Official Statistics

Winter Coronavirus (COVID-19) Infection Study: publications explained

Updated 14 March 2024

Applies to England and Scotland

Summary 

The Office for National Statistics (ONS) and UK Health Security Agency (UKHSA) have jointly launched the Winter Coronavirus (COVID-19) Infection Study (Winter CIS) and will be producing separate publications to share the results. 

The Winter CIS, which started on 14 November 2023, will run until March 2024. It will provide vital information on the epidemiology of SARS-CoV-2, the virus that causes COVID-19. This will help us understand the potential winter pressures on the healthcare services due to the virus.

ONS Winter CIS publication

The ONS will release a set of data tables reporting both the unweighted and weighted test positivity rate over time, including breakdowns by region, age and sex. These results reflect the actual data collected via the survey and are published to ensure transparency around the data collected.

The unweighted test positivity rate refers to the proportion of the sample population in England and Scotland that tested positive for SARS-CoV-2 using a lateral flow device (LFD) in a given week, with or without symptoms. These analyses have not been adjusted for demographic biases in the survey sample, or for the diagnostic performance of the LFD tests.

Weighted test positivity was produced by the ONS to ensure the results are more representative of the target population, as differential response rates among different demographic groups can lead to underrepresentation in the sample. The methods used by the ONS are not the same as those applied by UKHSA. This is also different from the proportion of the population that are infected with the virus, as tests are imperfect. Therefore, weighted test positivity differs to the weighted prevalence produced by UKHSA in that positivity does not adjust the results for the performance of the test used. Prevalence estimates provide a better estimate of the proportion of the population who are infected with SARS-CoV-2.

The ONS data table publication will also include proportions of participants reporting COVID-19 and respiratory related symptoms, as well as impacts on daily activities and health services due to a respiratory infection or personal health reason.

For more information on the weighting methodology, refer to the ONS Quality Methodology Information report released on 18 January 2024.

UKHSA Winter CIS publication

UKHSA will release a report setting out the main findings together with a set of data tables. The content of the report and data tables will build over time, initially reporting prevalence and adding further metrics as more data becomes available. The metrics that they will present include: 

• prevalence, which measures the estimated proportion of the population who are infected with SARS-CoV-2 at a given point in time
• incidence rate, which measures the estimated number of new infections occurring on a given day per 100,000 individuals
• the infection hospitalisation risk (IHR) which measures the risk of hospital admission given that individuals have been infected with the SARS-CoV-2 virus, which is important for healthcare planning
• the infection fatality risk (IFR) which measures the risk of death given that individuals have been infected with the SARS-CoV-2 virus
• how effectively vaccinations protect people from health outcomes such as infection, symptomatic disease, hospitalisation, and mortality, which is known as vaccine effectiveness

A full quality and methodology information (QMI) report for the study, detailing the strengths and limitations of the data, methods used, and data uses and users, was released by UKHSA on 21 December 2023.

How these statistics can be used 

The statistics published by the ONS and UKHSA will use the same underlying data to develop different metrics. 

For the initial publications, the ONS will report the unweighted test positivity rate and UKHSA will report weighted prevalence. Prevalence is calculated by adjusting the positivity estimate for the test sensitivity and specificity to provide a closer reflection of the proportion of people who are actually infected by SARS-CoV-2 and reweighted to make the estimate representative of the wider population.

In later releases ONS positivity estimates will also be weighted to correct for bias in the sample.

The prevalence and incidence estimates provided by UKHSA’s Winter CIS publication provide insight into how much SARS-CoV-2 is circulating in the community and how quickly it is circulating, respectively. The IHR measures the risk of hospital admission given that individuals have been infected with the SARS-CoV-2 virus, which is important for healthcare planning. Incidence and the IHR are planned to be phased in to reports over time.  

The insights from both the ONS and UKHSA publications aim to allow users to understand the current trends in SARS-CoV-2 infections and make informed health decisions about their behaviour.

Publication dates 

The first ONS data release on test positivity was on 7 December 2023, followed by a publication on 21 December 2023 alongside the initial UKHSA data release on prevalence. 

ONS data tables will be published approximately every 2 weeks and ONS and UKHSA will coordinate publications so that they are published at the same time. These have been pre-announced on the ONS and UKHSA statistics release calendars.  

Incidence requires more detailed data on diagnostic performance of LFD tests to produce reliable estimates, and therefore UKHSA will add this metric to the publication when this data becomes available (expected early 2024).

