Guidance

UK COVID-19 alert level methodology: an overview

Updated 29 March 2023

As of 29 March 2023, the UK COVID-19 alert level system has been suspended.

The suspension of the system reflects the transition to Living with COVID-19, which is thanks to the ongoing success of the vaccination programme and the availability of treatments for those who need them.

The UK Health Security Agency (UKHSA) continues to track the latest COVID-19 epidemiology through numerous surveillance systems, including the weekly National flu and COVID-19 surveillance reports.

Introduction

One of the key objectives of the UK Health Security Agency (UKHSA) is to provide advice to the UK chief medical officers (CMOs) who in turn advise ministers on the UK coronavirus (COVID-19) alert level. The alert levels were originally defined in the UK government’s COVID-19 recovery strategy ‘Our Plan to Rebuild’ in May 2020 (most recently revised in August 2022) and communicate the current risk at a UK-wide level.

The alert levels are:

• level 1: COVID-19 is present in the UK, but the number of cases and transmission is low
• level 2: COVID-19 is in general circulation in the UK, but direct COVID-19 healthcare pressures are low and transmission is declining or stable
• level 3: COVID-19 is in general circulation in the UK
• level 4: COVID-19 is in general circulation in the UK; transmission is high and direct COVID-19 pressure on healthcare services is widespread and substantial or rising
• level 5: as level 4 and there is a material risk of healthcare services being directly overwhelmed by COVID-19

This document outlines the criteria used by UKHSA as it considers its recommendation for the CMOs. The initial methodology was developed following consultation with national public health experts, reviewed and informed by the Scientific Advisory Group for Emergencies (SAGE) and agreed by the UK’s CMOs.

The methodology will evolve as UKHSA learns from current operations and as information streams develop. It is therefore subject to future review by the UKHSA Technical Board, chaired on a rotating basis by the 4 UK CMOs.

Approach

UKHSA’s approach is focused on the criteria to move between levels, rather than criteria that define an individual level. The indicators presented below are considered in the context of a range of measures. Determining the alert level is not an automated or purely statistical process. The approach aims to blend expert judgement and risk assessment with the more quantifiable indicators and thresholds outlined below. This informs an overall assessment of the situation and an alert level recommendation to CMOs.

The UK COVID-19 alert level is focused on data that directly relates to COVID-19 impacts. The one exception is the move between level 4 and level 5, where the UKHSA and CMOs consider all source pressures as this informs the probability of healthcare services being overwhelmed.

Our operational aim is to avoid unnecessary, unpredictable or frequent changes to the alert level, account for the nationwide complexity of the epidemic and reflect our evolving understanding of COVID-19 and its transmission.

Timing

It will sometimes be necessary to escalate the alert level as rapidly as possible, to signal an urgent, escalating national public health crisis. Conversely, as the risk posed by COVID-19 drops, it will be important to ensure changes to the alert level are undertaken in a stable fashion and that a long-term downward trend in new infections has been established. As a guiding principle, UKHSA will ensure that most, if not all, indicators have been met when recommending to CMOs a reduction in the alert level.

Following any recommendation to de-escalate, a minimum of 4 weeks is allowed before any subsequent recommendation to de-escalate further. During the 4 weeks immediately following a de-escalation, epidemic trajectory will continue to be monitored and a recommendation to escalate the alert level may still be made during this time if required.

Indicators for escalation and de-escalation of UK COVID-19 alert levels

A range of indicators and thresholds are used to support the underpinning analysis for alert level recommendations. A recommendation to raise the alert level is most likely to be made based on a combination of the indicators described for each threshold. The main exception to this is the risk posed by the emergence or spread of a new variant. Should a new variant be more transmissible or have vaccine escape characteristics it could be highly likely to lead to changes in the national epidemiology in the short-term. As a result, in exceptional circumstances a change to the alert level could be recommended based on the risk posed by the emergence of a new variant alone.

To avoid unnecessary, unpredictable or frequent changes to the alert level, at least 4 weeks of consecutive decline in important metrics would be required before considering de-escalation between levels.

Escalation and de-escalation indicators and thresholds for each alert level are listed below.

Alert level 1

Indicator

There is a single holding indicator, with recommendation for level 1 only possible if that indicator is met:

• Are current COVID-19 cases and transmission sufficiently low to support being at level 1?

Rationale

A recommendation to de-escalate to alert level 1 would require consensus between UK CMOs and UKHSA that case numbers and UK transmission of COVID-19 are sufficiently low to pose minimal direct acute threat to the UK population. Evidence will be considered using information provided by health protection and central surveillance teams in each UK nation and other sources as appropriate. Should the above indicator not be met, the alert level recommendation would be escalate to, or remain at, level 2.

