Letter to the Editor

(Comment on the Letter to the Editor by Sergei V. Jargin concerning “Atomic bomb testing and its effects on global male to female ratios at birth” by Victor Grech, The International Journal of Risk and Safety in Medicine 27 (2015), 35–44. DOI: 10.3233/JRS-150641)

Sir,

In response to findings by Victor Grech of biased global human gender proportions at birth following environmental radiation exposure [1–4], Sergei V. Jargin concludes that ‘a significant role of radiation from nuclear testing and Chernobyl fallout as a factor modifying the male to female ratio at birth is improbable’. Jargin’s reasoning mainly builds on two arguments: (1) ‘social factors that could have influenced this ratio were not comprehensively analyzed’ and (2) ‘the natural radiation background was not mentioned, although additional doses due to the contamination were often negligible compared to doses from the background’.

A social factor that may skew the birth sex ratio is gender selective abortion, a practice reported from parts of Asia and parts of North Africa [5]. However, this method requires advanced techniques for prenatal gender ascertainment that were not available at the time of the Windscale fire in 1957 and during the era of the major atmospheric nuclear weapons testing prior to 1963. Moreover, in western countries prenatal sex-specific selection is not part of the culture where females are valued as much as males. Therefore, the jumps in the secular sex ratio trends after Windscale in Europe [6] and the nearly synchronous jumps after the partial nuclear test ban treaty of 1963 [7] in Western Europe and in the USA can hardly be explained by sex selective abortion. Figure 1 shows an updated version of our synoptic sex ratio trend analysis for Europe and the USA [8] now based on 520 million officially recorded births from 1970 to 2010. In epidemiology, temporality and biological gradient are two indispensable requisites for inferring causality. Temporality means that the presumable cause precedes the observed effect in time, and biological gradient refers to the presence of an exposure-response association. Thus, Fig. 1 yields strong evidence of causality in as much as a clear sex ratio change occurs immediately after Chernobyl in the exposed parts of Europe only, and not in the USA, which remained essentially unaffected by Chernobyl fallout. In other words, Fig. 1 represents a highly significant spatiotemporal dose-response association based on precise and extremely powerful human data. A similar conclusion further highlighting the relative importance of internal versus external radiation exposure can be derived from the synoptic consideration of Russian and Cuban sex ratio trends [9]. In view of these facts, the tacit assumption by Jargin that soft and only gradually changing social factors could have caused abrupt sex ratio changes is implausible, all the more so since we found a significant dose-response association of Chernobyl fallout with subsequent sex ratio increases at the district level in Germany [10]. This is consistent with recent findings form a high background radiation area in India [11, 12] and with a report from a nuclear testing area in Kazakhstan [9, 13].

It is generally accepted that proof of safety is more difficult than proof of hazard [14]. Suggestively, this has been rephrased that ‘no evidence of effect is not evidence of no effect’. Therefore, the negative animal low dose experiment cited by Jargin [15], which is based on only 1053 baby mice distributed over 18 generations (i.e. 58.5 mice per generation) is practically meaningless due to its negligible statistical power. Another reportedly negative observation referred to by Jargin concerns the sex ratio shifts among infants born to survivors of the atomic bombings of Hiroshima and Nagasaki [16]. However, scrutiny of these data reveals clear and significant sex ratio jumps after the bombings, not only in Hiroshima and Nagasaki but also in all of Japan [17].

The dose (Gray or Sievert) in the radiation sciences is a surprisingly old and crude concept developed long before the discovery of the molecular structure of the DNA. The basis of this concept is unit energy deposited per unit mass (Joule/kilogram). Therefore, it is clear that dose can only be physically meaningful for energy deposits in volumes of at least a few milliliters, e.g. of water, and not for the interaction of photons, electrons, and α-particles by external or internal emitters with cellular and sub-cellular structures like cells, DNA, and proteins in animate beings. Therefore, comparing the additional radiation exposure of human populations after radiological incidents and accidents with the natural background radiation level is misleading in principle as the specific microbiological consequences of artificial external and internal ionizing irradiation and radionuclides are not taken into account. Jargin’s view that the natural annual background radiation dose of 1 mSv/a is trivial and that even multiples of this dose after nuclear accidents were ‘often negligible compared to doses from the background’ points to a key dissent in science: The doubling of the background radiation level, say, from 1 mSv/a to 2 mSv/a, represents a doubling of an important physical environmental parameter relevant for the development of life on earth, and cannot as such be considered a ‘low dose’ and of ‘no effect’. It is hard to imagine any ‘environmental parameter’ a doubling of which would have no consequences, be it the exposure to the sun, the oxygen and the carbon dioxide contents in the air, or the individual cigarette and alcohol consumption per year. Moreover, Jargin seems not to be aware of or not to acknowledge more recent findings by comprehensive and statistically powerful studies of increased childhood cancers in the vicinity of nuclear power plants [18, 19], in areas of high background radiation [20], and following computerized tomography examinations [21, 22].

In conclusion, the reservation by Sergei V. Jargin towards results of observational epidemiological investigations is certainly warranted in principle in so far as single studies cannot prove causality. However, the findings presented by Victor Grech must not be considered in isolation, they are rather consistent with early observations after the atomic bombing of Japan, and are clearly in line with our results after Chernobyl and in the vicinity of nuclear facilities [8, 9, 17, 23, 24]. The relief of ionizing radiation as a possible culprit for the sex ratio increases after radiological events offered by Jargin: social factors and negligibility of doses is not convincing as gradually changing social factors cannot entail abrupt sex ratio changes and a doubling of the background radiation is certainly not trivial. Considering the findings of the last two decades in radiation epidemiology synoptically, evidence continues to show that radiation risks have been systematically underestimated in the past, due to inadequate scientific concepts and underpowered or poorly designed experimental and human studies.