Hospital admissions with Covid in England increased by 33% in the last week of April, having been low and stable for a month.
Most regions saw significant in-week increases. The fasted increases were in NE+Yorks (+52%), Midlands (+39%) and South East (+38%).
There’s been a bit of an unpleasant and inaccurate narrative over recent days that the majority of COVID-19 deaths were people who were “at death’s door” in any case. That they died “with” rather than “of” the disease. I want to address that. 1/8
“Do you actually know how many people are dying of other stuff” asks Isabel, condescendingly.
“I’m an actuary” replies Trevor, politely.
Marvellous! 👏🏻
Ok, gather round, it’s time for todays statistics lesson. This week we are covering Simpson’s Paradox. I promise it won’t take long.
Simpson's Paradox when a relationship appears within individual groups of data but disappears or reverses when the groups are combined.
On
@BBCMoreOrLess
this morning I was delighted to have the opportunity to debunk the most ludicrous claim I’ve yet been asked to address!
John Campbell shared a video with his 2.9 million followers with claims of over 150,000 excess deaths compared to if no-one was vaccinated.
This rhetoric is divisive and unhelpful, particularly the headline.
It’s correct to say we are experiencing significant excess deaths. It’s also correct that much of the excess didn’t arise directly from COVID.
But I disagree with calling these deaths “lockdown effects”. 🧵
ONS deaths data has been released for week ending 23 Dec.
2,982 more deaths were recorded in-week compared to the 2015-19 average. That’s 26% more, the largest excess we’ve seen all year.
Year-to-date there have been 567,379 deaths recorded which is 8% more than 2015-19 avg.
Our new commentary published in
@LancetRH_Europe
discusses the ongoing excess deaths in the UK, and for the first time provides granular analysis by age, place and cause.
Excess deaths now highest for middle-aged adults, driven by cardiovascular causes.
Hopefully this goes some way to addressing the false narrative that these people would have died soon anyway, and we’ll hear rather less of it from journalists and commentators who ought to know better. 8/8
An important new report from my
@LCP_Actuaries
colleagues shows that most of the increase in the number of economically inactive people during the pandemic can be explained by long term sickness.
This runs counter to the “great retirement” narrative.
Important as this is, a reminder that is has no effect whatsoever on excess deaths, which is the definitive way to measure the impact of the pandemic. In fact it’s just this sort of recording issue that is why excess deaths are the gold-standard.
BREAKING ....
Matt Hancock calling "urgent review" into a staggering fiasco at Public Health England.
They are counting people who die of natural causes as coronavirus deaths if they tested positive at any time previously.
Not clear how many deaths were recorded in this way
Fine. Those of us analysing death data have been ignoring daily updates for mnths. They significantly undercounted on the way up; have slightly overcounted on the way down.
ONS will continue to release reliable deaths data each week, from which we can determine excess deaths.
The first thing that jumps out from the medical history section of Table 1 is that 93% of those critically ill with COVID-19 were “able to live without assistance in daily activities” prior to developing the disease. That typically suggests reasonable health. 3/8
Updated odds of catching COVID-19:
Mar 1-in-4600
Apr 1-in-140
May 1-in-5
Jun 1-in-2
Jul 1-in-12
Aug 1-in-150
Assumes continue current trajectory, without radical interventions.
Hopefully this gives context to Govt comments about timing interventions right.
ONS deaths data has been released for week ending 13 Jan. 2,837 more deaths were recorded in-week compared to the 5-year average. That’s 20% more - another significant excess.
Year-to-date there have been 32,364 deaths recorded, 17% more than the 5-year average.
📢RCEM: ‘This is the start of another extremely difficult winter, and we don’t have the beds that we desperately need’
@RCEMpresident
Our response to today's A&E performance figures, which show some appalling trends.
Read our full statement here:
Secondly, just 7% of 2,124 intensive care admissions had “very severe comorbidities”. That’s significantly less than typical viral pneumonia patients (as illustrated in table 1)👆
So 19 out of 20 were free of the most severe life-limiting conditions. 4/8
Today’s Twitter nonsense includes:
- people trying to educate me about mortality forecasting
- people trying to educate me about the long-term effects of Spanish flu
And the pièce de résistance, someone telling me I need to read a paper in the Lancet which I co-authored.
It’s been a week since THAT thread using Critical Care data (from ICNARC) to challenge the narrative that those getting very seriously ill or dying from COVID-19 were already “at death’s door”.
