Fig. 5: Heterogeneous effects of the COVID-19 pandemic across geography.
From: Increase in suicide following an initial decline during the COVID-19 pandemic in Japan
a–c, Effects of the pandemic on suicide rate during the first and second outbreaks across locations with different original suicide risk before the pandemic (a), different prefectural level confirmed COVID-19 cases (b), and different base income per capita (c). Blue and red circles denote the DID estimates during the first and second outbreaks, respectively, and error bars show 95% CI. If the variable in a city or a prefecture is lower than the median, the city or the prefecture is assigned to the low group; otherwise, it is in the high group. Full results are presented in Supplementary Table 7. All regressions include city-by-year fixed effects and city-by-month fixed effects and are weighted by population, and standard errors are clustered at the city level. N = 30,134 (a, low), 31,075 (a, high), 24,246 (b, low), 36,963 (b, high), 26,166 (c, low), and 35,043 (c, high). The separated observations are excluded (Methods).