The IHR requires data on hospitalisations from COVID-19 for England and Scotland. The admission data used to inform the IHR is made publicly available by the NHS. This means that UKHSA reporting timelines of the IHR will be determined by NHS reporting dates. There is a time delay between infection and attending hospital, due to the progression of the disease. Therefore, it is necessary to wait until sufficient time has passed before it is possible to calculate the IHR. UKHSA will add this metric to the publication when this data becomes available (expected early 2024).

Differences between the Winter CIS and previous CIS 

The Winter CIS is a new study and is not the same as the original Coronavirus (COVID-19) Infection Survey (CIS).  

The CIS ran from April 2020 to March 2023. It produced temporal estimates of weighted SARS-CoV-2 positivity from its cohort. Participants from England, Northern Ireland, Scotland, and Wales answered questionnaires and were tested using polymerase chain reaction (PCR) tests. For some participants, a blood sample was taken.

Data for the Winter CIS is collected using self-reported LFDs and self-completed online questionnaires. A follow-up questionnaire will be sent to the participants if a positive test result for SARS-CoV-2 has been declared.

LFDs have a lower average test sensitivity than the PCR tests used in the previous CIS. Test sensitivity is the probability that a person infected with SARS-CoV-2 will test positive. Those who test negative despite being infected are termed false negatives. Test sensitivity depends on 2 things. First, it depends on how much virus is circulating within the infected person, also known as the viral load. Viral load increases after infection and declines as the infection is cleared. Second, different types of tests are more or less likely to pick up low viral loads. Lower test sensitivity leads to a higher false negative rate. 

Tests specificity is the probability that someone who is not infected with SARS-CoV-2 tests negative for the virus. Someone who was not infected but tested positive would be called a false positive. For LFD tests specificity is very high. Higher test specificity leads to a lower false positive rate.

To weight the data, UKHSA are taking a multi-level regression and poststratification (MRP) approach. Changes in prevalence through time were modelled for each age group, region and sex simultaneously, accounting for LFD test sensitivity and specificity. A daily modelled approach is more easily able to show changes in trends (compared to the ONS weekly weighted positivity where data is reported for the week as a whole). As the study progresses and more data is collected the estimated prevalence will be updated.

As survey participants are not representative of the population, UKHSA re-weights these estimates according to the demographic profile of the general population in England and Scotland. This means estimates of prevalence are less biased towards groups with higher response rates in the survey. 

Confidence intervals and credible intervals are used to quantify the uncertainty in estimated the statistics. Different statistical approaches require different descriptions of uncertainty.  

ONS will report 95% binomial confidence intervals on positivity. These intervals will be larger when sample sizes are smaller.  

UKHSA will report 95% credible intervals on estimated prevalence. The 95% credible intervals reflect uncertainty in the estimated parameters of the MRP model. They account for variations in the survey data and the adjustments made for sensitivity and specificity, providing a probabilistic range of values for the estimated prevalence.

Winter CIS participant selection 

The Winter CIS will collect data from around 140,000 individuals living in England and Scotland only. Each participant is asked to test once each month, during a scheduled 7-day testing period.

While the devolved administrations of Wales and Northern Ireland will not be taking part, we will continue to work closely with them on disease surveillance across communities in all parts of the UK.   

All those sampled for the Winter CIS are participants who took part in the previous CIS and who had agreed to be approached about other ethically approved research studies. 

This includes adults aged 16 years and over who are willing and able to consent, as well as children aged between 3 years and 15 years, for whom a parent or carer is willing and able to consent to their participation.  

Official statistics in development  

The statistics produced by the Winter CIS are official statistics in development. 

This means they are new official statistics that are undergoing a development and will be tested with users, in line with the standards of trustworthiness, quality and value in the Code of Practice for Statistics.  

Until September 2023, official statistics in development were called ‘experimental statistics’.  

Official statistics in development are developed under the guidance of the Head of Profession for Statistics (HoP). The goal is to develop statistics that can be produced to the standards of the Code of Practice. 

Further information 

If you have any further questions about the ONS Winter Coronavirus (COVID-19) Infection Study, contact Winter.Covid19.Infection.Study@ons.gov.uk.

If you have any further questions about the UKHSA Winter Coronavirus (COVID-19) Infection Study, contact sipmo@ukhsa.gov.uk.