B.1.1.529 Omicron variant

[lpm]

Scotland.

Big range for the estimates of the initial ramp up, let alone the subsequent tsunami - 20 December best case 1,250, central 15,000, worst 25,000.

https://www.gov.scot/binaries/content/d ... -paper.pdf

Based on data to 6th December, we can estimate a doubling time for Scotland of between 2.18 - 2.66 days using S-gene target failure as a proxy for Omicron cases.

As a result Sturgeon is taking more action than England. Cancel those Christmas parties and don't expect Hogmanay. Any household infection requires isolation for all regardless of vaccination status - none of this English nonsense of people heading off to the work's Christmas bash while their partner is ill with a confirmed Covid infection.

[lpm]

Where's the Brexit stockpile thread? Time to get in January's supplies I think.

[Woodchopper]

Trevor Bedford thread: https://twitter.com/trvrb/status/146915 ... 72708?s=21

Omicron Rt for UK estimated to be about 6, doubling time of 2.2 days. Though that will probably slow down a bit.

Also reiterated that Omi is infecting the vaccinated and naturally immune at a far greater rate than Delta. But those people will be protected to a certain extent from serious illness.

[headshot]

none of this English nonsense of people heading off to the work's Christmas bash while their partner is ill with a confirmed Covid infection.

Almost everyone I know is being instructed to work from home as of Monday and their Christmas outings have been cancelled.

Looks like a good proportion of businesses are taking the responsible line themselves.

Hospitality is going to be devastated…again.

[lpm]

Trevor Bedford thread: https://twitter.com/trvrb/status/146915 ... 72708?s=21

Omicron Rt for UK estimated to be about 6, doubling time of 2.2 days. Though that will probably slow down a bit.

Also reiterated that Omi is infecting the vaccinated and naturally immune at a far greater rate than Delta. But those people will be protected to a certain extent from serious illness.

This is the moment when I get irrationally pissed off with moron maths pedants who say "Ah ha, R=1.1 is exponential too you know".

This is what f.cking exponential means in a pandemic:

[Woodchopper]

Trevor Bedford thread: https://twitter.com/trvrb/status/146915 ... 72708?s=21

Omicron Rt for UK estimated to be about 6, doubling time of 2.2 days. Though that will probably slow down a bit.

Also reiterated that Omi is infecting the vaccinated and naturally immune at a far greater rate than Delta. But those people will be protected to a certain extent from serious illness.

This is the moment when I get irrationally pissed off with moron maths pedants who say "Ah ha, R=1.1 is exponential too you know".

This is what f.cking exponential means in a pandemic:

Pure speculation but perhaps the difference between SA and UK is that its summer in the southern hemisphere and winter in the north.

[jimbob]

Trevor Bedford thread: https://twitter.com/trvrb/status/146915 ... 72708?s=21

Omicron Rt for UK estimated to be about 6, doubling time of 2.2 days. Though that will probably slow down a bit.

Also reiterated that Omi is infecting the vaccinated and naturally immune at a far greater rate than Delta. But those people will be protected to a certain extent from serious illness.

This is the moment when I get irrationally pissed off with moron maths pedants who say "Ah ha, R=1.1 is exponential too you know".

This is what f.cking exponential means in a pandemic:

I still don't see R as particularly useful except as a shorthand. There's no reason to believe that it's become less clumpy in each cases individual R, and a doubling time just seems far more useful.

[lpm]

R is most useful as a way to sense how much lockdownicity needs to be added to the system.

R=1.1, don't bother
R=1.5, add some significant measures
R=2.0, proper lockdown
R=3.0, do everything possible to try and flatten the curve while preparing for the overwhelmed NHS

[wilsontown]

It seems that Neil Ferguson broadly agrees with lpm's analysis, however much weight you want to put on that...

https://www.theguardian.com/world/2021/ ... guson-says

[jimbob]

It seems that Neil Ferguson broadly agrees with lpm's analysis, however much weight you want to put on that...

https://www.theguardian.com/world/2021/ ... guson-says

It's simple maths. And it has to be significantly less virulent than Delta in order for it to NOT overwhelm the NHS given its infectivity and lack of mitigation actions

[shpalman]

Two doses of a Covid vaccine offer less protection against symptomatic infection with Omicron than with Delta

The vaccine efficacy data suggests that people who have had two doses of the Oxford/AstraZeneca jab 25 or more weeks ago have far lower protection against symptomatic infection with Omicron than with Delta. While the data suggests about 40% protection against Delta at this time point, it suggests protection from Omicron could be less than 10%. There is a great deal of uncertainty around that figure, though, because of the small number of people studied and the fact that the Oxford/AstraZeneca jab was a bit sh.t.

f.cking glad I got a Pfizer booster recently.