Escalating from level 2 to level 3

Indicators

• Is the national R reliably estimated to be ≥1?
• Are case numbers and transmission persistently rising?
• Are direct COVID-19 healthcare pressures rising?
• Is current direct COVID-19 absolute healthcare pressure sufficiently high to support escalation to level 3?

Rationale

At this level many statistical or mathematical measures, such as R, are unlikely to be reliable. Weekly confirmed case rates and other markers of infection burden dependent on testing will be considered alongside representative surveillance measures such as the ONS community infection survey positivity estimates. At all levels of the alert system, the absolute level of direct healthcare pressures will be compared with previous phases of the pandemic and considered alongside contextual factors like disease severity, health inequalities, population immunity (including vaccination uptake) and likely upcoming behavioural changes.

Sources include laboratory test results, public health surveillance systems and modelling from SPI-M/SAGE.

Escalating from level 3 to level 4

Indicators

• Is the national R reliably estimated to be R>1?
• Is the doubling time of confirmed new infections less than 7 days?
• Are there more than 30,000 estimated new infections in the UK per day?
• Are COVID-19 related hospital admissions increasing at ≥25% over the same 7-day period?
• Is COVID-19 related hospital occupancy increasing at ≥25% over the same 7-day period?
• Are COVID-19 related high dependency units (HDU) or intensive care units (ICU) admissions and/or occupancy increasing?
• Are new daily COVID-19 related deaths increasing?
• Is current direct COVID-19 absolute healthcare pressure sufficiently high to support escalation to level 4?

Rationale

Hospital activity and severe health outcomes are the key indicators representing healthcare pressures, but these will be subject to a lag from the point of infection. Hence the inclusion of transmission dynamics, doubling time and estimated incidence.

Sources include laboratory test results, hospital admissions and death data (available on the GOV.UK dashboard). Estimated new infections will be informed by a range of sources, including survey data (for example Office for National Statistics (ONS) and ZOE/KCL) and mathematical modelling provided by SPI-M/SAGE.

Alert level 5

Indicator

There is a single holding indicator, with level 5 only recommended if that indicator is met:

• Has UKHSA, in consultation with NHS senior leadership and CMOs, estimated that forecasted healthcare demand will outmatch forecasted capacity across the UK, regions or devolved administrations within the next 21 days?

Should the above indicator not be met, de-escalation to, or remaining at, level 4 can be considered. This ensures that the focus remains on NHS capacity and operational pressures when COVID-19 infections are at high levels.

Rationale

A recommendation to escalate to COVID-19 alert level 5 should be made in consultation with health service directors and contingency planners, and should be based around their predicted capacity, which includes surge capacity and mutual aid. In principle, escalation to level 5 should allow sufficient time for the implementation of urgent national measures to protect healthcare services from being overwhelmed.

De-escalating from level 4 to level 3

Indicators

• Is the national R reliably estimated to be <1?
• Are there estimated to be less than 30,000 new infections per day?
• Have new daily COVID-19 confirmed infections been on a downward trend, or stable at a low level, for at least 4 weeks?
• Have COVID-19 related hospital admissions been on a downward trend, or stable at a low level, for at least 4 weeks?
• Has COVID-19 related hospital occupancy been on a downward trend, or stable at a low level, for at least 4 weeks?
• Have COVID-19 related HDU or ICU admissions and/or occupancy been on a downward trend, or stable at a low level, for at least 4 weeks?
• Have new daily COVID-19 related deaths been on a downward trend, or stable at a low level, for at least 4 weeks?
• Is current direct COVID-19 absolute healthcare pressure sufficiently low to support de-escalation to level 3?

Rationale

Estimated transmission dynamics of R<1 for a sustained period, combined with a demonstrable reduction in the number of people becoming severely unwell and/or dying, would give confidence that this relatively high degree of risk is receding.

Sources include laboratory test results, hospital admissions and death data (available on the GOV.UK dashboard), survey data (for example ONS and ZOE/KCL), and mathematical modelling provided by SPI-M/SAGE.

De-escalating from level 3 to 2

Indicators

• Is the national R reliably estimated to be <1?
• Are cases and transmission falling?
• Are direct COVID-19 healthcare pressures falling?
• Is current direct COVID-19 absolute healthcare pressure sufficiently low to support de-escalation to level 2?

Rationale

Sources include laboratory test results, hospital admissions and death data (available on the GOV.UK dashboard), mathematical modelling provided by SPI-M/SAGE, and public health surveillance systems.