Modeling the Covid-19 Pandemic Response of the US States

Abstract

Background: The United States of America (USA) has been the country worst affected, in absolute terms, by the Covid-19 pandemic. The country comprises 50 states under a federal system. The impact of the pandemic has resulted in different responses at the state level, which are driven by differing intervention policies, demographics, connectedness and other factors. Understanding the dynamics of the Covid-19 pandemic at the state level is essential in predicting its future evolution. Objective: Our objective is to identify and characterize multiple waves of the pandemic by analyzing the reported infected population curve in each of the 50 US states. Based on the intensity of the waves, characterized by declining, stationary, or increasing strengths, each state's response can be inferred and quantified. Methods: We apply a recently proposed multiple-wave model to fit the infected population data for each state in USA, and use the proposed Pandemic Response Index to quantify their response to the Covid-19 pandemic. Results: We have analyzed reported infected cases from each one of the 50 USA states and the District of Columbia, based on the multiple-wave model, and present the relevant parameters. Multiple waves have been identified and this model is found to describe the data better. Each of the states can be classified into one of three distinct classes characterized by declining, increasing, or stationary strength of the waves following the initial one. The effectiveness of intervention measures can be inferred by the peak intensities of the waves, and states with similar population characteristics can be directly compared. We estimate how much lower the number of infections might have been, if early and strict intervention measures had been imposed to stop the disease spread at the first wave, as was the case for certain states. Based on our model's results, we compute the value of the Pandemic Response Index, a recently introduced metric for quantifying in an objective manner the response to the pandemic. Conclusions: Our results reveal a series of epidemic waves, characterizing USA's pandemic response at the state level, and also infer to what extent the imposition of early intervention measures could have had on the spread and impact of the disease. As of June 11, 2020, only 19 states and the District of Columbia (40% of the total) clearly exhibit declining trends in the numbers of reported infected cases, while 13 states exhibit stationary and 18 states increasing trends in the numbers of reported cases.

Environment influences SARS-CoV-2 transmission in the absence of non-pharmaceutical interventions

“…Dynamics and reporting between USA states are known to be variable 38 , introducing a level of uncertainty to our findings. Furthermore, lockdown measures were (and continue to be) quite heterogeneous across the USA, with different states displaying different levels of response to COVID-19 39 . Through our epidemiological modelling approach we are able to account for these different state-level responses using google mobility data.…”

Temperature and population density influence SARS-CoV-2 transmission in the absence of nonpharmaceutical interventions

“…Dynamics and reporting between US states are known to be variable (46), introducing a level of uncertainty to our findings. Furthermore, lockdown measures were (and continue to be) quite heterogeneous across the United States, with different states displaying different levels of response to COVID-19 (47). Through our epidemiological modeling approach, we are able to account for these different state-level responses using Google mobility data.…”