Thread by @ewanbirney

COVID thoughts from cold, Christmas London. TL;DR The Omicron storm clouds are glowering above Europe, but its unclear whether this will just be rain and hail of COVID or rain, hail, tornados and hurricanes of infection and subsequent hospitalisation of COVID.

Context: I am an expert in human genetics and computational biology. I know experts in infectious epidemiology, viral genomics, immunology, public health and clinical trials. I have COIs - I am consultant and shareholder to Oxford Nanopore and was on the Ox/Az vaccine trial.

Reminder: COVID is horrible disease triggered by the infection of SARS-CoV-2 in a subset of people. A combination of reducing contacts (early), reducing travel (some countries) and vaccination brought SARS-CoV-2 towards a somewhat endemic trajectory over 2020 and 2021. However >>

A new variant, Omicron, with a large number of mutations, emerged and was rapidly characterised by excellent infectious epidemiologists in South Africa. Huge credit to the turn around and transparency of this team.

Similar to the 4 other endemic coronaviruses, Omicron has evolved to have its most exposed antigen (spike) to look different from previous variants, meaning previous infection and vaccination is less effective, sometimes not at all. In addition, Omicron is a "fast transmitter"

This has lead to eye-watering doubling times (2 to 3 days! ~100 fold in 2 weeks!!) for Omicron spread in European countries (probably elsewhere), and it is pretty inevitable that this will become now the third globally dominant variant after Alpha and Delta.

Some countries (eg DK, UK, IS, BE, IL) have fast variant sensitive PCR and/or sequencing, and they largely all see fast doubling times of Omicron if introduced. Most other countries that have not noted much omicron mainly because they can't see it or their data flows are slower.

Due the speed of spread scientists are still characterising it whilst it is rapidly spreading. Decisions therefore have to be taken with far from perfect data, parameters and models about the future.

A key piece of information has emerged from the UK Health Security Agency; they confirm the fast doubling rate and the "micro-study" of household attack rate, but importantly can characterise the impact of vaccination on exposure to Omicron.

This preliminary data shows that Ox/Az : Ox/Az vaccines provide close to 0 (!) protection to Omicron; Biontech:Biontech (and likely Moderna) ~40% but either Ox/Az x2 + Biontech or Biontech x2 + Biontech bring protection against symptoms back to just under the protection vs Delta

(phew). Basically, boosters are critical for Omicron. Get Boosted; help your friends and family to get boosted. But we're not out of this storm just by boosting...

Another parameter is the severity of the disease from Omicron. There is crudely less hospitalisation in South Africa for the case levels in Delta. However, these statistics are complex because the transmission over previously infected means denominators of cases is high

Furthermore South Africa's path through the pandemic (far more natural infection, often Beta) and its demographics (younger) means it is not an easy mapping from this setting to, say, Europe.

All this means there is a lot of uncertainity for Europe's hospitalisation via Omicron, though some things are clear. With a substantial amount of unvaxed individuals, in particular in older ages, the Omicron storm will hit soon and they must decide what to do *early*.

The worst situation would be to have overflowing healthcare triggering harsh lockdowns when this is pretty predictable now given the rate of infection that will happen.

It's worth noting that even if Omicron was 10-fold less severe- which sounds great - this is just ~8 days of higher transmission due to this doubling time. (As I said, eye-watering). With a substantial amount of unvaxed individuals one will have to control transmission hard

For countries who have vaxed and boosted their older age groups (eg, IL, IE, UK, PT, DK) they arguably have a harder decision surface - can they make it through the Omicron storm? If there is a chance, it likely will still need need their healthcare to process many many people.

(this is because even in these countries vaccination is not full so there are unvaxxed people at risk, and there will be breakthrough infections leading to hospitalisations at some rate)

As healthcare is a linear process (one has a certain number of beds, can add linearly more) but infection is exponential (doubling every 2 days! 100 fold in 2 weeks!) impedence matching infection to healthcare is very very hard.

This is one of these horrible "I hope I am wrong" advice scenarios - you prepare for the worst, hope for the best; if the best happens, relax. But everyone doesn't like really thinking about the worse case and plenty of "oh it will work out somehow" thinking can creep in.

Stepping back, as other commentators have said, arguably Omicron is our first taste of what "endemic SARS-CoV-2" looks like, with the evolution of the spike to evade immunity. It is likely we all must get it a number of times over the coming years (definition of endemic)

I think it is going to become more and more important to focus on COVID the disease, as well as (note: not instead of) SARS-CoV-2 the infectious agent. Antivirals is one step here (let's hope they are really as good as they seem in the trials).

I suspect there will be other ways to calm down our immune system, and indeed refine the prediction of who is at risk potentially of the infectious disease trigger. This is going to be hard miles of basic biology, clinical research and then interventions. A long road.

Two final points. Huge credit to the scientists here who acted fast; many many people to credit (it's definitely a worldwide effort), but the @nicd_sa and the @UKHSA teams stand out to me.

The second is a reminder that this is a humanity vs virus thing, not a country thing. We will need to have a system that manages this virus across the planet, for the benefit of everyone.