News

Childhood Leukemia Clusters Around 16 German Nuclear Power Plants

Rudi H. Nussbaum, Portland State University, Portland, OR
Long-term tabooing of the truth can sometimes give it away inadvertently.
Background
Several years after the dropping of "atomic bombs" on Hiroshima and Nagasaki, formerly secret reports surfaced about continuing long-term health effects from radiation exposure, raising concerns about workers' and the public's exposures from nuclear facilities. This led in the 1950s to the life-time follow-up of Hiroshima-Nagasaki survivors who had been exposed over a large range of doses by the bomb's flash, depending on their distance to ground zero. It is significant that the "Life Span Study" cohort (LSS) was assembled by the US Atomic Bomb Casualty Commission (ABCC) 5 years after the bombing. Members of the Atomic Energy Commission (AEC) and others in the radiation health establishment expected long-term effects to be negligible. The highly secretive ABCC later transformed into the Radiation Effects Research Foundation (RERF which included Japanese scientists). Despite the optimistic expectations, the radiogenic cancer mortality remained elevated among the surviving population at lower and lower exposure levels over years of subsequent follow-up periods due to wide individual variations in latency periods. At the start of the cold war, the advocates for a new nuclear arms race had to overcome the public's continuing trauma from the August 1945 use of atomic bombs on non-military targets. Consequently, the military/nuclear industry establishment, using President Eisenhower as spokesperson, started a well-subsidized large-scale "Atoms for Peace" promotion. Over the decades it has promised safe, cheap, and by now even "green" salvation from the looming energy crisis by nuclear power generation. Significantly, however, all the promotional publicity to date has not been able to convince the insurance industry to underwrite it.
The burgeoning radiation health establishment provided well-paid careers for scientists and health professionals, including appointments to prestigious national and international boards and commissions. These same experts were called upon to review grant applications and manuscripts submitted for publication to medical/scientific journals. Their opinions determined which studies would be published and which would not. Moreover, most radiation related research has relied on funding by agencies that had as their mission the promotion, facilitation, and regulation of the uses of ionizing radiation. Consequently, many government/industry-sponsored radiation health studies were specifically designed to fend off litigation for workers' or citizens' compensation claims.
The A-bomb survivor study, as interpreted by various authoritative bodies such as the National Academy of Sciences (BEIR), the International Commission on Radiation Protection (ICRP), the United Nations' UNSCEAR committee, or the World Health Organization (WHO), has been the primary basis for evaluations of radiation risks. The internationally adopted radiation protection standards for nuclear workers had to be such that they did not unduly burden the profitability of the industry. Hence, any occupational or environmental health study suggesting radiogenic detriment at exposures well below defined "safe" levels presented a direct challenge to these official interpretations. Such studies and their authors have always met fierce denunciation in the literature - if they succeeded to get published at all [Nussbaum 2007]. Yet, such challenges have persisted from health studies among nuclear workers [Wing 1991; Kneale 1995], from studies of malignancies in populations around the ill-fated Three Mile Island nuclear reactor [Wing 1997], or around some British [Gardner 1990] and French [Pobel 1997] nuclear installations, as well as from studies of the long-term health legacy at large distances from the Chernobyl explosion [Baverstock 2006;. Nussbaum 2006]. There is documented evidence that the estimates of population doses from Chernobyl fallout in Western Europe by UNSCEAR, as well as the officially adopted low-dose radiogenic risk factors, extrapolated from the A-bomb survivor study are probably too low by as much as a factor 10 or more [Nussbaum 1994, 1998]]. Yet, most of these peer-reviewed "discrepant" health studies in and around nuclear plants, or those among populations exposed to radioactive fallout from nuclear tests or reactor accidents, have persistently been ignored in authoritative periodic reviews of radiogenic risks by BEIR, ICRP, UNSCEAR or WHO. In the context of disputed findings of excess cancers around the Sellafield, UK nuclear installation, a 1991 editorial in the American Journal of Industrial Medicine states: ". . . [the radiation experts] concluded, on theoretical grounds, that these [doses] could not have caused any major excess risk: ‘it could not have happened, so it didn't happen'" [Greenberg 1991].
As a consequence of this mindset, several effective critics of the "mainstream scientific consensus" found themselves cut off from research support, professionally ostracized, and their careers stunted [Nussbaum.2007].
