Vital statistics and psycho-social stressors
The Japanese Statistics Bureau publishes demographical information compiled by the Ministry of Health, Labor, and Welfare. Statistics include the annual numbers of live births and the annual counts of children with a low birth weight of < 2500 g (LBW), see Table 1 or the internet platform Vital Statistics of Japan. We investigated the spatiotemporal distribution of 26.158 million live births, of which 2.366 million births (9.04%) with weights < 2500 g, across 47 Japanese prefectures from 1995 to 2018. A considerable proportion of the Japanese population was physically and psychosocially affected to a significant degree by the Great East Japan Earthquake and subsequent tsunami [50]. Therefore, the counts of earthquake related deaths, the counts of the dead and missing after earthquake and tsunami, as well as the numbers of evacuees to and within any prefecture were obtained from official sources [51] and served as additional explicit ecological confounding variables in our logistic regression models. Since medical supply including medical information may also impact the general health-behavior and thus the prevalence of LBW, annual physician density by prefecture in Japan was deployed as a surrogate confounder variable in the spatiotemporal logistic regression models, see Table 2.
Cs-137 deposition
Yasunari et al. published average prefecture-specific Cs-137 deposition after the Fukushima nuclear power plant accidents for the 47 prefectures of Japan [48]; see Table 3. Yasunari et al. assumed that half of the total cesium deposited was Cs-134. Therefore, it is easily possible to re-scale all calculations in this paper to total cesium in place of Cs-137. The Yasunari et al. data understate the true Cs-137 deposition, which underestimation may be supported by the following aspects:
- Yasunari et al.'s data based on measurements restricted to March 20th to April 19th, 2011.
- Yasunari et al. report a value of 24.7 kBq/m2 Cs-137 for Fukushima prefecture (see Table 3), whereas the UNSCEAR data set 2013/2014 documents a mean value of 153.957 kBq/m2 Cs-137, which amounts to a factor 6 underestimation of the deposition in Fukushima by Yasunari et al. [48], see the Excel-file provided by UNSCEAR as referenced in Table 3 [49].
- It is implausible that Fukushima prefecture would be less contaminated than Miyagi and Ibaraki prefectures, see Table 3.
- The Yasunari et al. data decay with r-3.27 at distance r form the FDNPP, see Figure 1, whereas a theoretical decay law of r-1.42 is expected according to UNSCEAR [52], and as empirically confirmed for the Fukushima prefecture [40].
Since strong underestimation of radiation exposure would exaggerate any dose-specific radiation risk estimates, we suggest and propagate a correction and a rescaling of the Yasunari et al. deposition data for all of Japan despite the disadvantages of these data listed above. The rationale behind this is that the Yasunari et al. data, while restricted to a narrow time frame nevertheless reflect a valid mutual relative exposure status amongst the prefectures. To this end, we firstly increased the original deposition value 24,718.4 Bq/m2 of the Fukushima prefecture by a factor of 4.3 to 106,867.0 Bq/m2 based on the MEXT/UNSCEAR data [49] excluding 20 locations in the immediate vicinity of FDNPP with more than 2.0E+6 Bq/m2 with less likely importance for public exposure. Secondly, we rescaled the deposition data with the original decay-rate r-3.27 to a decay of r‑2.00, which is a compromise between the theoretical decay r-1.42 by UNSCEAR [52] and the Yasunari decay r-3.27. The rescaling details and results are depicted in Figure 1 and listed in Table 3. Figure 2 shows a geographic region value plot for the rescaled Cs-137 deposition in the Japanese prefectures.
Dose-rate (µSv/h) derived from Cs-137 deposition
The United Nations Scientific Committee on the Effects of Atomic Radiation (UNSCEAR) published Cs-137 deposition and corresponding dose-rate readings at 2180 locations in the Fukushima prefecture or close to the borders of the Fukushima prefecture [49]. These data were provided by the Government of Japan as described in the report titled ‘Summarized version of the results of the research on distribution of radioactive substances discharged by the accident at TEPCO's Fukushima Daiichi NPP’. The Japan Atomic Energy Authority (JAEA) conducted the survey with cooperation of universities and research institutes. The Ministry of Education, Culture, Sports, Science, and Technology in Japan (MEXT) was responsible for the measurements and their validity. UNSCEAR reviewed and published the dataset [49]. The single dose-rate readings resulting from all relevant deposited radionuclides including Cs-134 range from 0.040 µSv/h to 54.800 µSv/h, with mean 1.259 µSv/h and median 0.40µSv/h. The single C-137 measurements range from 590 Bq/m2 to 15,450,928 Bq/m2, with mean 153,957 Bq/m2 and median 39,714 Bq/m2. A 2nd degree regression of the dose-rate on the Cs-137 deposition allows the translation of fallout to dose-rate. The functional details of this association are presented in Figure 3 and the resulting dose-rates are listed in Table 3. Unfortunately, fine-resolution contamination data as available for the Fukushima prefecture, does not exist for all of Japan.
Statistical methods
A powerful method to assess data that contains spatial as well as temporal information is spatiotemporal (logistic) regression [38, 53-57]. The basic idea is to adjust a regression model for region-specific trend functions and to allow for local or global level-shifts at certain points in time, or, preferably, to allow for the heights of local drops or jumps to be proportional to the contamination level of the regional strata. The advantage of this spatiotemporal method is optimization of adjustment and minimization of confounding by considering partial trends of regional units. Values of the outcome variable (here LBW) in those partial trends are modeled and compared within the same regional stratum, as the target variable describing the characteristic of interest varies from year to year. Information on several regional units is combined in a global spatiotemporal model, giving rise to tests of local or global change-points in time as well as spatial trends in the outcome variable with regionally determined contamination or exposure. In the present study, we considered the 47 prefectures of Japan with their documented individual annual LBW proportions (1995 to 2018), their original [48] and rescaled Cs-137 depositions, and the associated dose-rates compiled in Table 3. We tested whether possible changes of LBWp from 2012 onward were dependent on Cs-137 fallout at the prefecture level. We further adjusted the spatiotemporal logistic regression model for the annual population size of the prefectures (1995 to 2015), for the number of physicians per 1000 population, as well as for the counts of the dead, the missing, and the evacuees due to the earthquake and the tsunami as surrogate measures for the available prefecture-specific medical infrastructure and the stress associated with this triple catastrophe and its aftermath, see Table 2. For data processing, statistical analyses, and results display, we used Microsoft Excel 2016 (Office 365), R 3.5.1 (Version 2017-10-04), Wolfram MATHEMATICA 11.3, and mostly SAS/STAT software 9.4 (SAS Institute Inc: SAS/STAT User’s Guide, Version 9.4, Cary NC: SAS Institute Inc, © 2002–2012).