Here’s a story about a #COVID19 datapoint that looks very scary, but turns out to be quite reassuring.
The statistic is buried in a recent @PHE_uk release.
I’ve ringed it here.
It’s the no of double-jabbed people who’ve died with Delta variant assets.publishing.service.gov.uk/government/upl…
More vaccinated people are dying w/ the Delta variant of #COVID19 than unvaccinated.
Indeed, look at those aged over-50 and the proportion is even higher:
279 vaccinated people dying vs 131 unvaccinated.
More than double.
Scary, right? Well not necessarily…
Let’s run through it
There are two things to bear in mind here.
First, the vaccines are not 100% effective. They reduce your likelihood of dying of the disease. But even after vaccination some age groups are still relatively vulnerable. @d_spiegel explained this well here theguardian.com/theobserver/co…
Here’s a useful @Johnnmurdoch chart illustrating this (I’ve done a similar one for @skynews tonight).
Post-vax, a 90 year old faces similar risk as an unvaxxed 60yo. Same thing for an 80yo: after being jabbed, similar risk profile to an unjabbed 50yo.
Risk reduced. Not eliminated
The second thing to bear in mind is that the vast majority of over 50s in the UK - the age group most likely to face these higher mortality levels - are vaccinated. About 91% at present. And, over the period that PHE data covers, around 89% had at least one jab…
Now, back to the figs from that PHE dataset - the one that shows the majority of hospitalisations and deaths among over-50s were vaccinated. First thing to do is to compare the case numbers with the hospitalisations.
Like this. Notice anything?
That’s right: the dark blue bit of the bottom bar - the hospitalisations - is smaller than in the top bar - all the cases.
To put it another way, consider the proportion of Delta cases among over-50s that turn into hospitalisations:
3.5% among vaccinated
8.4% among unvaccinated
In other words, the hospitalisation-per-case ratio for those over-50s who are unvaccinated is 2.4 times higher than for those who have had one or more jab.
So they ARE effective.
Yet, and this is the crucial bit, the NUMBER of vaxxed people hospitalised is nonetheless higher
Similar thing for the death figures.
Remember those numbers above👆: 279 vaccinated people dying, vs 131 unvaccinated.
How do they look when we put them into the same kind of context…?
Compare the blue/red proportions in the bars
Of those vaccinated over-50s who caught Delta, 2% died.
But of those unvaccinated over-50s who caught Delta, 5.6% died.
That’s a significant difference, yet is only really apparent when you compare cases with deaths.
Looking at deaths alone is misleading.
Indeed, with 2.8 times more deaths per case among the unvaccinated portion of cases, you’re talking about high levels of vaccine effectiveness. How high? Perhaps 70-80%, but I’d be wary of putting too much weight on this dataset as I’m not sure it’s representative of the pop’n
It’s a reminder of something we’ve often encountered throughout #COVID19. CONTEXT IS EVERYTHING.
As populations become mostly vaccinated, simple numbers of hospitalisations/deaths will be dominated by the vaccinated.
It doesn’t mean the vaccines aren’t working as expected
Why are so many of those being hospitalised and dying of #COVID19 in the UK vaccinated? Here's a short video breaking down the numbers

• • •

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

Keep Current with Ed Conway

Ed Conway 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 @EdConwaySky

24 Jul
Remember how, back in April/May, the decision was made that under-30s and then under-40s should be given the option to have a vaccine other than AstraZeneca, because of the risks of side-effects? Well, the calculus has now changed and the benefits now probably outweigh the risks
It comes back to this chart. You might remember this from when Jonathan Van-Tam announced the decision. The point was that the bars on the right (the risks of blood clot events) were bigger than the bars on the left (the risks of getting #COVID19 and going into ICU) for under-40s
That chart was predicated on the notion that the risk of getting #COVID19 was pretty low. It was, back in April, because only one in 600 people had the virus. But since then things have changed. Now it’s one in 75 who have #COVID19. Meaning those bars on the right are bigger.
Read 8 tweets
12 Jul
Quick Covid data update. Cases heading higher, so too admissions. But the link between them is weaker than in previous waves.
In short, the coming months will be nerve-wracking.
Cases will get v v high. This will feel like a wave in that respect. But in other respects v different
Same data but on a log axis - now you can see that hospital admissions are now growing at almost the same rate as cases.
Roughly doubling every 11 days.
V unsettling. Especially given cases have further to rise. So what next?
Here’s a simple extrapolation of cases and hospitalisations (second chart is same data with log axes).
Extrapolate current growth rates and you’re talking abt 100k cases by late July.
Hospital admissions up to 1000 by 19 July, possibly touching 2000 by end July.
Read 10 tweets
7 Jul
Been a week and a bit since I did a COVID data update. Tbh the pattern is broadly the same as it was then: cases & hospitalisations growing. Both a bit faster than two weeks ago. But the link between them remains much weaker than in previous waves
Last time around we looked at the growth rate in cases and saw that if that continued it would mean 40k plus daily cases by “Freedom Day” July 19.
As you can see, that line is steeper. Points towards a higher level.
Current extrapolation suggests maybe 80/90k…
But hospitalisations are still broadly in line with that extrapolated line from a couple of weeks back. Ie heading up but less rapidly than cases. Where are they by 19 July? On basis of past fortnight’s daily growth rate, still 500-600. On basis of past week, 600-700.
Read 9 tweets
25 Jun
Few thoughts on UK COVID data.
There’s good news and bad news.
But the good news is that the good news outweighs the bad IMHO.
Let’s start with cases. They’re growing in the UK at a decent snap, albeit not as fast as winter. You can see this better in the log chart (2)
But the real question is not just about cases but whether they’re turning into deaths.
That was the case for most of the pandemic: look at how correlated these lines are.
But now look bottom right. Do they seem to be diverging? Looks like it (albeit still quite early days)
That brings me to the most encouraging bit of data out recently: the case fatality rate of the Delta/Indian variant vs Alpha/Kent and others.
Look! While 2% of those catching Alpha died, the CFR for Delta is only 0.3%. Massive improvement.
Read 17 tweets
15 Jun
What kind of impact will the Australia-UK trade deal have? Short answer is we don’t know, in part because we still have scant detail abt it, in part cos it generally takes a LONG time for these deals to be felt in our lives.
Even so, I think this is a big deal, for a few reasons:
The first thing to remember is the historical context. There are two lazy assumptions knocking around: 1. That the UK could never trade much with Aus given how far away it is and 2. Our trade with them really collapsed when we joined the EU and started imposing their tariffs…
Actually look at Australian trade as % of UK total, it’s clear it used to be REALLY high. Yes, this was partly because of war and partly because it was part of the sterling zone, thus locked into UK economy. But even in ‘50s/60s Aus accounted for 1 in 8 of all imports/exports
Read 18 tweets
14 Jun
There we have it. According to @borisjohnson the UK hasn’t passed the four tests laid out in Feb when the reopening timetable was announced.
A pretty clear pass on tests one and two (vaccines) and a pretty clear fail on test four (variants). Test three unclear.
Reopening delayed.
Here’s the q: what changed between the PM saying 12 days ago that he could see “nothing in the data” to change re-opening timetable and today?
Variant growth rate v similar.
Case growth rate v similar.
Not sure there’s any data we have now that we didn’t on June 2.
What changed?
The point of basing policy on DATA is that you actually follow the data.
Better still, you make it clear what levels of data will trigger what kind of action.
Why? That kind of transparency means people can use that data to plan their lives rather than hoping for leaks from No10
Read 6 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!

:(