Estimating and forecasting the burden and spread of SARS-CoV2 first wave in Colombia

Abstract

Following the rapid dissemination of COVID-19 cases in Colombia in 2020, large-scale non-pharmaceutical interventions (NPIs) were implemented as national emergencies in most of the municipalities of the country starting by a lockdown on March 20th of 2020. Using combinations of meta-population models SEAIIRD (Susceptible-Exposed-Asymptomatic-Infected-Recovered-Diseased) which describes the disease dynamics in the different localities, with movement data that accounts for the number of commuters between units and statistical inference algorithms could be an effective approach to both nowcast and forecast the number of cases and deaths in the country. Here we used an iterated filtering (IF) framework to fit the parameters of our model to the reported data across municipalities from march to late October in locations with more than 50 reported deaths and cases historically. Since the model is high dimensional (6 state variable by municipality) inference on those parameters is highly non-trivial, so we used an Ensemble-Adjustment-Kalman-Filter (EAKF) to estimate time variable system states and parameters. Our results show that the model is capable of capturing the evolution of the outbreak in the country and providing estimates of the epidemiological parameters in time. These estimates could become the base for planning future interventions as well as evaluate the impact of NPIs on the effective reproductive number (ℛeff) and the key epidemiological parameters, such as the contact rate or the reporting rate. Our approach demonstrates that real-time, publicly available ensemble forecasts can provide robust short-term predictions of reported COVID-19 deaths in Colombia. This model has the potential to be used as a forecasting and prediction tool to evaluate disease dynamics and to develop a real time surveillance system for management and control.