THREAD – Mortality 2020

Today we published England & Wales mortality statistics for the w/e 25 Dec

Nearly a full year’s data for 2020

In the last 52 weeks there were:
c.604k deaths registered across E&W

This is:
c.73k deaths (14%) above the 5-yr avg = excess deaths

1/11

Remember the bank holiday effect which affects weekly figs

And the lag between deaths occurring and being registered

We now provide modelled estimates of death occurrences to remove these

The chart below shows both registrations and modelled occurrences across 2020

2/11

Looking at excess deaths, we began the year with death levels below the 5-yr avg

This was followed by a huge peak in the spring, driven by COVID, until lockdown1 took effect

Note the small Aug heatwave peak

And then a gradual pick up through autumn as CV took hold again

3/11

A seasonal effect means more deaths in winter than summer

The pattern of excess deaths as a % of normal weekly deaths is similar

There is a noticeable jump in the latest w/e 25 Dec

Given the recent alarming rise in CV infections this will increase further in Jan

4/11

For longer term context we can compare annual deaths back to 1900

Deaths have been rising recently as life expectancy improvements have stalled, but the jump this year to over 600k is clear

Only one previous year has topped 600k – 1918, the year of the “Spanish” flu

5/11

Numbers of deaths are affected by population size

We can also consider death rates – as a proportion of the population

“Crude” death rates are shown below

All improvements in the first decade of this century will be reversed this year

“Spanish” flu did the same in 1918

6/11

Numbers of deaths are also affected by population structure, such as age. We will produce age standardised rates soon

Whilst the UK-born population is ageing, with post-war baby boomers now in their 50’s to 70’s, recent immigration will have predominantly been younger

7/11

Our world changes too

Medical advances, the NHS, welfare state, better living conditions, all mean that 14% of babies do not die before their 1st birthday. They did in 1900

Excess deaths vs the last 5 years gives current context

This year will see the most since 1940

8/11

We can adjust by population to get crude excess death rates

2020 will also be the highest since 1940

Even with measures taken to limit COVID spread, 2020 will still top 1951, the year of a major flu epidemic

Without all our efforts, 2020 could have been much worse

9/11

A week is a long time in a pandemic

But the basic COVID logic remains the same

Thank you to everyone caring for the sick in such difficult circumstances, you are heroes

My deepest condolences to all who have lost loved ones this year

10/11 https://twitter.com/NickStripe_ONS/status/1344033059432255489

Thread on today’s figures from the main ONS account below

Best wishes to all suffering with COVID today

And best wishes for an eventual recovery to all of you with Long COVID – I know how you feel

Stay safe and keep others safe

11/11 - ENDS https://twitter.com/ONS/status/1346751126637604864

UPDATE – Mortality 2020

Yesterday we published the last week of Eng & Wales death registration stats for 2020

There were c.608k deaths registered across the full calendar year (1 Jan - 31 Dec)

My colleague Sarah’s blog puts these figures into context https://twitter.com/ONSfocus/status/1348946050875355136

The blog also gives the provisional age-standardised mortality rate (ASMR) for 2020

= 1043.5 per 100k

This was c.13% higher than the ASMR for 2019

And c.8% higher than the recent 5-yr average ASMR

We have annual ASMRs published back to 1994 – charted here

We can therefore look at annual change against the 5-yr average ASMR back to 1999, as in the chart shown here

The long-run trend is one of improvements driven by better healthcare, public health, the way we live, less risky occupations etc

Although we do not have ASMRs back further than 1994, @EdConwaySky’s thread on Monday had calculations provided by @ActuaryByDay and colleagues

Note the next tweet in his thread shows year-on-year change, not vs 5-yr avg https://twitter.com/EdConwaySky/status/1348708802716196864

All our data for 2020 is still provisional, subject to final cross-checks and QA

Comprehensive annual analysis will be provided later this year

Next Tuesday will see the first release of weekly data from 2021 registrations

Please stay safe

ENDS