How will this continue? Modelling interactions between the COVID-19 pandemic and policy responses

Much of the uncertainty about the progression of the COVID-19 pandemic stems from questions about when and how non-pharmaceutical interventions (NPI) by governments, in particular social distancing measures, are implemented, to what extent the population complies with these measures, and how compliance changes through time. Further uncertainty comes from a lack of knowledge of the potential effects of removing interventions once the epidemic is declining. By combining an epidemiological model of COVID-19 for the United Kingdom with simple sub-models for these societal processes, this study aims to shed light on the conceivable trajectories that the pandemic might follow over the next 1.5 years. We show strong improvements in outcomes if governments review NPI more frequently whereas, in comparison, the stability of compliance has surprisingly small effects on cumulative mortality. Assuming that mortality does considerably increase once a countrys hospital capacity is breached, we show that the inherent randomness of societal processes can lead to a wide range of possible outcomes, both in terms of disease dynamics and mortality, even when the principles according to which policy and population operate are identical.. Our model is easily modified to take other aspects of the socio-pandemic interaction into account.