Jamie Song is Tradeoffs Operations Assistant and a soon-to-be graduate of the University of Pennsylvania’s MPH program. His work focuses on spatial epidemiology and health equity, and his capstone research project explores the associations between structural inequities and the burden of COVID-19 on marginalized communities.

In response to the disproportionate burden of COVID-19 on people of color and people working in high-risk industries, prominent figures in health policy including Dr. Marcella Nunez-Smith and Dr. James Hildreth have called for the use of a “vulnerability index” to guide the prioritization of vaccines to those who need them most urgently.

A vulnerability index combines numerous factors (like poverty rate, disability status and English proficiency) to measure how vulnerable a certain community is to a particular risk. Though there are many vulnerability indices designed specifically for COVID-19 — including the Pandemic Vulnerability Index and the COVID-19 Community Vulnerability Index — the original is the CDC’s Social Vulnerability Index (SVI), which was developed to assess vulnerability to emergency and disaster events. It is also the one being used by the Biden administration to prioritize vaccination efforts around the U.S.

In my capstone project — which I completed under the mentorship of Gina South, Sara Solomon and Doug Wiebe, and am planning to publish on medRxiv later this month — we investigated the relationship between a county’s SVI and its COVID-19 cumulative incidence (total cases divided by population) and case fatality risk (total deaths divided by cases) using data from the CDC and Johns Hopkins University. We used a statistical mapping tool called Optimized Hot Spot Analysis, which identified statistically significant clusters or “hot spots” of counties with high COVID-19 cumulative incidence or case fatality risk. Then, we looked for associations between counties’ SVI scores and whether or not they were in a hot spot.

Ultimately, we found that the most vulnerable counties (those with SVI values above the 80th percentile) were 1.63 times more likely to be in a hot spot for case fatality risk than the least vulnerable (SVI values below the 20th percentile). Perhaps surprisingly, we also found that despite having a lower case fatality risk, those counties below the 20th percentile of SVI values were 1.54 to 2.2 times more likely to be in a cumulative incidence hot spot than all other counties. Though we can’t say for sure why this happened, it’s possible people in counties with lower SVI values were more likely to travel or attend social gatherings or had better access to testing and health care. We urge further research on associations between SVI and factors facilitating transmission. We also found that Black people were overrepresented (with respect to the overall U.S. population) in both types of hot spots.