Share this page!

John Roberts Profile picture
Twitter logo
Jan 18 • 5 tweets • 2 min read
Latest CH deaths show a big increase this week (to 14th Jan), for the second week running, with numbers up threefold in a fortnight.

We don't have vax status or whether COVID was the underlying cause in these data (or the next chart).

1/
All over 70 COVID related deaths are up this week (to 7th Jan), but that's distorted by Bank Holidays, so I wouldn't read anything into just the one data point.

2/
As usual @ActuaryByDay is looking at the overall picture, but with a divergence in comparator year to the ONS benchmark, as described here.

3/
What benchmark we use is important in any analysis of excess deaths, as I describe here in the latest blog for @COVID19Actuary, when I compare the contrasting ONS and CMI approaches.

4/
A final point is that Week 1 data are usually difficult to interpret, because the BH effect varies by year, so the lag and catch-ups occur in different weeks. Week 2 may also display a similar effect too. (I don't believe that's the case for the CH death data though.)

5/5

• • •

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
Jan 19
Good news, with the latest ONS infection study advance data showing that three of the four nations have now started to fall, with NI the exception (though the rise there is within the conf interval bounds).

E 🔽6.9% to 5.5%
W🔽5.6% to 3.7%
NI🔼5.4% to 5.7%
S🔽5.7% to 4.5%
1/ Image
By region, all are now falling, except for the NE, which showed a further increase. You can see clearly how London peaked earlier, and is now falling fast.

2/ Image
Similarly, only one outlier by age band - that of the youngest from age 2 to School Year 6, which still shows an increase. Though 70+ is flat at 3.1% - would be good to see the oldest age group fall too.

3/ Image
Read 4 tweets
John Roberts Profile picture
Jan 7
JCVI recommend no second booster dose for the most vulnerable yet, although the attached quotes that after 10 weeks protection against mild symptomatic infection is reduced to around 30% in those age groups.

1/
 gov.uk/government/new…
For protection against hospitalisation, after 10 weeks it has fallen from 94% to 89% on the central scenario. Put another way, the efficacy gap has nearly doubled, from 6% to 11%.

The earliest boosted (and most vulnerable) are now well past 10 weeks.

2/
You can see here how protection against symptomatic infection falls off rapidly against Omicron after the booster. A similar picture is seen where AZ was used for the primary course.

3/
Read 4 tweets
John Roberts Profile picture
Jan 7
ONS has now provided the usual data to follow up its early release of the headline data on Wednesday.

Of particular interest is the regional position, most notably the NW, which looks to have overtaken London during the last few days of the year.

1/
Note that whilst the figures below show the NW still lower, these are based on the average for the week. Looking at the daily estimates through to the 31st, the NW is modelled to be 10.6% by the 31st, against London's 7.8%.

2/
Younger adults appear to be driving the increase, but equally of concern is the increase in the oldest age groups, most likely to put pressure on the health system.
The over 70 group has risen from 0.61% to 2.88% using the daily estimates in just 14 days.

3/
Read 4 tweets
John Roberts Profile picture
Jan 7
Signs of a reversal in trend in recent days as to the proportion of beds occupied where COVID is the primary diagnosis.

As cases/admissions appear to be levelling off in London, this is consistent with my recent thoughts, here.

1/
We only have data for acute hospitals, but from these we can see that those being treated primarily for COVID have continued to increase in the last week, regardless of any incidental admissions, up 47%.

2/
In London there's been an increase too, despite signs of levelling in admissions, although the last few days appear a bit odd (data issues maybe?).

NE/Yorks is up 87% and NW 72% on the week.

3/
Read 4 tweets
John Roberts Profile picture
Jan 5
ONS has released its infection survey results early this week, and they show a big, y-axis changing, jump for the week ending 31st Dec.
E ⏫from 3.7% to 6.0%
W⏫from 2.5% to 5.2%
NI⏫from 2.6% to 4.0%
S⏫from 2.6% to 4.5%

In E, 3.3m are estimated to have been infected.

1/
In London the modelling estimates that prevalence peaked on Xmas Day and is now falling back, albeit more slowly than it rose. That's good news, but...

...we also need to look at the age profile of London though.

2/
Here we see a slightly less optimistic picture, with infections at older ages continuing to rise towards the year end, although even the oldest ages appear to peak by the 31st, so hopefully it's downhill across all ages in the new year.

3/
Read 8 tweets
John Roberts Profile picture
Jan 5
Latest COVID CH deaths (to 31st Dec) show a marked increase in comparison to previous weeks, undoing much of the reductions we saw in the previous two months.

As usual, we don't have vax status, nor whether COVID was the primary cause of death.

1/
In contrast, Over 70 deaths (to 24th Dec - a week earlier) show a steeper fall after three more gradual steps downwards. (Overall registrations are up, so there doesn't appear to be a holiday effect involved in the fall.)

2/
You can see here how the proportion of Over 70 deaths fell in the Spring as the initial vax effect benefited older ages first, then reverted to some degree as other age groups benefited too, and more recently has fallen again with the booster effect kicking in first.

3/
Read 4 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

Or Donate anonymously using crypto!

Ethereum

0xfe58350B80634f60Fa6Dc149a72b4DFbc17D341E copy

Bitcoin

3ATGMxNzCUFzxpMCHL5sWSt4DVtS8UqXpi copy

Thank you for your support!

Follow Us on Twitter!

Like this author's thread?

Get emails when @john_actuary has new unrolls!

Check out other threads from @john_actuary!

Read all threads by @john_actuary

:(