A similar trend was seen for those who have had two doses of the Pfizer jab, with about 60% protection against Delta at 25 or more weeks since the second dose, and just under 40% protection against Omicron at the same time point – although again, there are uncertainties around the figures.

[monkey]

Two doses of a Covid vaccine offer less protection against symptomatic infection with Omicron than with Delta

The vaccine efficacy data suggests that people who have had two doses of the Oxford/AstraZeneca jab 25 or more weeks ago have far lower protection against symptomatic infection with Omicron than with Delta. While the data suggests about 40% protection against Delta at this time point, it suggests protection from Omicron could be less than 10%. There is a great deal of uncertainty around that figure, though, because of the small number of people studied and the fact that the Oxford/AstraZeneca jab was a bit sh.t.

f.cking glad I got a Pfizer booster recently.

A similar trend was seen for those who have had two doses of the Pfizer jab, with about 60% protection against Delta at 25 or more weeks since the second dose, and just under 40% protection against Omicron at the same time point – although again, there are uncertainties around the figures.

Also, Pfizer are saying that it looks like 3 shots of theirs gives similar protection to Omicron than two jabs gives to other variants - clicky. It's all preliminary though.

[lpm]

In chart form. AZ on the left, Pfizer on the right.

Shows how feeble two AZs are against Omicron, but getting the Pfizer booster takes it right back up again. (ETA: it's not a great presentation of information, but the the dots on the left of each chart are the first two doses, with the final right hand dot being the booster dose.)

Unfortunately, the main heavy lifting in the UK was AZ. Very large number of two AZs in the 50-75 age groups.



https://twitter.com/kallmemeg/status/14 ... 0864322560

[jimbob]

https://www.bloomberg.com/news/articles ... han-double

The daily number of people admitted to hospital in South Africa with Covid-19 more than doubled on Tuesday from a day earlier.

According to the National Institute for Communicable Diseases 383 people have been admitted to hospital with the disease in the last 24 hours compared with 175 in the preceding period.

Of course, I can't tell if there is a day of week effect

[lpm]

The "immunity wall" chart from John Burn-Murdoch Twitter.

https://twitter.com/jburnmurdoch/status ... oHM7g&s=19

This is the chart version of what I did weeks ago - think in terms of protection points. Half the population has 3 points (some 4) now. Crucially 85% of the over 65s have 3 points.

But the chart or points don't have a recentness metric. All boosters are recent, plus a lot of young people scored a recent point either via 2nd dose or infection.

I'm sure there's a lockdown path through this, despite Omi being so damn fast. Just need to win an extra four weeks of boosters.

Ironically, the EU countries with current Delta problems are the best placed. They have lockdown measures already. And they have recent infection protection points. The UK has the best booster rate, but older and more waned two doses, plus more weaker AZ. The UK has the fatal weakness of Johnson, obvs.

[bob sterman]

Another important factor that influences how bad things will be is how high are booster rates among frontline health and social care workers?

On the one hand, many will have had 2 x Pfizer - which was commonly used for hospital staff. On the other hand many of these were jabbed about 12 months ago.

When it comes to NHS staffing - protection against LFT detectable infection - is going to be the key metric. Since they are testing regularly - and not working if positive.

[shpalman]

let's see if the public is as indifferent to 4000 deaths per week as they obviously are to 1000 deaths per week

Well, 75,000 deaths before the end of April, which is 20 weeks away, would be 3,750 a week but they're obviously going to ramp up from where they are now (>800) over the next few weeks. The lower estimate of about 24,000 wouldn't be that much worse to what's been going on since September (there have been ~13,000 deaths in the past ~14 weeks but there were 7-day periods with >1,100).

Of course it depends if what they mean is that there'll be that many Omicron deaths in addition to the Delta deaths, but I don't really think it will work like that.

[shpalman]

The "immunity wall" chart from John Burn-Murdoch Twitter.

https://twitter.com/jburnmurdoch/status ... oHM7g&s=19

This is the chart version of what I did weeks ago - think in terms of protection points. Half the population has 3 points (some 4) now. Crucially 85% of the over 65s have 3 points.

But the chart or points don't have a recentness metric. All boosters are recent, plus a lot of young people scored a recent point either via 2nd dose or infection.

I'm sure there's a lockdown path through this, despite Omi being so damn fast. Just need to win an extra four weeks of boosters.

Ironically, the EU countries with current Delta problems are the best placed. They have lockdown measures already. And they have recent infection protection points. The UK has the best booster rate, but older and more waned two doses, plus more weaker AZ. The UK has the fatal weakness of Johnson, obvs.

When I first read the headline alert for Having two doses of a Covid vaccine offers less defence against symptomatic infection from the Omicron variant than Delta I thought it meant that a Delta infection gave you better protection against Omicron than two doses of the vaccine[/url]. That would have been interesting.

