Sunday, 29 March 2020

How America Dies

US “conservative” values that prioritise economy over lives are risking both.

Since 9 March 2020, over 90% of new US COVID-19 cases have occurred over the previous 10 days.

Stop and think about that. EVERY DAY, count back 10 days and the total new cases is over 90% of the total EVER reported in the US (up to that day).

This is why US hospitals will saturate. This is also why the death rate doesn't look super high (1.8% of cases as reported by 28 March). Most of the people who are going to die are still in hospital.

The numbers are falling slightly off an exponential trend but not enough to matter – each time I get a new data point, it pushes my projections out a day or two.

I wrote a paper describing this but doubt it will get to peer review in time to matter.

A scenario: about 10% of COVID-19 patients have to be admitted to hospital to survive – this number could be out a bit but that just shifts the day of reckoning slightly.

The US hits 3-million cases on 8 April, resulting in a need of at least 300,000 hospital beds. The country has about 1-million. Allow 30% surge capacity and you may be able to admit 300,000 patients. But you are less than a week out from there from needing all 1-million hospital beds – currently projected for 12–13 April – see first chart. So at that point, all COVID-19 patients are sent home to die.

What if my numbers are a bit out – the number of hospital beds is double, the fraction needing hospitalisation is lower, etc.? Those make little difference with such a rapid growth rate.

It only ends when either you have infected everyone or the magic of herd immunity kicks in. We don't really know what that level is –  it seems the kids are unlikely to get it badly and could become the core of the immune herd.

As a what-if, the second chart shows how long it takes to reach the entire US population. It will obviously stop long before we get there because we know that many people don’t get it seriously enough to know and therefore are not included in the stats. The point of this chart is to show how rapidly an exponential grows – the next step after 24 April is nearly 400-million.

The only thing that could slow it significantly is if those who had it mildly and are over it are a large fraction of the population. Unfortunately we have no idea what this number is – estimates from various parts of the world vary widely. A likely reason: testing standards and coverage vary a lot.

Children could be a significant component so let’s focus on them.

The last US Census was in 2010; that put the fraction under 18 as 24%. This fraction declined from 25.7% of the last census but gives us some idea.

But here’s the problem: many children could have been infected and not know it; they could be a significant part of the reason it spreads so fast.

That leads to first step of a solution; the rest follows from that.
  • Take as many kids as possible out of circulation and put them in isolated places with a few adults who have had it and been cured or have it so mildly that they are not at risk.
    • This is not so contrary to US culture: you call this Summer Camp.
    • Monitor each camp for any cases that turn serious.
  • Next step: lock down the whole country to prevent further spread so all available resources can go to the real hotspots. This would require making food and meds available to the poor and anyone without the means to go without income for 3 weeks, the minimum you need to stop the spread seriously.
  • Meanwhile you are growing a population of people with immunity – the kids and their camp counselors. They and others who recovered can increasingly be drawn into voluntary and paid work to ease the lockdowns.
  • In any areas where numbers are under control, progressively unlock them but be prepared to lock down again as needed.
  • If at any time a vaccine or cure is discovered, this can all be scaled back. A reliable test to determine immunity from prior exposure – e.g. measuring antibodies – would also be useful but the priority research in the short term has to be testing for the virus efficiently and finding either or both vaccine and cure.
Since the required response is a massive all of society intervention, the right will likely veto this as “socialist” or an attack on civil liberties.

I find this civil liberties argument less than compelling when people are prepared to sacrifice millions of their own people for an economic goal. This is the language of a brutal dictator like Stalin, not the language of free people.

Millions will die and mostly without even basic hospital care, let alone ventilators or ICUs. It is very hard to see the US coming out of this as a viable society.

That is what “conservative” values brings you. If only the right was dying, you could say it is a valid choice but they are taking out a lot of people who don't support their death cult.

Numbers here.


Philip Machanick said...


The US growth trend will stop at some point when enough people who had it mildly have immunity. So far there is no sign of that – they are less than 2 weeks out from collapsing their hospital infrastructure.

All they can hope for is the big unknown: how many asymptomatic or unreported mild cases are there really?

WHO suggests 80%, which is good – it limits growth to 20% of the population. If that is true, the US case count should level off at some point before 66-million. Which actually is not so good – that means most who need hospitalisation to recover will die. On this, the WHO report is less sanguine: it estimates 15% as severe, requiring oxygen, and 5% requiring ventilation.

My article, so as not to appear too alarmist, puts the % who die without hospital intervention at 10%. That would mean about 6-million US deaths; at 20% it becomes, about 12-million.

Philip Machanick said...

On second thoughts, the WHO numbers are for everyone who catches it including the asymptomatic and mild cases.

So the upper limit on deaths is more like 20% of the US population – about 65-million.