Excess deaths

Deaths at Home Rise by a Third as Patients Stay Away From Hospital

There were more deaths from all causes in homes in each month of 2020 than in a normal year, according to new data from the Office for National Statistics (ONS), as treatment waiting lists and the “protect the NHS” drive kept patients away from hospitals. The Times has the story.

A total of 166,576 deaths in private homes from all causes were registered in 2020, compared with an average of 125,255 between 2015 and 2019, according to the ONS.

This means there were 41,321 extra deaths, or “excess deaths”, in private homes during the year, although Covid was responsible for 8% of the total.

The majority of deaths in 2020 where coronavirus was the main cause occurred in hospitals and care homes. In contrast, many deaths from other causes, such as breast cancer and prostate cancer, happened in private homes, to people who in a non-pandemic year would probably have died elsewhere, such as in hospital.

The figures show that deaths from diabetes in private homes were 60% higher in 2020 compared with the average for 2015-19, while those from heart disease and Parkinson’s disease were both up 66%.

For dementia and Alzheimer’s disease, deaths were up 65%, with increases of 44% and 37% for prostate cancer and breast cancer respectively…

The rise in deaths in private homes comes amid concern that during the pandemic people have not been receiving the medical attention for serious illnesses they would have in normal circumstances. In the early stages of the pandemic GPs and hospitals reported a drop in the number of patients and urged people to contact their doctor if they had a health issue.

Worth reading in full.

England and Wales See Seventh Consecutive Week of Negative Excess Deaths

The ONS announced today that there were 9,941 deaths in England and Wales in the week ending April 23rd, which is 497 fewer than the previous week. In addition, this week’s number is 5% below the five-year average, and marks the seventh consecutive week of “negative excess deaths”. Here’s the chart from the ONS:

Over the last seven weeks of ONS reports, there were 5,511 fewer deaths than you’d expect based on the average of the last five years. And recall that, because the population is ageing, the five-year average slightly understates the expected number of deaths. So the true level of “negative excess mortality” is even higher.

The number of deaths registered in the week ending April 23rd was below the five-year average in eight out of nine English regions. (Only London saw positive excess deaths.) Compared to the five-year average, weekly deaths were 6.8% lower in Wales, and 8.1% lower in the South East.

At the beginning of April, David Spiegelhalter and Anthony Masters wrote a piece suggesting a number of possible reasons for the low number of deaths in England and Wales: mild weather; fewer road accidents and flu deaths due to lockdown; and deaths having been “brought forward” by the pandemic.

Given that we are no longer in winter or the flu season, and there has been an increase in mobility since March, it seems unlikely that the first three factors they mentioned can account for more than a small share of the “negative excess deaths” observed in April. Rather, this phenomenon is probably explained by deaths having been “brought forward” by the pandemic.

Geography, Not Lockdowns, Explains the Global Pattern of Excess Mortality

Throughout the pandemic, commentators have relied on ‘COVID-19 deaths per million people’ as a measure of the disease’s lethality. While this is not unreasonable for making comparisons within Europe, it is less justifiable for other parts of the world, where there has been substantial underreporting of COVID-19 deaths.

A better measure to use is excess mortality, i.e., the number of deaths in excess of what you’d expect based on previous years. This measure does not vary with factors like testing infrastructure or the criteria for assigning cause of death. (Though the best measure is age-adjusted excess mortality.)

In an unpublished study, the researchers Ariel Karlinsky and Dmitry Kobak have compiled all the available data on excess mortality. Their database – which they’ve termed the ‘World Mortality Dataset’ – allows us to make more meaningful comparisons across countries. 

Unfortunately, data are not yet available for most of the countries in Sub-Saharan Africa, Asia and the Middle East. However, the database does encompass Europe, much of Central Asia and the Americas, and parts of East Asia. Countries with available data are shown in blue in the map below:

Note that Karlinsky and Kobak did not take the usual approach of using the average of the last five years as the baseline. Rather, they took the superior approach of using a linear trend over the last five years. If deaths have been increasing year-on-year, their approach will yield a higher baseline, compared to using the five-year average. (See here for a visual explanation.)

Deaths In England and Wales Now 5% Below the Five-Year Average

The ONS announced today that there were 10,045 deaths registered in England and Wales in the week ending March 26th. This is 266 fewer than the previous week (which was the lowest since 2014 for that week). In addition, this week’s number is 5% below the five-year average, and marks the third consecutive week of “negative excess deaths”. Here’s the chart from the ONS:

Over the last three weeks of ONS reports, there were 1,800 fewer deaths than you’d expect based on the average of the last five years. (And note that, because the population is ageing, the five-year average slightly understates the expected number of deaths.)

