UK Announces Plan B in response to Omicron.

The UK Government is going to introduce the dreaded 'Plan B' in response to the omicron variant.

For me the key point of any announcement or action is the form this takes. I support imposing these kinds of measures now but with an explicit proviso that they will be reviewed in two to three weeks time when we should have a firm idea about the nature of the new variant. At that point, unless a particular set of things prove to be the case, the precautionary measures should be stopped on cost-benefit grounds. If however the worst does emerge then continue them. I fear though that we will just get an indefinite and open ended announcement.

Why is there such concern about this variant when there wasn't for the several others that have emerged between delta's appearing and becoming dominant and this one?

The reason is the truly extraordinary number of mutations it has (52 by the best estimates). This is a low probability event. What it means is that it is quite possible that vaccines will not work against it or have much reduced efficiency, because it is now so different from the version they were designed to combat. It's important to be clear what 'not work' means here. It does NOT mean unable to prevent you getting infected. The existing jabs already do not protect people against infection. The point is they protect people against severe illness as a result of infection (not the same thing) which means that even large numbers of infections do not lead to as many people having a severe illness and requiring hospitalisation as was the case before. What the vaccines do is to reduce the proportion of those infected who require hospitalisation. That is important because in every country it is the need to protect the hospital system that has driven policy - it is the single and only reason for all of the controls and lockdowns.

Right now definite information about omicron is just emerging. There are three big questions. The first is whether it is more infectious than delta and if so by how much. The second is whether it causes severe illness in those infected at a similar rate to delta, a higher rate or lower rate - is it more virulent than delta, about the same, or milder. The third is whether it evades the effect of vaccines and if so by how much. The evidence is coming in and you can also look at aggregated prediction sites such as Metaculus to get an idea.

So what is the emerging picture, granting that it is still very uncertain?

It is already pretty clear that it is much, much more infectious than delta, with a doubling time of 2-3 days. That is more infectious by a whole order of magnitude. Now that it is clearly loose everywhere around the world it will rapidly supplant delta and will also spread throughout almost every population very, very quickly. A lot of the press and media coverage will focus on the higher infectiousness but we should ignore this as by itself that doesn't matter.

The second question is severity.

Here the evidence is still conflicted. Metaculus gives an even money (50%) chance that it is milder than delta, possibly much milder. There is though a 25% chance that it is more severe. This is what the government is waiting for or should be waiting for. In Metaculus the shift is away from 'as severe/more severe' to 'milder'.

The third question is whether it can evade vaccine protection against serious illness and if so to what degree.

Here it is starting to become clear that having two shots of the main vaccines does not provide as much protection as against delta (although still a lot more than if you are not vaccinated at all). Having a booster seems to mean you are still protected to the same degree. Again, this is early, we will know more certainly in a couple of weeks.

The nightmare would be for omicron to be simultaneously more infectious, more severe, and highly vaccine resistant. This would be a disaster and would put us back where we were in January 2019 but worse. Fortunately it looks as though that very unlikely event is not happening. What is more likely is a combination of the three variables that is not good but not disastrous, e.g. more infectious, same severity, slightly weaker vaccine protection or more infectious, milder, weaker vaccine protection.

In the most likely combinations the rational thing to do once things are clear is to reverse the temporary plan B but really ramp up the vaccine booster programme.

Two things to add.

If it turns out that omicron is much more infectious and also clearly milder then this is good news as it means that the virus will spread very rapidly through the population and give most people natural immunity, without there being too many serious cases. On the other side, if it is more infectious and as severe as delta but with slightly reduced vaccine protection (these two combinations look the most likely) then given the rate at which it spreads we can expect a lot of cases among the 30% who are currently unvaccinated (but mitigated by the number of those who have had it already). This will make the next couple of months really tricky. The thing is, with the level of infectiousness we are talking about I don't think anything is going to even slow down the spread significantly. We really have to hope the mean prediction at Metaculus is correct.

The longer term lesson from not just this but the whole of the last two years is that everyone everywhere has to rethink how to organise the healthcare system, given that it's the demonstrated vulnerability of the systems we have in most of the world that has made a bad event into a disastrous one.