COVID-19 Progression Timeline and Effectiveness of Response-to-Spread Interventions across the United States

Pai Liu; Payton Beeler; Rajan K Chakrabarty

is the (which was not peer-reviewed) The copyright holder for this preprint . https://doi.org/10.1101/2020.03. 17.20037770 doi: medRxiv preprint Figure 3 (a) indicates that it is already too late (up to now, March 16, black symbols) for a wholesale traffic restriction to be effective in containing the epidemic spread, since virtually all US states have been seeded with COVID-19 patients. The previous modelling study on the COVID-19 spread in mainland China (2) has also arrived at a similar conclusion regarding the futility of mobility reduction in delaying epidemic spread. Nonetheless, it is worth noting that the reduction in interstate mobility in the US could slightly desynchronize the state-wise epidemic dynamics (i.e. the moderate broadening of the time window for the local epidemic peaks shown in panel (a)). This desynchronization could potentially alleviate the burnden on the limited available medical resources on a national level. The side-by-side comparison with the forecast made on March 2 (red symbols) indicates that should an aggressive (c.a. 90% mobility reduction) traffic restriction have been emplemented two weeks before, the national epidemic peak could have been delayed substaintially. Panel (b) shows that 90% reduction of interstate mobility, if implemented on March 2, could have reduced the epidemic peak magnitude by up to 30%. However, up to March 16, the influence of mobility reduction on epidemic peak magnitude has dwindled to a negligible level. (d) show that till date reducing COVID-19 transmissibility remains to be an effective intervention approach. A 25% reduction in transmissibility ( 0 reduced to 2.01) across all states could delay the national epidemic peak by about 35 days and reduce its magnitude by 39%. A 50% reduction in transmissibility ( 0 reduced to 1.34) will contain the spread of COVID-19 (with the national epidemic peak postponed to winter 2020 and the peak magnitude reduced to 1%). These observations agree with Ref.