Scherb H1, Voigt K.
1Institute of Biomathematics and Biometry, Helmholtz Zentrum Muenchen, German Research Center for Environmental Health, Ingolstaedter Landstrasse 1, Neuherberg, 85764, Germany.
Abstract
BACKGROUND, AIM, AND SCOPE:
Ever since the discovery of the mutagenic properties of ionizing radiation, the possibility of birth sex odds shifts in exposed human populations was considered in the scientific community. Positive evidence, however weak, was obtained after the atomic bombing of Japan. We previously investigated trends in the sex odds before and after the Chernobyl Nuclear Power Plant accident. In a pilot study, combined data from the Czech Republic, Denmark, Finland, Germany, Hungary, Norway, Poland, and Sweden between 1982 and 1992 showed a downward trend in the sex odds and a significant jump in 1987, the year immediately after Chernobyl. Moreover, a significant positive association of the sex odds between 1986 and 1991 with Chernobyl fallout at the district level in Germany was observed. Both of these findings, temporality (effect after exposure) and dose response association, yield evidence of causality. The primary aim of this study was to investigate longer time periods (1950-2007) in all of Europe and in the USA with emphasis on the global atmospheric atomic bomb test fallout and on the Chernobyl accident. To obtain further evidence, we also analyze sex odds data near nuclear facilities in Germany and Switzerland.
DATA AND STATISTICAL METHODS: National gender-specific annual live births data for 39 European countries from 1975 to 2007 were compiled using the pertinent internet data bases provided by the World Health Organization, United Nations, Council of Europe, and EUROSTAT. For a synoptic re-analysis of the period 1950 to 1990, published data from the USA and from a predominantly western and less Chernobyl-exposed part of Europe were studied additionally. To assess spatial, temporal, as well as spatial-temporal trends in the sex odds and to investigate possible changes in those trends after the atomic bomb tests, after Chernobyl, and in the vicinity of nuclear facilities, we applied ordinary linear logistic regression. Region-specific and eventually changing spatial-temporal trends were analyzed using dummy variables coding for continents, countries, districts, municipalities, time periods, and appropriate spatial-temporal interactions.
