The effect of non-pharmaceutical interventions on COVID-19 cases, deaths and demand for hospital services in the UK: a modelling study

<jats:title>Abstract</jats:title><jats:sec><jats:title>Background</jats:title><jats:p>Non-pharmaceutical interventions have been implemented to reduce transmission of SARS-CoV-2 in the UK. Projecting the size of an unmitigated epidemic and the potential effect of different control measures has been critical to support evidence-based policymaking during the early stages of the epidemic.</jats:p></jats:sec><jats:sec><jats:title>Methods</jats:title><jats:p>We used a stochastic age-structured transmission model to explore a range of intervention scenarios, including the introduction of school closures, social distancing, shielding of elderly groups, self-isolation of symptomatic cases, and extreme “lockdown”-type restrictions. We simulated different durations of interventions and triggers for introduction, as well as combinations of interventions. For each scenario, we projected estimated new cases over time, patients requiring inpatient and critical care (intensive care unit, ICU) treatment, and deaths.</jats:p></jats:sec><jats:sec><jats:title>Findings</jats:title><jats:p>We found that mitigation measures aimed at reducing transmission would likely have decreased the reproduction number, but not sufficiently to prevent ICU demand from exceeding NHS availability. To keep ICU bed demand below capacity in the model, more extreme restrictions were necessary. In a scenario where “lockdown”-type interventions were put in place to reduce transmission, these interventions would need to be in place for a large proportion of the coming year in order to prevent healthcare demand exceeding availability.</jats:p></jats:sec><jats:sec><jats:title>Interpretation</jats:title><jats:p>The characteristics of SARS-CoV-2 mean that extreme measures are likely required to bring the epidemic under control and to prevent very large numbers of deaths and an excess of demand on hospital beds, especially those in ICUs.</jats:p></jats:sec><jats:sec><jats:title>Research in Context</jats:title><jats:sec><jats:title>Evidence before this study</jats:title><jats:p>As countries have moved from early containment efforts to planning for the introduction of large-scale non-pharmaceutical interventions to control COVID-19 outbreaks, epidemic modelling studies have explored the potential for extensive social distancing measures to curb transmission. However, it remains unclear how different combinations of interventions, timings, and triggers for the introduction and lifting of control measures may affect the impact of the epidemic on health services, and what the range of uncertainty associated with these estimates would be.</jats:p></jats:sec><jats:sec><jats:title>Added value of this study</jats:title><jats:p>Using a stochastic, age-structured epidemic model, we explored how eight different intervention scenarios could influence the number of new cases and deaths, as well as intensive care beds required over the projected course of the epidemic. We also assessed the potential impact of local versus national targeting of interventions, reduction in leisure events, impact of increased childcare by grandparents, and timing of triggers for different control measures. We simulated multiple realisations for each scenario to reflect uncertainty in possible epidemic trajectories.</jats:p></jats:sec><jats:sec><jats:title>Implications of all the available evidence</jats:title><jats:p>Our results support early modelling findings, and subsequent empirical observations, that in the absence of control measures, a COVID-19 epidemic could quickly overwhelm a healthcare system. We found that even a combination of moderate interventions – such as school closures, shielding of older groups and self-isolation – would be unlikely to prevent an epidemic that would far exceed available ICU capacity in the UK. Intermittent periods of more intensive lockdown-type measures are predicted to be effective for preventing the healthcare system from being overwhelmed.</jats:p></jats:sec></jats:sec>