Epidemiological modelling to support the global COVID-19 response: how to mitigate impact in low-income and crisis-affected settings

The current COVID-19 pandemic is the greatest threat posed by a respiratory virus since the 1918 H1N1 influenza pandemic. To date, the majority of epidemiological modelling analyses have focused on high-income countries. However, there is an equivalent need for models appropriate to low- and middle-income countries (LMICs) that comprise 85% of the world's population and have differing demographics and behaviours that are not captured by existing models. To address this, we will use a model of SARS-CoV-2 transmission to forecast epidemics and healthcare needs in LMICs, explore the potential impact of proposed interventions, and estimate their impact in real-time. The model will fit to individual country surveillance data to support estimation of the reproduction number, and projections will be made of the potential impact of alternative mitigation and suppression strategies, including household quarantine and social distancing, both generally and in vulnerable populations. The fit of the model to COVID-19 case count and mortality data collected after the implementation of various interventions will be used in real-time to evaluate their effectiveness in individual LMIC countries and the criteria for lifting social distancing measures explored using the best fit model.