But still there are only 367,386 people in Italy who are officially considered to have recovered from covid within the past 6 months, whereas about 30 million have been "fully vaccinated" in that time.

The UK has had a "Delta problem" since July, it has just ignored it.

[lpm]

I'm puzzled by so much of the commentary.

We know Delta breaks through vaccines and puts people in hospital.

We know vaccines aren't as good against transmission for Omicron.

It would be really weird if vaccines are better at preventing hospitalisations from Omi than Delta. It's a weird virus, but that's taking weird too far. The best we could assume is that vaccines protect against hospitalisations at the same level as for Delta.

But more likely is that protection against hospitalisations is lower, to correspond with the known fact of vaccine efficacy being lower for transmission.

Yet everyone seems to be seduced by the "milder" hypothesis. Omicron might have less natural virulence, sure. But I've no idea why people make this their implicit starting assumption. Makes far more sense to assume the same virulence, plus slightly less benefit of vaccines, to give a slightly higher underlying rate of hospitalisations - when comparing a vaccinated person with Delta to a vaccinated person with Omicron.

I think people are confusing mix effects with underlying effects. The mix will tilt to more vaccinated people getting infected. But every commentator should be used to mix effects by now, because we've understood how it works when infections concentrate in young people.

[lpm]

The LHSTM study assumes Omicron is less severe than previous variants if you have been vaccinated

I don't get it. Why would they assume this? Is there some finding from South Africa that I've missed?

[shpalman]

The LHSTM study assumes Omicron is less severe than previous variants if you have been vaccinated

I don't get it. Why would they assume this? Is there some finding from South Africa that I've missed?

Prof Hunter said he suspects these models “overstate” risk of hospital admissions and deaths and the “worst case” scenarios are “unlikely to be seen”.

Although we will not know for certain for a few weeks indications from South Africa do suggest that Omicron does cause less severe disease than Delta.

There is also early as yet not peer reviewed data suggesting that although Omicron mutations are enough to escape antibody, T cell immunity would be less compromised.

It is thought that T cell immunity is more important for reducing risk from severe disease than it is for reducing the milder nose and throat infections.

If Omicron is indeed associated with less severe disease as is, in my view, likely to be the case then these models would overestimate hospital admissions and deaths, possibly substantially.

It may be slightly wishful thinking, we won't know if the indications from South Africa aren'y confounded by lots of undetected prior infection or the younger population or downplaying whatever.

[headshot]

I’ve listened to a number of podcasts that have featured doctors from SA, the USA and the UK who have all talked about Omicron causing less serious disease - but without citing anything but data reported by SA. None of which seems particularly reliable evidence.

[Woodchopper]

The thing is that Delta became significantly less severe as a consequence of mass vaccination. As far as I remember from about 10% of cases needing hospital treatment down to about 2%.

I assume that when people talk about Omi being less severe they mean in comparison to the early phases of the other variants. I haven’t seen any evidence that Omi would be less virulent in a population without any antibodies (if such a population even exists now).

[jimbob]

Good thread with lots of links about severity etc here

https://twitter.com/DrTomFrieden/status ... 95328?s=20

[lpm]

The thing is that Delta became significantly less severe as a consequence of mass vaccination. As far as I remember from about 10% of cases needing hospital treatment down to about 2%.

I assume that when people talk about Omi being less severe they mean in comparison to the early phases of the other variants. I haven’t seen any evidence that Omi would be less virulent in a population without any antibodies (if such a population even exists now).

That's ratios. It's misleading and is making people go wrong.

If Delta puts 2% of cases in hospital, Omicron will put 1%, say. Far less. That's guaranteed. It's simply the outcome of the mix.

We saw it with Delta. A huge number of kids in the case numbers, the hospitalization ratio falls. A huge number of vaccinated breakthroughs and the ratio falls.

But that's just aggregation across a population and is irrelevant to real world outcomes. It's individuals and their personal risk that matter. I'm never sure what people are talking about with "milder" because they seem confused by mix vs individuals.

I'm not managing to explain it very well though. What we need to discover is individual impacts.

Take David. He is 80 years old. Two Pfizers. Likes to spend time with grandchildren.

He didn't catch Delta. Either luck or the good vaccine protection. He's now quite likely to catch Omicron, because it'll be so widespread and his Pfizers are less protective against infection.

He catches it. Has David's risk of hospitalization changed vs if he'd caught Delta?

- Higher risk, because two Pfizers are still very good protection against hospitalization, but not quite as good as before
- Lower risk, because Omicron is milder underlying virulence

It's impossible to determine individual risk from aggregated national statistics because the vax vs unvax ratio changes in such a dramatic way for Omicron. Look at any aggregated statistic and you'll see a far lower hospitalization ratio - tricking you into telling David he's got a lower risk of going into hospital now he's caught Covid. This might well be false.