What’s more, the number of deaths registered in the week ending March 26th was below the five-year average in seven out of nine English regions. (Only the East Midlands and West Midlands saw positive excess deaths.) Compared to the five-year average, weekly deaths were 7.5% lower in London, 9.3% lower in the South West, and a remarkable 10.7% lower in the East of England.

As I’ve noted before, there are several possible reasons why the number of deaths is so low at the moment. But whatever the exact reason, or mix of reasons, today’s numbers are surely cause for optimism.

It’s Time to Retire the ‘Deaths Within 28 Days of a Positive Test’ Chart

The BBC and other media outlets continue to post the chart showing the number of deaths within 28 days of a positive test. And this is the first chart you see after clicking ‘Deaths’ on the government’s COVID-19 dashboard. Here it is below:

However, the chart gives a very misleading impression of the relative severity of the first and second waves. It has been claimed, for example, that the second wave was “more deadly” than the first. And in fact, if you sum the figures under each of the two curves in the chart above, you find that the total for the second wave is more than double the total for the first wave.

But simply counting the number of deaths within 28 days of a positive test isn’t the best way to gauge COVID-19’s impact on mortality. This method understates the number of deaths in the first wave because some people died of COVID-19 without being tested. And it overstates the number of deaths in the second wave because some people who would have died anyway happened to test positive.

As Sarah Caul (the Head of Mortality Analysis at the ONS) has noted, “For the best comparisons, we really need to look at age-standardised mortality rates.” These take into account the ages of those who died, as well as the age-structure of the overall population. The ONS recently calculated weekly age-standardised mortality rates going back to 2015. They then calculated excess mortality for 2020 and 2021 by taking the age-standardised mortality rate in each week, subtracting the average over the last five years, and then expressing the difference as a percentage of that average. Here’s the chart they produced:

Although the figures only go up to February 12th, they indicate that the second wave was actually less deadly than the first. It’s time to retire the ‘deaths within 28 days of a positive test’ chart. At the very least, it should only be posted alongside the chart showing age-standardised excess mortality – which gives a much more accurate picture of the UK’s pandemic.

Covid Involved in Fewer Than 10% of Deaths in England and Wales by Mid-March, ONS Says

Covid was involved in fewer than 10% of all deaths in England and Wales by mid-March, according to the latest figures from the ONS. New data also shows that the number of excess deaths has fallen below zero for the second week in a row. Here are the key findings from the ONS.

The number of death registrations in England and Wales involving the coronavirus decreased from 1,501 in Week 10 to 963 in Week 11 – a 35.8% decrease. Of all deaths registered in Week 11, 9.3% mentioned Covid on the death certificate.

In England, the number of deaths involving Covid in Week 11 was 912, accounting for 9.4% of all deaths compared with 13.9% in Week 10.

In Wales, there were 49 deaths involving Covid in Week 11, accounting for 7.9% of all deaths compared with 9.9% in Week 10.

And on excess deaths:

The provisional number of deaths registered in England and Wales decreased from 10,987 in Week 10 (week ending March 12th 2021) to 10,311 in Week 11 (week ending March 19th 2021). The number of deaths was 8.0% below the five-year average (894 deaths fewer).

In England, the number of deaths decreased from 10,277 in Week 10 to 9,673 in Week 11, which was 774 deaths (7.4%) fewer than the Week 11 five-year average. This is the second consecutive week that deaths have been lower than the five-year average in England.

In Wales, the number of deaths decreased from 685 in Week 10 to 621 in Week 11, which was 106 deaths (14.6%) fewer than the Week 11 five-year average. This is the third consecutive week that deaths have been lower than the five-year average in Wales.

Deaths in private homes continue to run significantly above average, as they have since the first lockdown, with 715 excess deaths or 27.7% above the five-year average in the week ending March 19th. Many of these deaths may be avoidable as they reflect people not accessing medical care.

Worth reading in full.

Excess Deaths Fall Below Zero

The number of excess deaths in England and Wales has fallen below zero for the first time in six months, according to ONS data.

The ONS reports:

The provisional number of deaths registered in England and Wales decreased from 11,592 in Week 9 (week ending March 5th 2021) to 10,987 in Week 10 (week ending March 12th 2021). The number of deaths was 4.4% below the five-year average (511 deaths fewer). 

In England, the number of deaths decreased from 10,882 in Week 9 to 10,277 in Week 10, which was 468 deaths (4.4%) fewer than the Week 10 five-year average. This is the first time that deaths have been lower than the five-year average in England since the week ending September 4th 2020 (Week 36).

In Wales, the number of deaths decreased from 689 in Week 9 to 685 in Week 10, which was 35 deaths (4.9%) fewer than the Week 10 five-year average. This is the second consecutive week deaths have been lower than the five-year average in Wales.

Given this data, and the fact that half of all UK adults have received a Covid vaccine, is it not about time that Britain unlocks?

The ONS’ findings are worth reading in full.