Thread by @DFisman

Global R(t) has been remarkably stable since the first wave. Here it is plotted against global doubling time. Will try to unpack this further when I get a chance but it’s a good news/bad news story: many infections ahead, but this isn’t open-ended.

“Moving average” = 7 day moving average for doubling time.

Just to unpack this a bit, there's a direct relationship between R and doubling time, inasmuch as R(t) is a function of growth rate, as is doubling time. Doubling time may just be a bit more intuitive.

We can see in the graph that both R and doubling time have actually changed little since April.

We are currently around 42 million COUNTED cases globally; let's say we're under-estimating infections by a factor of 20...probably about right.

https://twitter.com/DFisman/status/1262141885986082817?s=20

That would mean that the 42 million counted cases represent around 800 million infections. If doubling time remains around 75 days (as a function of this on again, off again, institution of distancing followed by relaxation of measures)...

we're at 1.6 billion by mid-January, and around 3.2 billion by March. That's likely getting to the point where an uncontrolled epidemic would peak due to approaching the critical fraction of susceptibles to sustain exponential growth.

This assumes R of 2-3 with that critical fraction being a bit lower due to the non-random nature of population mixing.
That can still result in a very large future stream of cases due to high prevalence, so the pandemic doesn't screech to a halt.

But the dynamic is likely to change as our R effective is near 1 even without intervention. The longer term future will likely result some kind of endemic equilibrium with "covd season" as we now have flu season, unless we have a potent vaccine that makes elimination viable.

But hopefully it's helpful for folks to understand that this isn't open-ended. And the timeline I describe, while very approximate, would look very much like historical pandemics, like 1918-19.

I know that billions of infections by early 2021 sounds implausible, but that's the wonder of exponential growth. We're currently well above 400,000 counted cases globally each day. At 5-10% case identification fraction, that's 4-8 million infections...

it adds up!

Meanwhile, the United States is approaching 230,000 counted COVID deaths in a 6 month period. In terms of deaths, that's 4 Vietnam Wars, and the Vietnam War lasted 19 years.

Yes, thanks everyone. My bad on this tweet, where the "deaths" I referred to were US military casualties. As many have pointed out there were millions of deaths that resulted from conflict in S.E. Asia over multiple decades.