COVID estimates revised after change to methodology

May 12, 2021

What’s changed?

We’ve adjusted our figures to account for the high number of vaccinated contributors using the ZOE COVID Study. As a result incidence rates have been revised to be higher. The numbers remain low and this is no cause for alarm. 

We’re constantly looking at how our contributors reflect the UK population, and in recent weeks the number of vaccinated ZOE COVID Study contributors started to outpace the rate of vaccination in the wider population. After investigating how this impacts our numbers, we decided to adjust our data to take into consideration the rate of vaccination in the UK, not just our study.

Why is ZOE updating incidence reports?

We are making these changes because:

  • The current low prevalence of the virus, in part due to the success of the vaccine roll out, has meant regular reviews of methodologies are needed to ensure the accuracy of data.
  • There’s a disproportionately high number of vaccinated ZOE COVID Study contributors compared to the general population

The pandemic is changing, and as a result, we need to make sure our data takes this into account. We’re committed to reviewing our methodology as part of our quality assurance process, this ensures the ZOE data remains as accurate as possible.

COVID is changing

Due to the vaccine roll out many peoples’ COVID-19 symptoms are now less severe than in 2020 and there are significantly fewer cases of serious respiratory responses.

Although your risk of contracting coronavirus is drastically reduced after even one dose of vaccine, there’s still a small chance that you could catch it, but again with much milder symptoms.

The rates of infection have reached some of the lowest levels ever, comparable to summer 2020. As a result of this our confidence intervals have become very big meaning it’s hard to give exact numbers. 

The percentage of adults in the UK who have received at least one dose of a COVID vaccination, as of 11th May 2021, is 67.3%. While this is a brilliant achievement, the number of contributors to the ZOE Symptom Study who have reported at least one vaccination as of 8th May 2021 is even higher at  89%.

Since our contributors have been disproportionately vaccinated against the virus, we made the decision to adjust our figures to account for the difference in infections that we expect to see in the wider UK population.

You can see the huge increase in the ZOE app contributors who have received a vaccination, with higher rates than the UK population percentage.


The unvaccinated group is more likely to test positive for COVID-19 than those who have received at least one dose of the vaccine.

How is ZOE updating incidence reports?

Here’s how we’ve been calculating COVID incidence estimates up until now:

  1. We calculated positivity rates on ZOE COVID Study contributors based on the self-reported PCR test results (we currently don’t include lateral flow test results), as well as newly symptomatic rates from the health reports logged in the ZOE app.
  2. The rates were then extrapolated to the UK population to produce a figure for daily incidence rates (the number of new cases reported each day).

However, because of the higher percentage of vaccinated contributors in the ZOE COVID Study, we’ve updated our methods to include an extra step to adjust for the fact that our contributors have outpaced the rate of vaccination of the wider population:

  1. We split the tests and symptoms data reported through the app between those who have had at least one vaccine and those who are unvaccinated.
  2. We calculate positivity rates and newly symptomatic rates on these two groups based on the self-reported PCR test results (still excluding lateral flow) 
  3. We now can extrapolate this data to the UK population, including the government-reported rates of vaccination, to come up with a figure for daily incidence rates (the number of new cases reported each day).

By splitting positivity rates in these two groups of vaccinated and unvaccinated contributors, we can calculate an incidence number that reflects the rate of vaccination in the wider UK population.

Since we compute our prevalence estimates (how many people have COVID-19 today) from these incidence reports (how many new COVID infections today), both estimates will be impacted by this change.

For example, our incidence reports (using our previous methods) on the 12th of May 2021, show the total number of new cases to be 1,615 as of the 8th of May. However, after applying our adjustments based on the percentage of the UK population vaccinated, this number will now be reported as 2,782.

You can see the change in our estimates over the last few months below with previous methods (orange line) and our updated methods (blue line)



What does this mean for me?

While it might be surprising to see the rates increase, we want to reassure you that these numbers are still incredibly low, and it should not be a cause for concern. It’s essential for us to remain flexible and adapt to changes to make sure we’re as accurate as possible in our daily reports.

This is not the first time we have had to adapt to reflect the rapidly changing COVID situation in the UK, and you can be reassured that we’re committed to reviewing our methodology as part of our quality assurance process.

Your daily contributions mean a lot for science, so please keep logging your vaccines, symptoms and tests, and encourage any unvaccinated people you know to download the app and log their symptoms and updates too. That way we can continue to provide you with the most accurate data every day.

Stay safe and keep logging!

Text Link

Other updates

Sign up to get the latest research first, direct to your inbox.

Close

Get in touch

Please complete the reCAPTCHA and try again

Thank you

Your submission has been received
Oops! Something went wrong while submitting the form.