Share this page!

John Roberts Profile picture
Twitter logo
24 Nov, 4 tweets, 2 min read
The latest data on registered deaths shows a continued increase in excess deaths, and we've now passed the grim milestone of exceeding 2019 total deaths. I've seen comment that the recent xs is due to a catch up of registrations. There's no evidence of that as we can see... 1/3
The ONS helpfully provides an estimate of weekly deaths by date of death. Recent weeks have more uncertainty, as a smaller proportion of the eventual total has been registered, but it's clear that the view is still of continuing increases. 2/3
This data is all just based on the weekly raw death numbers. As @ActuarybyDay notes, a better view, adjusting for population growth and ageing of the population will be published later today by the CMI, and we will link to it with analysis when released. 3/3
Addendum: Quick of the mark as ever, @Longevitymatt has already added this week's figure onto a graph which plots each week for the last 5 years. The excess we are seeing is well clear of any normal variation we might see at this time of year.

• • •

Missing some Tweet in this thread? You can try to force a refresh
　

Keep Current with John Roberts

John Roberts Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!

PDF

Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @john_actuary

John Roberts Profile picture
17 Nov
The actuarial profession's CMI has produced its weekly update on mortality during the pandemic. It reports that in the latest week (to 6th Nov) excess deaths were 9% above expectation. Some more context to follow. 1/11
The analysis uses age standardised mortality rates, which adjusts for changes in the size/age dist'n of the pop'n. This gives a better answer than using numbers of deaths - typically you get a lower excess as our population is ageing, so we would expect more deaths each yr. 2/11
This chart compares 2020 with the average of 10-19. Note that as mortality has improved over the decade you would expect more recent years to be lower, and vice versa. Note also how close 20 was tracking to 19 before late March. You can see the line drifting up again now. 3/11
Read 11 tweets
John Roberts Profile picture
8 Nov
A summary of some key points from @ICNARC's latest weekly update on COVID patients in ICU, on behalf of @ActuaryByDay and @COVID19actuary. New admissions have now risen to 3129 since 1 Sep, with 1263 in the last 14 days. It's still heavily weighted towards the North. 1/9
The slower growth than in March is very apparent, though numbers are steadily climbing. The last few days will have reporting delays, so need to be taken with a degree of caution. We know that the picture is very varied regionally, so let's take a look at that data too. 2/9
Except for NI, all areas are well below their first wave peak. Note how London, by population, was hit much worse first time around, so any comparisons with it appear more favourable. In fact, this time rates are broadly consistent with most areas, except for the NW. 3/9
Read 9 tweets
John Roberts Profile picture
6 Nov
The latest @ONS infectivity survey has been released. For England ONS suggests that new infections have levelled off at around 50,000 per day. (The estimate has reduced from 51,900 to 45,700, but this must be seen in the context of the confidence intervals around each one.) 1/6 Image
Overall infections have increased, albeit more slowly, from 568k to 619k, being 1 in 90 of the community. Remember that this survey excludes healthcare settings, including care homes. 2/6 Image
The regional picture shows a clear reduction in the NW, consistent with the levelling off in admissions that @COVID19actuary has been reporting of late. Other regions appear to be stabilising too. 3/6 Image
Read 6 tweets
John Roberts Profile picture
29 Oct
The interim report of Round 6 of @imperialcollege's #REACT study makes grim reading. It estimates 96,000 new infections a day, with an R of 1.56, and doubling every 9 days. Nearly 1m are estimated to be infected. Covering 16-25 Oct, it's very up to date. Here's some detail. 1/8
Regionally, Yorks/Humber has overtaken the NW, consistent with admissions data @COVID19actuary has been reporting recently. Lower prevalence areas have higher growth rates, with R over 2, so there should be no complacency elsewhere, with short doubling times. 2/8
Similarly for age, although 18-24 is highest the most rapid growth is now at older ages, with a threefold increase at 55-64, and doubling above 65. It's these groups which will be driving the increases in admissions we've already seen (and sadly, subsequent deaths). 3/8
Read 8 tweets
John Roberts Profile picture
27 Oct
The latest #REACT study by @imperialcollege shows decreasing levels of antibody prevalence. Overall levels fell by 24% over a 12 week period, from 6% to 4.8% in Round 2, and now 4.4%. Let's take a closer look in this short thread. 1/6
By age the % fall was greatest in the oldest groups, with an increasing trend from age 45. There is some variability though, eg in the data table in the report we see 75+ increased in Rd 3 by 25%, reflecting some wide CI's. 2/6
Showing the wide regional variation, London fell 27% from 13% to 9.5%, whereas at the other end of the spectrum, the SW fell 42% from 2.8% to 1.6%. Note again the increase in Rd 3 for the SW, with a very wide CI for the change we shouldn't read too much into this. 3/6
Read 6 tweets
John Roberts Profile picture
23 Oct
The latest weekly report from @ONS on infectivity shows a further increase for England, with 433k (up from 336k) estimated to be infected in the most recent period. That's 1 in 130 (1 in 160), and with 35,200 (27,900) new infections per week. 1/6
The regional trends are starting to show some variation, with some signs of levelling in the last couple of weeks for the worst affected regions. Wider confidence intervals make these more uncertain though, but hopeful signs maybe. 2/6
By age there's a clear fall in early teens, a levelling off for younger adults, but still growth at older ages, albeit at a lower level. Those are the ages which matter in terms of hospitalisation though. 3/6
Read 7 tweets

Did Thread Reader help you today?

Support us! We are indie developers!

This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Too expensive? Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal Become our Patreon

Thank you for your support!

Follow Us on Twitter!

Like this author's thread?

Get email notifications when new thread is out from this Author!

Check out other threads from this Author!

Read all threads by @john_actuary