A Guide to Covid Surveillance Metrics

An overview of the surveillance metrics employed by the UK to monitor levels and risks associated with respiratory viruses, including Covid.

Introduction.

The aim of this article is to provide an explanation of the surveillance metrics for respiratory viruses presently published by the UKHSA and how there metrics are used to assess risk. It is intended as a guide to the surveillance metrics that I present in my weekly Covid Situation Reports which are published in the Substack folder found at this link.

Summary.

In the first few months of the Covid pandemic, the UK had limited testing capacity resulting in a lack of data on the prevalence of Covid in the community. Once testing capacity increased, reliable and unbiased estimates of community prevalence were provided through the Coronavirus (COVID-19) Infection Survey (CIS).

However, the CIS survey has now stopped and Covid is presently monitored in the same way as other respiratory infections like Flu. A wide range of surveillance metrics are used to track Covid activity including test positivity and hospital admissions. For some of these metrics experimental thresholds have been established that determine if Covid activity is at a baseline, low, moderate of high level.

Finally, test positivity and hospital admissions are shown to be good proxy measures for community prevalence by comparing data for these measures with prevalence from the CIS survey.

Background.

Effective surveillance, particularly through widespread testing, is crucial in the early stages of a pandemic. At the start of the Covid pandemic, the UK’s limited testing capacity meant that hospital testing was prioritised and community testing did not take place. This meant that there was very limited knowledge of Covid levels in the community with estimates based on Covid hospital admissions — a lagging indicator.

By May 2020, testing capacity had increased sufficiently to allow symptomatic testing in the community, providing a better source of information to assess Covid levels and monitor the progress of the pandemic. Unfortunately, testing capacity was still constrained and in September 2020 demand exceeded capacity as Covid levels surged following the summer lull. This meant that positive tests were not picking up the full extent of Covid in the community.

However, there were other more reliable sources of information available, principally from the UK’s Coronavirus (COVID-19) Infection Survey (CIS). Launched in May 2020, the CIS tracked the prevalence and spread of Covid within private households across the UK. Participants from randomly selected private households were invited to provide regular swab samples. This meant that CIS offered unbiased estimates of Covid prevalence in the community, capturing data on both symptomatic and asymptomatic cases.

The Coronavirus Infection Survey stopped on March 2023, although a scaled back Winter Infection Survey (WIS) took place later in the year. Running from November 2023 to March 2024, the WIS study aimed to monitor Covid prevalence and infection rates during the winter months to inform public health responses and prepare for potential healthcare pressures.

Finally, wastewater monitoring for Covid in England and Wales was also stopped in March 2023, leaving Scotland as the only home nation that continues to monitor Covid levels in wastewater.

In summary, the UK no longer has the detailed information on Covid levels in the community that was previously available. Covid is now monitored in the same manner as other respiratory viruses and the next section provides an overview of the surveillance metrics presently available.

Respiratory virus surveillance metrics.

The United Kingdom Health Security Agency (UKHSA) publishes a weekly National Flu and Covid-19 Surveillance Report during the winter months. This report includes a wide variety of metrics used to monitor a range of respiratory infections including Covid, Flu, Respiratory Syncytial Virus (RSV), and Human metapneumovirus (hMPV) in England.

UKHSA employs a comprehensive approach to monitor and assess respiratory virus activity using data from various surveillance systems to provide a holistic view of the virus's impact. The key components of this approach are:

1. Laboratory Surveillance. Data from testing hospital patients with respiratory infections is collated and analysed. This data is used to calculate test positivity (the percentage of positive tests among all tests conducted), offering insights into community transmission levels for the major respiratory viruses, including Covid.

2. Community Surveillance. Data on Acute Respiratory Infection (ARI) incidents in institutional settings, including care homes, hospitals, schools and prison, are tracked. These provide an indication of the number of outbreaks across these setting caused by various infections, including Covid and Flu.

3. Primary Care Surveillance. Two components of data are provided from General Practices — Syndromic and Virological surveillance. Syndromic surveillance includes GP consultation rates for influenza like illness (ILIs) as well as the reporting of ARI incidents. Virological surveillance covers data from testing GP patients with respiratory symptoms used to calculate test positivity rates.

