ABSTRACT
Colombia announced the first case of severe acute respiratory syndrome coronavirus 2 on March 6, 2020. Since then, the country has reported a total of 4,240,982 cases and 106,544 deaths as of June 30, 2021. This motivates an investigation of the SARS-CoV-2 transmission dynamics at the national and regional level using case incidence data. Mathematical models are employed to estimate the transmission potential and perform short-term forecasts of the COVID-19 epidemic trajectory in Colombia. Furthermore, geographic heterogeneity of COVID-19 in Colombia is examined along with the analysis of mobility and social media trends, showing that the increase in mobility in July 2020 and January 2021 were correlated with surges in case incidence. The estimation of national and regional reproduction numbers shows sustained disease transmission during the early phase of the pandemic, exhibiting sub-exponential growth dynamics. Moreover, most recent estimates of reproduction number are >1.0 at the national and regional levels as of May 30, 2021. Further, the 30-day ahead short-term forecasts obtained from Richards model present a sustained decline in case counts in contrast to the sub-epidemic and GLM model. Nevertheless, our spatial analysis in Colombia shows distinct variations in incidence rate patterns across different departments that can be grouped into four distinct clusters. Lastly, the correlation of social media trends and adherence to social distancing measures is observed by the fact that a spike in the number of tweets indicating the stay-at-home orders was observed in November 2020 when the case incidence had already plateaued. Author summaryAs the COVID-19 pandemic continues to spread across Colombia, studies highlighting the intensity of the pandemic become imperative for appropriate resource allocation and informing public health policies. In this study we utilize mathematical models to infer the transmission dynamics of SARS-CoV-2 at the regional and national level as well as short-term forecast the COVID-19 epidemic trajectory. Moreover, we examine the geographic heterogeneity of the COVID-19 case incidence in Colombia along with the analysis of mobility and social media trends in relation to the observed COVID-19 case incidence in the country. The estimates of reproduction numbers at the national and regional level show sustained disease transmission as of May 30, 2021. Moreover, the 30-day ahead short-term forecasts for the most recent time-period (June 1-June 30, 2021) generated from the mathematical models needs to be interpreted with caution as the Richards model point towards a sustained decline in case incidence contrary to the GLM and sub-epidemic wave model. Nevertheless, the spatial analysis in Colombia shows distinct variations in incidence rate patterns across different departments that can be grouped into four distinct clusters. Lastly, the social media and mobility trends explain the occurrence of case resurgences over the time.
ABSTRACT
The COVID-19 pandemic strained healthcare systems in many parts of the United States. During the early months of the pandemic, there was substantial uncertainty about whether the large number of COVID-19 patients requiring hospitalization would exceed healthcare system capacity. This uncertainty created an urgent need for accurate predictions about the number of COVID-19 patients requiring inpatient and ventilator care at the local level. In this work, we develop a Bayesian Susceptible-Infectious-Hospitalized-Ventilated-Recovered (SIHVR) model to predict the burden of COVID-19 at the healthcare system level. The Bayesian SIHVR model provides daily estimates of the number of new COVID-19 patients admitted to inpatient care, the total number of non-ventilated COVID-19 inpatients, and the total number of ventilated COVID-19 patients at the healthcare system level. The model also incorporates county-level data on the number of reported COVID-19 cases, and county-level social distancing metrics, making it locally customizable. The uncertainty in model predictions is quantified with 95% credible intervals. The Bayesian SIHVR model is validated with an extensive simulation study, and then applied to data from two regional healthcare systems in South Carolina. This model can be adapted for other healthcare systems to estimate local resource needs.
