Perinatal mortality after the Fukushima nuclear accident: An ecological study

OBJECTIVE: This study continues former studies on perinatal mortality in Japan after the Fukushima Daiichi nuclear power plant (FDNPP) accident in March 2011. An increased study region is chosen, and the study period is extended to 2019. METHODS: Japanese monthly perinatal mortality data are provided on a prefecture level by the Japanese government. The study region consists of 12 prefectures around the FDNPP; the rest of Japan is used as the control region. A combined non-linear regression of perinatal mortality rates in the study- and control regions is conducted. The regression model allows for a common asymptotic lower limit of perinatal mortality, seasonal variations, and periodic peaks in 2012–2019 in the study region. To determine the dependency of the effect on distance from the FDNPP, the study region is divided into four core prefectures and eight prefectures surrounding the core prefectures. RESULTS: Perinatal mortality rates in the study region show a significant 6.4% (95% CI: 1.8%, 13.4%) overall increase in 2012–2019 relative to the trend in preceding years with no attenuation during 2012–19. The increase translates to 590 (165, 1226) excess perinatal deaths (p = 0.016). It is characterized by annual peaks with maxima in April. A 13.6% increase is determined in the four core prefectures and a 4.3% increase in eight prefectures surrounding the core prefectures. Before 2012, there is a peak around April 2011 and a decline in October 2011; another significant peak is detected in November 2012. In the 4 core prefectures, large increases are found in the first quarter of 2018 (+70%) and in May 2019 (+130%). CONCLUSION: This study finds periodic peaks in perinatal mortality in spring 2012–2019 in 12 prefectures of Japan surrounding the FDNPP. In light of massive increases in 2018 and 2019 in the four core prefectures, continued investigation of perinatal mortality in contaminated regions of Japan is recommended.