There’s a new report out (and I’m a glutton for punishment) so let’s dive back in. /1
So what are the underlying causes? I think three main factors:
1. elevated cardiovascular risks after covid infection
2. current delays for urgent treatment in the NHS
3. missed and delayed diagnoses early in the pandemic
Lastly, for now, the age distribution. Most critical cases are aged between 50 and 80.
People of this age, without severe comorbidities, could reasonably have expected to have years of life ahead of them before they contracted this disease. 5/8
For ages 50-64, deaths involving cardiovascular diseases such as heart disease and stroke were 33% higher than expected.
Other causes with significant excess deaths at ages 50-64 were acute respiratory infections (43% excess) and diabetes (35% excess).
Hospital admissions with COVID in England increased by 27% in-week.
R-estimate incresed to 1.15.
Admissions were flat in the East (-2%) but rose sharply elsewhere. Largest increases in the South East (46%) and the South West (45%).
Acute bed occupancy is up by 26%.
Getting the train into London for the first time in a while. I’m one of the few people wearing a mask. Am trying to ignore the coughs and sniffles all around me. Why are so many people travelling when symptomatic?
I know it’s shameless self-promotion but I’m very proud to receive this professional recognition for my COVID work over the last 18 months.
I know I’m quieter of late (busy dayjob and we’re in for the long haul) but grateful to my
@COVID19actuary
colleagues for sharing the load.
The IFoA's President’s Award was presented to
@ActuaryByDay
for being the compelling actuarial voice in the fight against the COVID-19 crisis in 2020-2021. Read the full story here.
Updated chart of COVID-19 hospital admissions now including the newly published data from the last three days.
Admissions in England are rising sharply.
The 7-day moving average is now two and a half times the level it reached at its low point on 26 August.
A lot of people have asked for a
@COVID19actuary
view on this thread. I’ll give a personal reply for now, though others I’ve spoken to concur.
Big picture: broad agreement that the recent excess mortality is concerning and we need to better understand root causes. Detail below.
🧵on excess deaths. Concerning picture TLDR;
-excess deaths for 11 consecutive wks
->1000 excess deaths/wk
- ~500-800 excess deaths/wk at home
-excess in all ages including children & young people
-Almost all excess in hospitals explained by COVID
-What's causing home deaths?
Oh Boris!
There’s a lot wrong with this. For one thing he compared median to mean, but crucially he’s wrongly using life expectancy (LE) at birth.
LE at age 80 is 9yrs for men and 10yrs for women
LE at age 90 is 4yrs for men and 5yrs for women
ONS 👉
Analysis by
@LCP_Actuaries
for
@Independent
shows that the crisis in A&E can be linked to more than 15,000 deaths in the last six months.
Up to 500 patients a week are dying because of long waits for emergency care.
Hospital admissions with COVID in England increased by 19% in-week.
R-estimate is stable at 1.12.
Admissions rose in all English regions. Largest increases in the South West (61%) and NE&Yorks (35%).
Acute bed occupancy is up by 6%.
ONS deaths data has been released for week ending 16 June.
1,008 more deaths were recorded in-week compared to the 5-year average. That’s 10% more, which is a significant excess.
Year-to-date there have been 289,993 deaths recorded, 8% more than the 5-year average.
There‘s a danger of drawing conclusions from ICU admissions that I ought to bring out. Given capacity limits there will inevitably be some triage going on, ie admitting those with the best chance of survival. This will skew the data to some extent. 6/8
Are we seeing significant excess deaths in England and Wales at the moment? Yes, absolutely. Death rates have been around 11% above pre-pandemic over the last 8 weeks.
So why might skeptics suggest otherwise? If acting in good faith then they don’t fully understand the data 🧵
ONS deaths data has been released for week ending 24 Mar.
1,361 more deaths were recorded in-week compared to the 5-year average. That’s 13% more, a higher excess than seen in recent weeks.
Year-to-date there have been 160,016 deaths recorded, 7% more than the 5-year average.
Popping in to the twittersphere just briefly to say a HUGE thank you to everyone who has been in touch to congratulate me on the MBE. 🙏
I hope the excited kids and puppy will allow me a bit more time later to thank more people in person. 😊
Over 40s can now rebook earlier second jabs (from 8 weeks after first dose).
I just brought mine forward by 3 weeks (and moved it to a location I can walk to).
Via link below and “Manage my appointments”.
There are factual inaccuracies and misleading statements in this short thread - disappointing from a former Secretary of State
@DavidDavisMP
. Let’s quickly walkthrough them to set the record straight if possible.
