Excess mortality during and after the COVID-19 emergency in Japan: a two-stage interrupted time-series design

The COVID-19 pandemic has had unprecedented impacts on mortality worldwide. This study aimed to estimate excess all-cause mortality in Japan from 2020 to 2023, stratified by age, sex and prefecture, to assess the evolving impact of the pandemic, particularly in the latter years. The study period extends beyond Japan ending the public health emergency.

Using national vital statistics data from 2015 to 2023, we employed a two-stage interrupted time series analysis to estimate excess mortality during the COVID-19 pandemic (2020-2023) compared with the prepandemic period (2015-2019). Models were adjusted for seasonality, long-term trends, temperature and influenza activity. We calculated excess deaths during (14 January 2020 to 7 May 2023) and after (8 May 2023 to 31 December 2023) the COVID-19 emergency.

Japan experienced 219 516 excess deaths (95% empirical CI (eCI) 138 142 to 301 590) during the study period, corresponding to 3.7% (95% eCI 2.33 to 5.09) excess mortality. Excess mortality was negative in 2020 (-1.67%, 95% eCI -2.76 to -0.55), becoming positive in 2021 (2.19%, 95% eCI 0.9 to 3.49) and peaking in 2022 (7.55%, 95% eCI 5.96 to 9.13) before declining in 2023 (5.76%, 95% eCI 4.29 to 7.24). The <60 age group consistently showed the highest percentage excess mortality. Males had slightly higher excess mortality than females. By 2022, all prefectures exhibited positive excess mortality. The relative risk peaked in late 2022, with a smaller peak in summer 2023, coinciding with the post-emergency period. Comparing this post-emergency period with prior years shows the highest percentage excess mortality in 2022. Rural prefectures, and those with lower influenza cases, showed reduced excess mortality during the latter and post-emergency period.

Despite initial success in mitigating excess deaths, Japan saw increasing excess mortality as the pandemic progressed, with continued elevation post-emergency. The varying impact across age groups and time highlights the complex factors affecting mortality. This study's findings underline the importance of continuous monitoring of excess mortality as a key indicator for public health dynamics.