Independent scientists published evidence that the Japanese LSS survivor cohort constitutes a highly selected study population [Wing 1999; Stewart 2000]: Only persons with exceptionally strong health could have survived the first five years of a totally devastated social structure in the two destroyed cities. Thus, the most radiation-sensitive very youngest and oldest survivors had been eliminated from the LSS cohort, skewing RERF's conclusions about dependence of radiogenic risk on age at exposure. This explains at least in part the wide disparities between LSS-based age-dependent risk factors and well-documented findings of strongly heightened susceptibility for radiation detriment among both the young and the old in ordinary populations, including a particularly high sensitivity during early fetal development [Gilman 1988; Lord 1992]. In addition, RERF reports on dose-dependent radiogenic risk at low doses are also flawed by large uncertainties in the doses assigned to LSS survivors stemming from external exposures due to residual fallout in Hiroshima, never included in RERF's dose classifications. At distances between 1.5 - 1.7 km from the hypo-center the additional dose from fallout exceeded the doses from direct exposures to the bomb flash [Sawada 2007]. In addition, RERF neglected internal exposures from this fallout [Sawada 2007]. Nonetheless, ignoring all these important flaws in dosimetry, the standards for radiation protection, including those for young children and the elderly, continues to be based on the follow-up of that exceptionally healthy LSS cohort and the flawed dose classifications. No wonder, then, that e.g. nuclear worker studies, based on readings of personal dosimeters, found more than 10 times higher radiogenic risks at low doses than those adopted by official agencies [Kneale 1995; Wing 1991].
Mainstream reviews of radiogenic risk have also ignored large uncertainties in the concept of "relative biological effectiveness" of radiation (RBE), in particular on the molecular or cellular level. RBE varies over factors of 10 with type (alpha, beta or gamma) and energy of the radiation, as well as, most crucially, on whether exposure originates from external sources (like X-rays, or radiation flashes from A-bombs) or from radioactive emitters lodged in body tissue.
The German childhood malignancy study around nuclear reactors
After a childhood leukemia cluster was reported in the late 1980s in the vicinity of the nuclear plant Krümmel near Hamburg [Schmitz-Feuerhake 1997; Hoffmann 2007], and continued to be discussed in German media, public anxieties were raised.
In earlier so-called ecological studies disease rates were found to be elevated in regions around reactors, compared with those in regions without reactors. However, such studies, working with population averages (aggregates), can only suggest, but not confirm or deny a causal relationship between living in the vicinity of nuclear power plants and higher density of childhood cancer cases.
Therefore, in 2002 the German federal childhood cancer registry was contracted to initiate a case-control study of childhood cancer cases in the areas around all 16 German nuclear reactors, known in Germany under the acronym KiKK. A case-control study seeks to find common characteristics that distinguish individual children who suffer from a disease (cases) from those who do not have this disease, but share the same demographics (controls). Since radioactive emissions from the reactors had been suspected to be associated with the malignancies, and since direct measurements of individual exposure are not available, distance to the reactor was chosen as a plausible surrogate variable. Thus, residential addresses of the cases at the time of diagnosis were compared with addresses of controls. This type of study can confirm or contradict the working hypothesis, that incidences of childhood malignancy do not increase with proximity to nuclear power plants. The credibility of the KiKK study was enhanced by the appointment of a prestigious independent review committee of 12 scientific experts: 5 epidemiologists, 2 pediatricians, 2 statisticians and 3 physicists.
Addresses of all German children under the age of 5 at the time at which they had been diagnosed with malignancies during the period 1980 - 2003 (1592 cases) were compared with addresses of 3 times as many randomly selected children of the same age and sex, and residing in the same region of the country, who did not have this disease (4735 controls). Care was taken to match cases and controls in all demographic factors that are known or suspected to influence disease rates, leaving residential distance to the 16 power plants as the only significant variable. To my knowledge, no comparably authoritative health study has ever been conducted among residential populations around nuclear installations in Germany, the US or elsewhere [Kaatsch 2008; Spix 2008].
Based on the presumed very low emission data as provided by the reactor operators, combined with the internationally adopted low-dose risk factors for radiogenic cancers, the investigators expected that a state-of-the-art study design would allow them to conclude that radioactive emissions from these plants are not causally associated with the childhood cancer clusters. This could have silenced once-and-for-all the opposition, based on public health concerns, to the planned construction of new nuclear reactors.

