Death counts with COVID-19

DHSC and ONS death counts differ. Why?

The Department for Health and Social Care publish daily updates of COVID-19 deaths. The Office for National Statistics publish weekly registered deaths. These weekly numbers will now show deaths involving COVID-19.

These data series show different figures. This article explains that difference.

Update: I published this article on 4th April. Since that date, the DHSC changed their measure. The article is correct as of 1st June 2020.

Where do the figures come from?

DHSC publish daily updates to confirmed COVID-19 deaths. This is the number of patients who tested positive for COVID-19 and later died. The positive test must be in a Public Health, NHS or allied commercial laboratory. Public Health England also publish these figures.

The department compiles validated figures from each country. In England, the figures are from NHS England and Improvement. Public Health Wales provides counts for Wales.

Hospitals provide data, which NHS England extracts data at 5pm each day. Also, PHE Public Health Protection teams report notified deaths. Before 29th April, the England figures only included deaths in NHS-commissioned services.

On Saturdays, DHSC reports new recorded deaths between Thursday 5pm and Friday 5pm. That time window applies to England and Wales. For Scotland, the cut-off time is 9am on the day of publication. For Northern Ireland, that time is 9:15am.

These statistics exclude deaths of people with COVID-19:

  • who were not tested;
  • who tested negative, and later caught COVID-19.

After 1st June, death counts include positive tests with commercial partners. This change affected statistics from 24th May on-wards.

There is a delay between death and registration. (Image: Getty Images/BBC)

The Office for National Statistics publishes death registrations. Their weekly COVID-19 figures are counts of death certificates which mention the virus.

This publication of COVID-19 deaths is on every Tuesday, starting on 31st March. There is usually a delay of around five days between death and registration.

Why are they different?

There are major differences: covered countries, included deaths, and time.

Covered countries

  • DHSC: The figures are for England, Wales, Scotland, and Northern Ireland.
  • ONS: Registration numbers are in England and Wales.

Included deaths

  • DHSC: The patient must have tested positive for COVID-19. In Wales, there are two extra criteria. The death is in a Welsh hospital or care home. Also, a clinician must suspect COVID-19 was a causative factor in the death.
  • ONS: Registrations include any place of death, such as care homes. The death certificates must mention COVID-19. A certifying doctor could suspect a person had the virus, without a positive test result.


  • DHSC: The counts are daily, and include deaths without formal registrations. Reported deaths are after an official notifies the family.
  • ONS: Weekly registrations are 11 days behind, due to the time it takes to register, process, and publish. The ONS can publish figures by registration date and the actual date of death.

Which is ‘better’?

Two different data sources have different strengths and limitations. As Sarah Caul (ONS) highlights: ONS registrations follow a similar trend to the DHSC counts. There is a short time-lag.

ONS deaths by actual date of death are above daily DHSC death counts. (Image: National Statistical)

The DHSC figures provide an up-to-date view of the pandemic. These numbers are not how many confirmed cases die on a particular day. There can be a short delay between deaths happening and records entering NHS systems. In some cases, that delay is “a few days”.

The ONS figures provide a more complete view of deaths involving COVID-19. That completeness comes at the price of timeliness. Registrations take time to process and publish. Weekly registrations are 11 days behind.

In both cases, these counts are deaths involving the virus. COVID-19 does not need to cause each death.

We should thank all analysts working in statistical services. Providing quality and valuable statistics is difficult. During a global pandemic, it is a monumental challenge. There have been improvements to the quality and clarity of COVID-19 data in recent weeks.

As an RSS Statistical Ambassador, I helped write a short guide to COVID-19 statistics.

This blog looks at the use of statistics in Britain and beyond. It is written by RSS Statistical Ambassador and Chartered Statistician @anthonybmasters.

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store