Data-driven approach in a compartmental epidemic model to assess undocumented infections

Abstract

Nowcasting and forecasting of epidemic spreading, fundamental support for policy makers' decisions, rely on incidence series of reported cases to derive the fundamental epidemiological parameters. Two relevant drawbacks for predictions are the unknown fraction of undocumented cases and levels of nonpharmacological interventions that span highly heterogeneously across different places. We describe a simple approach using a compartmental model including asymptomatic and pre-asymptomatic contagions that allows to estimate both the level of undocumented infections and the value of Rt from reported case series in terms of epidemiological parameters. The method was applied to epidemic series for of COVID-19 across different municipalities in Brazil allowing to quantify the heterogeneity level of under-reporting across different places. The reproductive number derived within the current framework is little sensitive to both diagnosis and infection rates during the asymptomatic states, while being very sensitive to variations in case count series. The methods described here are general and we expect that they can be extended to other epidemiological approaches and surveillance data.