Assessing the impact of reduced travel on exportation dynamics of novel coronavirus infection (COVID-19)

Asami Anzai; Tetsuro Kobayashi; Natalie M. Linton; Ryo Kinoshita; Katsuma Hayashi; Ayako Suzuki; Yichi Yang; Sungmok Jung; Takeshi Miyama; Andrei R. Akhmetzhanov; Hiroshi Nishiura

As another measure of impact, we estimated the probability of a major epidemic. Figure 3A shows the probability of a major epidemic with three different levels of transmissibility assuming an R0 of 1.5, 2.2, or 3.7, and three different levels of contact tracing resulting in a success rate of isolation of the traced contacts of 10%, 30%, or 50%. Without the reduction in the travel volume, the probability of a major epidemic exceeded 90% in most scenarios. However, considering there have been six untraced cases in Japan under travel restrictions, the probability of a major epidemic more broadly ranged from 56% to 98%. Figure 3B shows the reduced probability of a major epidemic. Assuming an R0 of 2.2, the absolute risk reduction was 7%, 12%, and 20%, respectively, for contact tracing levels leading to isolation at 10%, 30%, and 50%. Figure 3B describes the absolute reduction in risk of a major epidemic. The largest reduction was 37% when R 0 =1.5 and 50% of contacts were traced. The smallest reduction was 1% when R 0 =3.7 and 10% of contacts were traced. Using those estimated relative reductions, the median time of delay gained by travel volume reduction is shown in Figure 4 . The time delay of a major epidemic was less than one day when R 0 is 2.2 and 3.7, and 1 to 2 days when R 0 is 1.5. . Delay in the time to a major epidemic gained by travel volume reduction. The median delay is shown for Japan, using relative reduction in the probability of a major epidemic. The vertical axis represents the time delay to a major epidemic (in days), and the horizontal axis represents the proportion of contacts traced. Each shaped dot represents different values of the basic reproduction number.