To the surprise of the KiKK investigators and of their government sponsors, the study found a continuous increase in childhood leukemia and cancer incidence within a radius of 50 km from the 16 German reactors, the closer the children had lived to a nuclear plant at the time of diagnosis. This translates into an increase in relative risk from 1.02 at 50 km to 1.27 at 5 km, with 1.01 the lowest value for the lower 95% confidence range. A relative risk of 1.27 means that children living 5 km or less from a nuclear plant are 27% more likely to develop a malignancy than children living outside the 50 km radius. Such a consistent correlation around all 16 reactors is one of a set of indicator for a causal relationship [Hill 1965]. Consistency with independent health studies near nuclear installations is another:.
* A meta-analysis of incidence and mortality rates of childhood leukemia near 136 sites worldwide. It showed a statistically significant increase near many nuclear facilities, although the authors [Baker 2007] do not suggest an explanation for these excesses.
* Other investigators of cancer clusters near nuclear installations, such as in Germany [Schmitz-Feuerhake 1997; Hoffmann 2007], in France [Pobel 1997], in the United Kingdom [Gardner 1990] and in the US [Mangano 2002, 2008], suggest that the most likely reason for these clusters is exposure to radioactive emissions, citing extensive supporting evidence in their references.

In contrast, the KiKK investigators concluded that, based on current radio-biological knowledge and epidemiological studies of health effects of radiation, the emitted radioactivity from normally operating German nuclear power reactors can "on principle" not have caused the observed excesses in leukemias and other cancers in children. They claim that radioactive releases from these nuclear plants would have to be at least 1,000 times higher to explain the observed effects. Therefore, since possible confounders of association of malignancies with proximity to the power plants could not be identified, they declared that their findings remain unexplained [Kaatsch 2008; Spix 2008]

The not-to-be-questioned "principles" the KiKK scientists invoked are the industry estimates of emissions, as well as the appropriateness of the applied radiogenic risks They ignored (1) the high likelihood of the cases' internal exposure to radioisotopes by ingestion and (2) the manifold higher radiosensitiviy of the developing fetus [Gilman 1988; Lord 1992]. The radiation risk models used in the KiKK team's conclusion ignore both of these factors.

This counter-intuitive interpretation of the study's findings by its investigators is in line with repeated claims by the WHO and the IAEA that well-documented findings of excess cancers, neo-natal mortalities, spontaneous abortions and other health detriment all over Western Europe, at large distances from the 1986 Chernobyl disaster, could not possibly be associated with radioactive fallout. Population exposures, as estimated by UNSCEAR, combined with the official radiation risk factors for such doses, would be several factors of ten too small to initiate the observed and reported detrimental health effects [Baverstock 2006; Nussbaum 2006].

The independent scientific panel, reviewing the KiKK study, came to a more plausible final conclusion which can be paraphrased as follows:

Epidemiological causality criteria as applied to the data of the KiKK study do suggest a causal relationship between the emissions from nuclear power plants and the shown increases of childhood cancer cases with decreasing distance from these plants. There exists no plausible alternative hypothesis. [Greiser 2008].
To summarize:
This exclusion of "common sense" (plausibility) from judgment in deference to the authority of a powerful government/industry/radiation science establishment exemplifies how a carefully manipulated mainstream canon can suppress healthy scientific skepticism.
Yet, in the end, the study, meant to dispel public fears of detrimental health effects from emissions around nuclear power plants, actually corroberates them.
That's probably why so few people in this part of the world have heard anything about the KiKK study.

