Sir Model

In January 2020, the first confirmed case of the novel severe acute respiratory syndrome coronavirus 2 emerged in the United States of America. By March 2020, the USA had declared a national emergency and implemented stay-at-home policies subject to the individual initiative of health authorities of each state. However, ambiguity in the literature exists about the extent to which temporal variation of stay-at-home implementation contributes to an effective stay-at-home order. To examine the role of the implementation of stay-at-home policy at the county level on outbreak progression, we compiled the case count data and dates of policy commencement for 1720 counties from the US Counties: Socio-Health Data database. Measures of central tendency and rate of change identified correlation between the change of confirmed case counts compared to time, quantified by comparing four successive time points of 5 days to the initial date of each county's stay-at-home implementation. We then used a deterministic county-level SIR epidemiological model to predict post stay-at-home case counts based on pre-stay-at-home parameters and compared the model to actual post-stay-at-home case counts to identify the degree of error Mean Squared Error (MSE). Our analyses demonstrated the high error between time since stay-at-home implementation and change in actual case counts compared to predicted case counts, which suggests an interaction between policy and COVID-19 transmission. Our findings shine light on the confounding variables of stay-at-home policy at the county level and the promising outlook of stay-at-home policy in the USA.