Thread by @ProfColinDavis

(THREAD) Here is my understanding of the UK's Covid situation, and why we need a circuit breaker NOW.

Whether we call it a circuit breaker or a lockdown, some dramatic cessation of normal activity is inevitable fairly soon. If we just carry on at the present rate, we should expect to be up to around 2000 deaths a day by the end of the year.

In fact, that’s probably optimistic, as the collapse of a number of local hospital systems by that point will mean a greatly increased case fatality rate and/or an increase in ordinarily preventable non-COVID deaths. It’s hard to predict exactly when it’ll happen, but ...

there must surely come some point when the public will not accept this level of mortality, and will demand that the government imposes the necessary measures to reduce it. Any delay in commencing this cessation will only extend the time for which it needs to run.

My understanding is that the goal of a circuit break is to get transmission down to the point where contact tracing can be effective. As such, it can be distinguished from longer lockdowns that seek to eliminate community transmission altogether, a la New Zealand or Victoria.

People often refer to circuit breakers as being for a fixed duration (e.g., 2-3 weeks). But the stopping point needs to be epidemiological rather than temporal. Perhaps it could’ve been only 2-3 weeks had we implemented it when Sage recommended, but now it’ll need to be longer.

Contact tracing hasn’t been going well, but it’s probably not reasonable to expect it to be able to cope with the volume of cases we currently have. The virus is out of control. Around 20,000 cases a day and an R rate of ~1.5 means that ...

there are just too many contacts for the system to handle (even a “world beating” system like ours). If we had only 1000 cases a day, for example (as was true back in August), it’s feasible that the tracers could be contacting more than 90% of those who need to self isolate.

In addition to its epidemiological effects, a circuit break could have powerful psychological effects. It would send a clear message that This Is Serious Now, which is critical for improving compliance with self isolation (without which our contact tracing system is pointless).

And a national circuit breaker might help to restore some of the “We’re all in it together” community spirit that was evident during the initial lockdown, which is also essential for compliance (this is one reason why I’m sceptical about regional approaches).

The lockdown until #ZeroCOVID option is still my preference, for what it’s worth. That option would be very costly (we’d need to make the rich pay for it!), although I suspect it’d still be cheaper in the long run, and would save many more lives. One can only look at NZ with envy

Either way, the current approach of tinkering at the margins and squabbling over paying for the necessary public health measures is wasting precious time. It will impose a terrible cost in lives, in addition to its economic cost.

A couple of addenda prompted by comments I’ve received. I’ll use graphs to make both points. One of them is quite positive, the other rather less so.

I’ll start with the gloomy one. When I said “around 2000 deaths a day by the end of the year”, that wasn’t a figure plucked from the air to be alarmist. It was the most optimistic I could be. Deaths per day is growing exponentially: plotted on a log scale it looks like this.

That trend line running through the data has an R^2 of 94%. If we stay on that line we'll get to 2000 deaths a day BEFORE December.

Anyone who thinks that there’s some reason why deaths won’t continue to follow the trendline on a log plot should go back and take a look at the plots that @bristoliver has been making for about 2 months of deaths in the NW.

https://twitter.com/BristOliver/status/1316431703213309960?s=20

But here’s a positive graph, showing progress on track and trace. Prior to mid-August there was never a week in which tracers contacted >5000 people within 24 hrs. Since then there’s been a huge increase: tracers are now beating the 24 hr contact threshold for >5000 people a day.

That increase hasn’t been enough to keep pace with the spread of infection, obviously. But with a circuit break we could get to a point where the caseload is manageable.