Estimates for relative impact of different COVID-19 control measures in UK

Caption: Empty London. Credit: LSHTM

The London School of Hygiene & Tropical Medicine’s Centre for the Mathematical Modelling of Infectious Diseases (CMMID) is conducting crucial work on the COVID-19 pandemic.

Since February, CMMID has been asked by governmental advisory bodies to project the possible trajectory of the COVID-19 epidemic in the UK under a range of different scenarios, including introduction of non-pharmaceutical interventions such as school closures, social distancing shielding of the elderly and adults in high-risk groups. These large-scale interventions, including home isolation, aim to reduce coronavirus transmission.

Estimates (not peer reviewed) in a new pre-print, combine some of the work CMMID has been conducting for the government in February and March, with more current data which has become available as this fast-moving epidemic evolves.

Method

The team developed a mathematical model to simulate the COVID-19 epidemic across 186 county-level administrative units of the UK. The model starts from an initial importation of infections, then simulates how these might transmit among the community.

It uses age-specific probability of developing symptoms on infection, age-specific contact rates, and age-specific hospitalisation and death rates. The team also modeled the likely requirements for non-ICU hospital beds and ICU beds over time.

The interventions are modelled by simulating changes to these contact patterns. For example, for school closures, contacts made in schools are decreased to 0 in the model, with potentially increased contact between children and older adults to take account of increased care by grandparents.

Findings

The team estimated that an unmitigated epidemic could result in 16-30 million symptomatic COVID-19 cases and 250,000-470,000 deaths. In this scenario the NHS’s hospital and ICU bed capacity would be massively exceeded.

Shorter-term (1 week to 12 week) interventions have typically been used to mitigate the burden of pandemic influenza and SARS. In this scenario the team found that these would likely be inadequate for COVID-19 in the UK. ICU bed requirements would exceed availability, by a factor of 10-30.

In a third scenario, longer-term (1 year) programmes of social distancing and protecting the most vulnerable, plus periodic ‘lockdowns’ with stricter measures (like the current UK one), have the potential to keep NHS capacities from being exceeded. However, these would likely need to be in place for many months at least.

Nick Davies from CMMID and member of the modelling team said: “The Centre is proud to be one of the modelling teams in the UK working around the clock on COVID-19. Scenarios on how the epidemic might play out from LSHTM and others are being used by the UK government to make evidence-based strategy decisions. However, it’s important to highlight that the effect of interventions on the epidemic trajectory and health burden is only one aspect of making these policy choices.

“Our estimates do not take into account pharmaceutical interventions – the soonest a vaccine could become available is probably a year. However, let’s hope the therapeutic drugs being tested or developed are available much sooner so we can save as many lives as possible.”

The team highlight that this work was conducted before the UK lockdown, while the outbreak was growing. The Centre is developing full analysis of potential exit strategies.

They also acknowledge limitations of their work. Because much remains uncertain about COVID-19, in many cases the best estimates for transmission rates and health burdens come from China and other countries where the coronavirus has been circulating for longer. Moreover, we are just now starting to get a full picture of the impact that social distancing measures have had in the UK. UK-specific estimates of health burdens and healthcare demand, and the effect of social distancing measures-combined with real-time tracking of transmission in the UK-would help us to have more confidence in projections. These are in progress for follow-up studies.

**This work has not been peer reviewed**

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

/Public Release. The material in this public release comes from the originating organization and may be of a point-in-time nature, edited for clarity, style and length. View in full here.