Changing measures again

The England count for daily COVID-19 deaths changes again.

Public Health England published their review of the daily COVID-19 death measure. That review changed the measure to:

Deaths are only included if the deceased had had a positive test for COVID-19 and died within 28 days of the first positive test.

This article looks at the consequences of that change.

The total count reduces by over 5,000

Public Health England’s original measure had no time limit.

Someone could test positive for SARS-CoV-2 in March. After recovery, they die of unrelated causes in July. Since that person had a prior positive test, their unrelated death would count as a ‘COVID-19 death’. Public Health agencies in the rest of the UK would not count that as a ‘COVID-19 death’.

After their review, Public Health England introduced two new measures:

  • Death within 28 days: the death occurs within 28 days of the first lab test confirming SARS-CoV-2.
  • Death within 60 days, or COVID-19 on death certificate: the death is within 60 days of the first positive test. After 60 days, a death counts if the death certificate mentions COVID-19.

On 12th August 2020, the measures stood at:

  • No time limit: 42,072
  • Death within 60 days, or COVID-19 on death certificate: 40,404
  • Death within 28 days: 36,695
Image for post
Image for post
The difference became very large in recent days. (Image: Public Health England)

The 28-day measure is on the daily report. The European Centre for Disease Prevention and Control collects that UK measure. Our World in Data uses ECDC as its data source.

Applying the 28-day time limit means a reduction of 5,377 deaths (12.8%).

Over time, the 28-day measure fell by more than the count without a time limit. According to the daily report, the new England measure has been below 50 since 9th July. This is by date of report, not date of death.

Aligned with Scotland and Northern Ireland

The daily measure in England now aligns to Scotland and Northern Ireland.

There were differences in definitions between the four nations. Public health agencies in Scotland and Northern Ireland shared their definition:

  • Scotland (Public Health Scotland): confirmed deaths in all places. The positive test result must be at most 28 days before the death.
  • Northern Ireland (Public Health Agency): confirmed deaths in all places. The person must have died within 28 days of their first positive test result.

Public Health Agency count death notifications to the agency. Public Health Scotland use the National Records of Scotland daily file.

People may not have understood these differences in definitions. Social media users highlight how different reported deaths in the nations were:

On that reporting date, all counted deaths were in England.

The population in England is about 10 times larger than that in Scotland. Before this change, ‘COVID-19 deaths’ meant something different in England and Scotland.

Public Health Wales maintains a different definition

The GOV.UK page for daily Coronavirus statistics states:

The new daily measure provides a UK-wide count of deaths under a consistent methodology for England, Northern Ireland, Scotland and Wales that has been endorsed by an external review.

This statement does not appear to hold.

There is not full alignment across the four nations. Public Health Wales has a different definition, with three criteria:

  1. Settings: Deaths of hospitalised patients in Welsh hospitals or care homes.
  2. Confirmed case: A lab-confirmed case of COVID-19 with a positive result.
  3. Suspicion of cause: Clinicians suspect COVID-19 was a causative factor.

There is no 28-day cut-off in this definition.

The notes in the daily dashboard recognise this difference:

The majority of deaths included occur within 28 days of a positive test result.

For the rest of the UK: this count is of deaths occurring within a set time of the lab result. This measure is of deaths with the virus. It does not mean the disease contributed to every counted death.

For deaths in 28 days of the positive specimen, 95% of certificates mention COVID-19. That figure includes samples drawn after death, and excludes deaths without a certificate.

Image for post
Image for post
That figure falls to 73% for 29–42 days. (Image: Public Health England)

Death certificates mention diseases as believed causes or contributory factors.

The three statistical offices (ONS, NRS, NISRA) publish death registration statistics. Their measures are of death certificates which mention COVID-19. Those stats have a common definition and coverage across the four nations.

Excess deaths are deaths from all causes above a calculated baseline. Another means is to compare changes in all-cause mortality. Interpretation is more difficult. Excess deaths can be due to COVID-19, disrupted healthcare, or other reasons.

The National Institute for Health Protection will replace Public Health England. Many thanks to everyone at Public Health England for their work during this pandemic.

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