 14th COMARE report, "Further consideration of the incidence of childhood leukaemia around nuclear power plants in Great Britain" (http://www.comare.org.uk/press_releases/documents/COMARE14report.pdf), paragraphs 6.12-6.15.

[10] EC (2008). Implied Doses to the Population of the EU arising from Reported Discharges from EU Nuclear Power Stations and Reprocessing Sites in the Years 1997 to 2004. Radiation Protection 153. Available at http://ec.europa.eu/energy/nuclear/radiation_protection/doc/publication/153.pdf (accessed December 2010).

[11] SSK (2008). Assessment of the Epidemiological Study on Childhood Cancer in the Vicinity of Nuclear Power Plants (KiKK Study). Statement of the Commission on Radiological Protection (SSK). Available at http://www.ssk.de/en/werke/2008/volltext/ssk0806e.pdf (accessed December 2010).


Privacy Policy
We have updated our privacy policy. In the latest update it explains what cookies are and how we use them on our site. To learn more about cookies and their benefits, please view our privacy policy. Please be aware that parts of this site will not function correctly if you disable cookies. By continuing to use this site, you consent to our use of cookies in accordance with our privacy policy unless you have disabled them.