Predicting the Impacts of a Pandemic: An Analysis of the Relationship Between Social Vulnerability and COVID-19 in the United States

• Activity Number: 75 - SPEED: Data Challenge and SPAAC
• Type: Topic-Contributed
• Date/Time: Monday, August 9, 2021 : 10:00 AM to 11:50 AM
• Sponsor: Section on Statistical Computing
• Author(s): Coby Warkentin* and Lydia Carter and Thomas J Fisher
• Companies: Miami University and Miami University and Miami University
• Keywords: DataExpo; COVID-19; Social Vulnerability; Predictive Modeling; American Community Survey

Abstract

The COVID-19 pandemic affected nearly every aspect of life over the past year; however, not all regions of the country have been impacted equally. We determine the impact of COVID-19 across the United States and how well it was predicted by “social vulnerability,” which is defined by the CDC as “potential negative effects on communities caused by external stresses on human health.” The 2018 Social Vulnerability Index (SVI) predicts the level of assistance communities would need after experiencing external stresses such as natural disasters or disease outbreaks. To quantify the impact of COVID-19 at the county level, we utilize multiple measures of COVID-19 impact (e.g., case rates, death rates, changes in unemployment) and develop a COVID-19 impact score. This constructed score is then compared to the SVI to determine how well social vulnerability predicted COVID-19 impact. In order to better predict impacts of COVID-19, and thus better predict vulnerability to a widespread pandemic, we create our own model using data from the American Community Survey from the U.S. Census Bureau, and other sources, to predict COVID-19 impact scores.