|
Abstract:
|
Viruses spread indiscriminately, yet the effects of COVID-19 have not been distributed evenly throughout the United States. CDC reports highlight the extant differences in infection and death among different demographic groups, in part driven by access to healthcare infrastructure. We examined these disparities by extending the network autoregressive model to a binomial response: number of hospital beds occupied by covid patients. Using a network autoregressive model allowed us to draw inference on community demographics while respecting the spatial and temporal dependencies of the COVID pandemic. We found evidence that different demographic groups did not bear equal COVID burdens. Because our model can identify communities in need, these findings may be useful in guiding public policy concerning vaccine distribution and future healthcare infrastructure investment.
|
