1870

Bright side, maybe much quicker sickness so less community spread? I know, single data point, but I’m reaching for the silver lining here.

1871

It would presumably make contact tracing easier if incubation is shorter.

1872

I was thinking more along the lines of responsible people staying home when they are sick, but I’ll take anything.

1873

I meant in addition to that, sorry

1874

Or his known exposure isn’t the one where he got infected.
Anyway, that particular anecdata-point is not especially encouraging, given that he is fully vaxed.

1875

Spread in Norway and Denmark is pretty significant already.

One single Christmas party in Norway in November has so far resulted in 130 people being infected - 70 of the people in the party and 60 more people who were just in the restaurant on the night in question. Worth noting that everyone participating in the party itself were vaccinated (as that was a demand) and everyone was tested before the party. On average, >80% of all adults have the vaccine.

Got sent back to home office last friday and expect I’ll be there until at least February - March. Christmas parties are still a thing, though; Omicron is just going to have some outstanding days this winter and is going to tear through the population in the coming weeks.

Edit: This was actually intended for the other thread.

1876

Yes. Denmark is at about 2% of cases Omicron last Saturday, doubling every 2-3 days. See page 2 of https://files.ssi.dk/covid19/omikron/statusrapport/rapport-omikronvarianten-07122021-1t6o. There’s also a breakdown of the vaccination status of the cases on page 5. 77% of cases are double-vaccinated, 6% boosted.

UK is at 2% last Saturday, doubling every 2 days: https://twitter.com/AlastairGrant4/status/1468151959957950466 (sorry, can’t find a primary source, but the author claims the data is from UKHSA and appears to be reliable)

Switzerland is at 3.6% according to the authorities, but I can’t find a time series.

Unless something changes, it’ll be the dominant variant in Europe in 2-3 weeks despite the huge ongoing Delta waves.

My expectation at this point is:

  • The efficacy of vaccines / previous infections against Omicron will be <10%.
  • The efficacy of vaccines / previous infections against serious cases will be <50%.
  • It will not be any milder than Delta.
  • The Omicron wave will be steeper and higher than any of the previous ones.

There will be huge political pressure to do just wait and do nothing in order to “save Christmas”, so this is looking kind of grim.

1877

What’s driving this number? Seems out of step with most of what I’ve read.

1878

We can consider South Africa to be almost entirely immunized (one way or another), and the numbers on severity from there just don’t look that good.

Just to be clear, none of that data can possibly give us an accurate number on the vaccine efficacy. But it at least does not feel compatible with the efficacy still being 90%+. I hope I’m wrong.

1879

Thanks for the links! One thing I don’t understand - in the twitter thread, the share of patients in ICU during the delta wave was higher than the fraction receiving oxygen? Did they have an oxygen shortage at that time?

1880
1881

I don’t understand why anyone thinks this is the case. I’m not saying it isn’t — I genuinely don’t know — but it seems pretty unlikely to me that it is the case, and I don’t really know where the supposition is coming from or why it has any constituency.

There are 60m people in South Africa, of whom about 15m are fully vaccinated. If the other 45m of those people had already had COVID, I think the data from that country would look quite a bit different than it does. Wouldn’t it? We’d see more than 3m cumulative reported cases, and more than 90k deaths, wouldn’t we?

1882

Depends on SA’s IFR, but I don’t see many reasons why it would be significantly off the 1.5-0.5% estimation in other countries.

Using it as a tool to project, deaths are about a third of what would be needed for full immunization/exposure.

Edit: ah, SA has recorded 270k excess deaths. Yes, that looks like full immunization.

1883

Here’s an earlier discussion:

It’s a young population, so the baseline expectation is that the mortality rate will be low (just based on the age distribution, you’d expect an IFR of 0.35%). Their excess mortality is actually higher than that: 0.39% of the population. And they already had 25% seroprevalence a year ago, before the variants hit.

1884

I confess I’m skeptical. It seems unlikely to me that they’ve somehow only detected 1 case in every 15. Is that a reasonable result?

1885

I guess that depends. In the US I’d say no, but given the availability of testing, timing of waves, access to medical care, etc maybe that is for South Africa?

It certainly seems like a far end of the statistical likelihood, but it does not seem impossible to me that some countries may only detect 1 in 15 cases, if they only test symptomatic cases and have low contact tracing.

1886

The excess death figures are hard to dispute, they are too high. IFR is higher than 0.39%, since that gives a too high number of cases.

1887

I don’t say it’s impossible. I just don’t know why you use something as your starting assumption on the basis that it isn’t impossible. And if the case counts are badly wrong, seems likely the other counts are wrong, too.

1888

Well the official number I see is ~90k deaths for COVID. @Juan_Raigada is using an excess deaths number of 270k, which is significantly higher than the official. I don’t know where that number is derived from, but if it is accurate at all, it would comport with the other data counts being wrong.

1889

Yes, that makes sense.