In this video, Dr John Campbell provides an excellent summation of an FOI request answered by the ONS last month, disentangling the statistics of people who have died with Covid and those who died solely because of Covid. (1/4)
The actuarial profession is beginning to express a view on the extent to which the legacy of the pandemic will affect life expectancy.
The CMI is consulting on the next version of its mortality projection model. The proposal would reduce life expectancy of retirees by 6 months.
A significant proportion of the age-standardised excess are deaths with COVID.
However, a similar number are not. It is these non-covid deaths that
@sarahknapton
’s article focuses on, calling them “lockdown effects”. This attribution is speculative and I’d say it’s inaccurate.
Despite what we experienced in March a lot of people still don’t really grasp exponential growth. I’m getting challenged about why I’d talk about admissions doubling when average “has only gone from 50 to 100”.
If conditions don’t change the pattern goes 100, 200, 400, 800, ...
Hospital admissions with COVID in England increased by 10% in-week.
Our R-value has increased to 1.06.
Admissions rose in all regions except NE&Yorks (-6%). Largest increaes in the South West (+35%) and the East (+29%).
Acute bed occupancy is up by 4%.
Prevalence is estimated to be highest in the London area, at just over 6% across the population. Note though that confidence intervals are wider due to lower sample sizes than in previous studies.
3/
No sign of a slowdown in the rate at which hospitalisations with COVID are increasing.
Whatever precautions you chose to take in the January and March waves, you should consider taking now. The only things that’s substantially different is the media coverage.
The last update of hospital data for the week shows admissions growing at 41% per week, which if maintained would suggest a doubling time of approximately 14 days.
With the daily average now 962, a further doubling would take us close to the BA.2 peak shown.
1/
So what the chart above is really showing is that within the 10-59 age band the average unvaccinated person is much younger than the average vaccinated person, and therefore has a much lower death rate. Any benefit from vaccines is swamped by the increase in death rate with age.
Today CMI released its quarterly mortality monitor for Q1 2023.
The analysis of mortality by age is fascinating but grim reading. Death rates are relatively high for all ages but especially high for young adults. Death rates at age 20-44 are as bad as the second wave in 2021.
In the next week there will be a bulletin on this from one of the team at
@COVID19actuary
but for now some brief thoughts from me.
I’ll refer to yesterday’s update from the Intensive Care National Audit and Research Centre (ICNARC). 2/8
Full report:
Still, this data covers all admissions from the start of the epidemic until 3 April. For much of that period we are told that spare capacity was available so this data should not be skewed too much away from the typical critical case. 7/8
I’ve again updated my chart of COVID-19 hospital admissions to include the latest numbers.
Admissions in England are rising. The 7-day moving average has increased by 70% since its low point on 26 August.
This is the first sustained increase in admissions since March/April.
I have mocked up some illustrative numbers below to hopefully show this in action. I’ve split 10-59s into 10-29 and 30-59. Within each group the death rate for unvaccinated people is twice as high as for vaccinated people. But within the combined group it is lower!
Based upon google searches, it appears that "long covid" only exists in the anglosphere nations: Ireland, Australia, UK, Canada, New Zealand, and the USA.
Does covid somehow know if you speak English?
Why no interest in long covid in Japan or Sweden or Italy or Brazil?
Let’s go through the claims one by one.
1. ONS data clearly shows that death rates are lower among vaccinated people than unvaccinated. This chart shows age-standardised death rates from all causes.
Source:
Excess deaths arise directly and indirectly from the Covid pandemic. This includes:
- increased pressure on NHS urgent care services, resulting in poorer patient outcomes
- direct effects of Covid-19 infection
- disruption to chronic disease prevention, detection and management.
Both Covid and flu admissions to hospitals increased in the last two weeks. Flu up 33% this week, with more admissions than from Covid, which is up 8%.
Concurrent waves of Covid and flu are piling pressure onto a hospital system which appears not to have the capacity to cope.
This analysis on
@telegraph
front page is fundamentally flawed.
It compares death counts in 2023 with the 2015-19 average, concluding that there have been 53,000 excess deaths.
But around half of that is explained by the ageing population which means deaths increase each year.
So that’s Simpson’s Paradox in a nutshell.
Hopefully the explanation above is clear and will be helpful, at least until we have updated data to conclusively prove the case.
With luck, this latest bit of vaccine misinformation will do a disappearing act!
