State and County-Level Factors Associated with the Effectiveness of Stay-At-Home Orders Issued in the United States in Response to COVID-19

TitleState and County-Level Factors Associated with the Effectiveness of Stay-At-Home Orders Issued in the United States in Response to COVID-19 BackgroundTo slow the spread of COVID-19 and protect medical facilities from overflowing, Stay-At-Home Orders (SAHOs) were issued in the United States during the spring of 2020. These orders had variable levels of effectiveness and profound consequences that continue to manifest long after their termination. This study aimed to assess if state and county-level population characteristics could explain variability in SAHO effectiveness as measured by the effective reproductive number (Rt). MethodsWe calculated the Rt for the 40 states which enacted SAHOs, and also for a sample of 289 counties that issued SAHOs in 2020, using EpiEstim R Package based on the states and counties daily case data. We determined SAHOs to be effective if, three weeks after their implementation, Rt was equal to or less than one. Wilcoxon rank sum tests and logistic regression were used to determine if population characteristics (age, income, level of education, political orientation, percent of non-English speaking people, racial and ethnic compositions), and percentage of frontline workers, percent of eligible people vaccinated by July 2021, level of viral transmission, and other Non-Pharmaceutical-Interventions (NPI) enacted before the SAHO, were associated with effectiveness of SAHOs. ResultsSAHOs were effective in 20 (50%) states. No significant differences were found in the characteristics studied between states with effective and ineffective SAHOs. SAHOs were effective in 54% of counties. Counties with effective SAHOs had fewer days of NPIs before the SAHOs in comparison to counties with ineffective SAHOs (Median 24 vs. 34, p-value 0.005). All other characteristics considered showed non-significant differences. In multivariate analysis, days of NPIs before the SAHOs remained the only significant factor for effective SAHOs in studied counties. ConclusionOur analysis suggests that SAHO effectiveness may be influenced by the implementation of prior public health interventions but is not likely to be related to the other characteristics studied. These findings should be considered when assessing when and how to implement SAHOs in future epidemics to limit the spread of an infectious respiratory disease.