State-dependent patterns and coupling between the vaccination schedule, population mobility and the COVID epidemic outline, in the US states

We study the evolution of the COVID-19 epidemic in the US, since January 2020 until May 2021. Our primary goal is to understand some of the complex coupled dynamics between factors that ultimately regulate the epidemic case load. As potentially crucial factors, we focus on population mobility and vaccination patters (both related to risk of contracting the SARS-Cov2 virus). These factors may in turn depend on demographic parameters (which are unrelated to the epidemic evolution), but also on the population response to the epidemic outbreak itself. In our work, we use correlation analyses, in conjunction with open source data from US states, to investigate the type and strength (1) of the relationships between demographic measures and epidemic, mobility and vaccination timelines during our established time window; (2) of the bidirectional coupling between these timelines. We showed that the wide between-state differences in epidemic outcome correspond to between-state differences in demographic measures (such as density, income, political affiliation). As a potential underlying mechanism, we found that demographic measures are also predictive of the degree of coupling between epidemic timelines (on one hand) and vaccination and mobility timelines (on the other hand), coupling which can be broadly interpreted as the populations behavioral response to the epidemic. In support of this idea, our analysis shows this response to be tightly correlated with epidemic outcome. This suggests that a states demographic profile may be invaluable to generating predictions on the epidemic evolution in the respective state, and that this information may be used to understand the weaknesses of a state and how to compensate for them to improve epidemic outcome (e.g., via state centralized incentives, and customized mitigation strategies).