Impact of Social Distancing Measures on COVID-19 Healthcare Demand in Central Texas

Xutong Wang; Remy F Pasco; Zhanwei Du; Michaela Petty; Spencer J Fox; Alison P Galvani; Michael Pignone; S. Claiborne Johnston; Lauren Ancel Meyers

As part of a CDC modeling network, we initially modeled a large number of school closure policies, with variable implementation time and duration. Here, we reported only two of these strategies to demonstrate the impact of implementation time: (i) closure immediately following the first confirmed case (March 14) and (ii) delayed closure two months after the first confirmed case (May 14). In both cases we assumed that schools remain closed through the end of the summer vacation (August 18, 2020) [33] , which corresponds to a 23-week duration for the early closure, and a 14-week duration for the late closure. The early closure scenario roughly corresponded to the city of Austin announcing the first two confirmed cases on March 13th and major school districts closing the next day. In our simulations, the median cumulative number of symptomatic COVID-19 cases by March 14th was 23 (IQR: 17-32) and 9.0 (IQR: 5.8-13) assuming a four-day and seven-day doubling period, respectively; by May 14th, median cumulative symptomatic cases climbed to 508,546 (IQR: 104,535-766,109) and 6,759 (IQR: 1,901-12,353), respectively.