Skip to main content
. 2021 Jul 31;203:111795. doi: 10.1016/j.envres.2021.111795

Fig. 2.

Fig. 2

Shows the interactive effect of the COVID-19 epidemic and TCC on prevalence of psychological distress. Red and blue represent May 2019 (COVID-19 non-epidemic period) and May 2020 (COVID-19 epidemic period), respectively, and shaded areas show the 95% confidence interval. The vertical dash line of x-axis was 22.3%, 26.45% and 33.21%, 38.06% represented the baselines for low, medium, and high residential TCC, respectively. The psychological distress prevalence of a1 subtracting a2 represents the (all possible) decrease in the low TCC level during non-epidemic period, which was obviously much higher than that of b1 subtracting b2 during the epidemic period. In the same way, in the medium TCC level, the decrease (a3 subtracting a4) during the non-epidemic period was lower than the decrease (b3 subtracting b4) during epidemic period. While at the high TCC level, the decrease (a5 subtracting a6) during the non-epidemic period was lower than that (b5 subtracting b6) in the epidemic period. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)