References
Baker PJ et al. 2007. Meta-analysis of standardized incidence and mortality rates of childhood leukemia in proximity to nuclear facilities. Eur J Cancer Care 16:355-363.
Baverstock K et al. 2006a. The Chernobyl accident 20 years on: an assessment of the health consequences and the international response. Environ Health Perspect 114:1312-1317.
Baverstock K et al. 2006b. The Chernobyl nuclear catastrophe: Baverstock and Williams respond. Environ Health Perspect 115:A239
Gardner MJ el al. 1990. Results of case-control study of leukaemia and lymphoma among young people near Sellafield nuclear plant in West Cumbria. BMJ 300:423-429.
Gilman EA et al. 1988. Pregnancy X-rays and childhood cancers: Effects of exposure age and radiation dose. J Radiol Prot (UK):3-8.
Greenberg M. The evolution of attitudes to the human hazards of ionizing radiation and to its investigators. Am J Ind Med 1991;20:717-721
Greiser E et al. 2008. Epidemiological assessment of the KiKK studies by the German registry for childhood cancers. Presentation by the external experts commission to the Federal Agency for Radiation Safety (BfS), April 17, 2008 [in German]. Accessible at http://www.strahlentelex.de/KiKK_Bewertung%20_Greiser.pdf
Hill, AB. 1965. The environment and disease: Association or causation? Proc of Roy Soc Med 58, p. 295-300.
Hoffmann W et al. 2007. Childhood leukemia in the vicinity of the Geesthacht nuclear establishments near Hamburg, Germany. Environ Health Perspect 115:947-952.
Kaatsch P et al. 2008. Leukaemia in young children living in the vicinity of German nuclear power plants. Int J Cancer 1220:721-726.
Kneale et al. 1995. Factors affecting recognition of cancer risks of nuclear workers. Occup Environ Med 52:515-523.
Lord BI et al.. 1992. Age-dependent uptake and retention of 239 Pu: Its relationship to haemopoietic damage. Rad Prot Dosim 41:163-167.
Mangano J et al. 2002. Infant death and childhood cancer reductions after nuclear plant closings in the United States. Arch Environ Health 57:23-30.
Mangano J et al. 2008. Chilhood leukemia near nuclear installations. Eur J Cancer Care 17:416-418.
Nussbaum RH et al. 1994. Inconsistencies and open questions regarding low-dose health effects of ionizing radiation. Environ Health Perspect 102:656-667.
Nussbaum RH. 1998. The linear no-threshold dose-effect relation: Is it relevant to radiation protection regulation? Med Phys 25:291-299.
Nussbaum, RH. 2006. The Chernobyl Nuclear Catastrophe: Unacknowledged health detriment. Environ Health Perspect 115:A238-A239.
Nussbaum, RH. 2007. Manipulating public health research:EThe nuclear and radiation health establishments. Int J Occup Environ Health 13: 328-330.
Pobel D et al. 1997. Case-control study of leukemia among young people near La Hague nuclear reprocessing plant: the environmental hypothesis revisited. BMJ 314:101-106.
Sawada S. 2007. Cover-up of the effects of internal exposure by residual radiation from the atomic bombing of Hiroshima and Nagasaki. Med Conflict Survival 23(1):58-74
Schmitz-Feuerhake I et al. 1997. Leukemia in the proximity of a boiling-water nuclear reactor: Evidence of population exposure by chromosome studies and environmental radioactivity. Environ Health Perspect 105(suppl 6):1499-1504.
Spix C et al. 2008. Case-control study on childhood cancer in the vicinity of nuclear power plants in Germany 1980-2003. Eur J Cancer 44:275-284.
Stewart AM et al. 2000. A-bomb survivors: factors that may lead to a re-assessment of the radiation hazard. Int J Epid 29:708-714.
Wing S et al. 1991. Mortality among workers at Oak Ridge National Laboratory. JAMA 265:1397-1402 and extensive subsequent Correspondence.
Wing S et al. 1997. A reevaluation of cancer incidence near the Three Mile Island nuclear plant: The collision of evidence and assumptions. Environ Health Perspect 105:52-57 and correspondence.
Wing S et al. 1999. The relevance of occupational. epidemiology to radiation protection standards. New Solutions 9(2):133-151.