Impact of Social Vulnerability on COVID-19 Incidence and Outcomes in the United States

Aditi Nayak; Shabatun J Islam; Anurag Mehta; Yi-An Ko; Shivani A Patel; Abhinav Goyal; Samaah Sullivan; Tene T Lewis; Viola Vaccarino; Alanna A Morris; Arshed A Quyyumi

The copyright holder for this preprint (which was not peer-reviewed) is the . https://doi.org/10.1101/2020.04.10.20060962 doi: medRxiv preprint Generalized linear mixed models, with negative binomial distribution or Poisson distribution when appropriate, 11 were used to examine the association of outcomes with SVI (reported as percentile of social vulnerability, with higher numbers representing increased vulnerability) and its sub-components including socioeconomic status, household composition, minority status, and housing type/transportation accessibility (Supplement). 5,6 Given differences in COVID-19 testing by state, state-specific random intercepts were incorporated in models to account for correlations among counties within the same state. Covariates included percentage of population aged >65 years and average HCC score. A sensitivity analysis was conducted using data from six states (New York, New Jersey, Washington, Massachusetts, Vermont, and Louisiana) with the highest levels of testing as of April 4 th , 2020. 12 Age-and HCC score-adjusted CFR and incidence were compared across medians of overall SVI using one-way ANOVA.