Latest modelling of the Omicron wave by
@cmmid_lshtm
suggests the following for England:
🦠 21 - 35 million infections
🏥 175 - 492 thousand hospitalisations
💀 25 - 75 thousand deaths
It is only this third element that can reasonably be called “lockdown effects”. Even here though, many would argue that with no restrictions an overwhelmed NHS would have led to more missed or late diagnoses, as well as more short-term deaths.
To state the obvious, we have not been in lockdown for a long time. It is over a year since the last restrictions were lifted. We saw excess deaths after restrictions were lifted last July and we are seeing them again now.
So what are the causes of the “non-covid” excess deaths?
ONS deaths data has been released for week ending 30 Dec.
1,563 more deaths were recorded in-week compared to the 2015-19 average. That’s 20% more. Bank holidays make this week harder to interpret.
Year-to-date there have been 576,896 deaths recorded, 8% more than 2015-19 avg.
In the video John repeated and endorsed incorrect claims made in an anonymous blog by “the nobody who knows everybody” that there have been over 150,000 excess deaths as a result of people getting vaccinated.
Here is how and why we can be certain this is completely false. 🧵
Today on
@BBCMoreOrLess
I will be debunking the most ludicrous claim I’ve been asked to look at! Though that didn’t stop
@Johnincarlisle
sharing it with his 3 million
@YouTube
followers.
9am on
@BBCRadio4
. Thread to follow!
For those who have queried it, when we talk about the "direct effects of Covid-19 infection" we refer to both acute Covid-19 illness and to elevated cardiovascular risk post-infection (which is now well established).
The latest
@ICNARC
intensive care weekly report has been published. A short thread to summarise, including a link to the full report. The report covers critical care admissions to 15 October and compares the second wave (1 Sep onwards) to the first. 1/10
I and others have written to ONS, alerting them to the concerns that this data is causing. It appears from a new blog they have released that they are aware of the issue and will use narrower age bands in the next release.
Wait a minute! This would imply that the trickledown economics idea of bankers wandering their local streets, looking for worthy people to share their huge bonuses with, might not be entirely accurate.
A year ago, I asked
@FinancialTimes
readers about bonus expectations:
💰 40% got a six-figure payout
💷 Most intended to invest all the money and not spend a penny
🏠 13% said they’d pay lump sum off mortgage
😱 7% bought crypto
2023's survey is here:
Newly published data from
@OHID
shows that, from June 2022 to June 2023, excess deaths were highest for ages 50-64, at 15% above normal.
In comparison, excess deaths were 11% higher than expected for under 25s and 25-49s, and were 9% higher for over 65s.
ONS deaths data has been released for week ending 30 June.
918 more deaths were recorded in-week compared to the 5-year average. That’s 10% more - another significant excess.
Year-to-date there have been 310,838 deaths recorded, 8% more than the 5-year average.
1. This recent article by
@saimamaysidik
in
@Nature
gives a good summary of concerns around post-COVID cardiovascular risk:
“studies suggest that the risk of cardiovascular problems, such as a heart attack or stroke, remains high even many months after a SARS-CoV-2 infection”
NEW - Data for December has just been published and paints an even worse picture.
2,200 additional deaths associated with A&E delays in December alone. That is 500 per week.
Method as per linked blog.
Are A&E pressures really leading to hundreds of additional deaths each week?
@ActuaryByDay
and colleagues
@LCP_Actuaries
crunch the numbers and conclude that a significant number of patients could be dying because of long delays accessing emergency care.
Some ‘interesting’ replies. It doesn’t matter whether I think lockdowns are ever necessary. Point is, lockdowns are a tool Governments will use when hospital capacity is threatened. If we want to avoid them we should think twice before opposing other measures (eg masks, wfh, vax)
@AlistairHaimes
Even 18 months in it still astounds me that most so-called “lockdown skeptics” vocally oppose any and all measures which might prevent lockdowns from becoming necessary.
This is a perfectly reasonable question.
#VaccinesWork
and the UK vaccination effort has been incredible, so why do so many people, including me, feel that unlocking on June 21 is just too much of a risk right now? 🧵 (TLDR: Delta/India variant, cases 📈)
Can anyone answer this basic question:
If the
#CovidVaccine
works against the Delta/Indian Variant, as it seems to, then people may get Covid but they won't be seriously ill or die.
Which means a rise in cases won't overwhelm the NHS.
So why not unlock on 21/6 as planned?
2. The claim that if no-one was vaccinated there’d have been under 340,000 deaths in England in a 12 month period is incredible! Anyone with relevant expertise should know straight away that this must be false.
There have been over 450,00 deaths a year for the last 25 years!