4. Secondary Care Surveillance. Covid, Flu and RSV admission rates to hospital and critical care units are reported. Whilst the data is primarily used to assess the impact on hospitals it also provides insight in to community prevalence. As most hospital admissions are for patents clinically at risk these measures provide an indication of the risk to individuals in the community from these diseases.

5. Genomic Surveillance: Tracking variants of Covid and Flu to identify emerging risks, and which lineages may be responsible for changes in Covid incidence, transmission and severity.

A fuller and more comprehensive description of the metrics monitored can be found in the UKHSA data quality report that can be accessed here.

Respiratory infection activity thresholds

To aid interpretation of the rates of influenza like illness and allow comparison with previous years, the UK has adopted a standardised method of reporting influenza activity, the Moving Epidemic Method (MEM). This method is based on the World Health Organisations pandemic influenza severity assessment (‎PISA) guidelines and is also used by the European Centre for Disease Prevention and Control.

The MEM method uses historical data to evaluate the timing and duration of an influenza epidemic through a series of cut points, baseline, low, medium, high and very high thresholds. The initial baseline threshold, once breached, typically denotes the start of influenza activity or circulation with the breach of subsequent thresholds denoting the intensity of influenza activity in a particular season.

The following chart illustrates the concept using the latest data for Flu hospital admissions and the current MEM thresholds.

The lowest band, shaded in light green, shows the baseline level of activity below which Flu activity is not considered a risk. As hospital admissions rise the baseline threshold is breached indicating the start of the winter Flu surge. Further increases may then breach the higher thresholds indicating the severity of the surge when compared to previous winters.

The latest data on the chart indicates that whilst Flu admissions are rapidly falling there is still a moderate risk from Flu in the community compared to previous years.

While MEM thresholds can be set for Flu and RSV due to the availability of historical and consistent data, this is not yet possible for Covid because of the recency of the virus and a lack of a defined seasonal pattern. However, UKHSA have established experimental thresholds for Covid using a mean standard deviation (MSD) method.

The following chart shows the test positivity rate for Covid from July 2022 to February 2025 mapped against the experimental threshold levels.

The chart shows the many waves of Covid activity and how they break through the threshold levels. At present the test positivity rate is well below the baseline threshold of activity and, consequently, the relative risk from Covid when compared to past levels is very low.

In summary, UKHSA use activity thresholds derived from historic data to assess the relative risk posed by respiratory viruses by monitoring key metrics, like hospital admissions and test positivity, against these thresholds.

However, these metrics do not directly measure the prevalence of the virus in the wider community so the question remains how good are they as a proxy for community prevalence? The final section in this article aims to answer that question.

What do the surveillance metrics tell us about prevalence?

As previously mentioned, community prevalence was measured in the UK by the Coronavirus (COVID-19) Infection Survey (CIS) and the 2023/2024 Winter Infection Survey (WIS). Although this data is no longer available it does provide the opportunity to compare the trends for two of the key surveillance metrics with community prevalence.

The following panel chart compares the test positivity rate for Covid, depicted in blue, with community prevalence for Covid measured by the Winter Infection Survey, shown in red.

The charts show that the overall trend for the two measures follow the same broad pattern indicating that test positivity is a good proxy for the trend in community prevalence. The two measures start rising at the same time and peak within a day of each other. However, the fall in test positivity is slower than the fall in prevalence and this makes it difficult to extrapolate community prevalence directly from the test positivity rate.

The following chart compares hospital admissions for Covid, depicted in green, with community prevalence for Covid measured by the Winter Infection Survey, shown in red.

Although hospital admissions follow the same broad pattern as prevalence there are some differences. First, the peak for hospital admissions is delayed by about 10 days. However, this may be due to the Christmas / New Year break which can impact the timing of admissions. Secondly, hospital admissions show a third peak when prevalence is falling. Based on this data, hospital admissions is not such a good proxy for prevalence and should be seen as a lagging indicator.

In conclusion.

Although the data available to assess Covid levels in the UK is now more limited, the surveillance metrics published by the UKHSA provide a useful guide to Covid activity in the community. Furthermore, the threshold levels provide a good indication of the level of risk.

In addition, test positivity rates are a better proxy for community prevalence than hospital admissions, which are more of a lagging indicator.

As always, if you have any comments on this Covid Situation Report or suggestions for topics to cover, please post a message below.