Interesting and quite different from the impression you might get from (social) media. Around twice as many say handling of covid was “not strict enough” versus “too strict”.
In hindsight, do you think the government’s handling of the Covid-19 outbreak was…
All Britons
🟰 About right: 34%
✖️ Not strict enough: 37%
❌ Too strict: 19%
Con voters
🟰 50%
✖️ 20%
❌ 25%
Lab voters
🟰 23%
✖️ 54%
❌ 14%
I've been poring over today's NHS data releases and the picture for emergency care isn't pretty.
Last year I estimated 400 additional deaths a week arising from A&E delays, Sep-Nov.
This year looks just as bad, despite less Covid and flu - so far!
I am deeply skeptical about the headline conclusions from this new Oxford uni study. A more considered response will follow but brief thoughts now on why I don’t believe the conclusion that over half the UK has had the virus. 1/9
fascinating, counter-intuitive research from Oxford University suggesting that coronavirus may have already infected half of UK population
@FT
via
@financialtimes
@chrischirp
Agree entirely. There is no doubt whatsoever that clinicians on the ground were making tough triaging decisions about where to direct their limited resources.
A note on triage. Here‘s the Decision Support Tool which aids clinicians deciding who to admit to ICU. I don’t know how long this has been in effect - I’ve been sent it a few times in last 24hrs but not before that.
If total score is 9 or more the guidance is not to admit. /14
Great show of support for vaccination from the intensive care community yesterday! Follows TV coverage of an anaesthetist with unusual views.
There are individuals with fringe views in every profession. Problems arise when they get disproportionate visibility on (social) media.
I’ve gained around 800 followers today - very unusual. Thank you to those who RTd or recommended me - it is much appreciated.
It’s usually a bad sign when people are interested in what a mortality expert has to say - sadly this time is no exception. Still, welcome new followers!
Across all ages, deaths in a private home were 22% higher than expected compared with 10% in hospitals, no excess deaths in care homes, and 12% fewer deaths than expected in hospices.
Deaths from cardiovascular causes in private homes were 27% higher than expected.
ONS deaths data has been released for week ending 21 Oct.
2,073 more deaths were recorded in-week compared to the 2015-19 average. That’s 21% more, which is another significant excess.
Year-to-date there have been 459,819 deaths recorded which is 7% more than the 2015-19 avg.
A classical example of Simpson’s Paradox is currently doing the rounds on social media. I’m always loathe to share misinformation, but at this point it is probably fair to say that this chart is already very much “out there”, so here it is.
2. Current delays for urgent NHS care are extraordinarily long.
See chart below for those waiting >12 hours in A&E for example. Ambulance waiting times and other statistics are equally concerning.
Staffing levels, ability to discharge patients and demand are all relevant here.
Hospital admissions with COVID in England increased sharply, up 41% in-week.
Our R estimate increased to 1.20.
Admissions slightly fell in London (-4%), but rose in all other English regions. The largest increase was 82% in the South East.
Acute bed occupancy increased by 43%.
In fact, we can go back further… much further.
You would have to go back to the 1840s, when the population of England was much smaller, so see so few deaths in one year! If it wasn’t so serious these claims endorsed by John Campbell would be laughable.
Hospital admissions with COVID in England increased by 7% in-week.
R-estimate volatile at the moment, currently 1.06
Admissions fell in the South East (-9%) but rose elsewhere. Largest increases in the North West (20%) and the Midlands (14%).
Acute bed occupancy remained flat.
ONS deaths data has been released for week ending 4 Nov.
1,904 more deaths were recorded in-week compared to the 2015-19 average. That’s 19% more, which is another significant excess.
Year-to-date there have been 482,859 deaths recorded which is 8% more than the 2015-19 avg.
@TonyHinton2016
@abirballan
@CMO_England
@sajidjavid
@ONS
Using such a broad age group (10-59) is highly misleading. Most of the unvaccinated people are young. Most of the vaccinated people are older. Since death rates increase sharply with age the protective effect of the vaccines can’t be seen here.
This chart is deception!
When John Campbell describes the stitching together of the data as “a really clever way to look at this” he is completely wrong.
This statistical bodge job gives estimated death rates for unvaccinated people which are about half what they should be.
ONS deaths data has been released for week ending 22 July, the week of the heatwave.
1,885 more deaths were recorded in-week compared to the 2015-19 average. That’s 21% more.
Year-to-date there have been 320,980 deaths recorded which is 4% more than the 2015-19 average.