Without decent epidemiological data any public health policy with an eye on containment or mitigation is simply a shot in the dark. Unfortunately, materials, equipment and lab bottlenecks have made testing numbers plateau at this level. Multi-day delays in receiving test results are widespread.
. . .The United States will be the country that has suffered the greatest recorded deaths from coronavirus within two weeks, and the death toll certainly will not peak within a week of that dire landmark.
Yesterday, the White House briefing gave out a forecast for a range of deaths between 100,000 and 240,000. With a population of 330 million, this works out to a death rate per million of 300 to 720. Just for benchmarks, Italy and Spain are currently at 200 deaths per million. New York state is at 80. All other U.S. states and major countries are below those numbers. But to be clear, I don’t find those benchmarks helpful in forecasting going forward.
Currently, for our nation as a whole the number of deaths per million stands at 12. To stay within the White House range, the number of deaths must double either 5 or 6 more times. If it only doubles 4 times, then we fall below the range. More than 6 times, and deaths will exceed the upper end of that range.
Most recently, the rate has been doubling every three days. How much would you want to bet on a forecast that between now and the end of the year it will double no less than 5 times and no more than 6 times? The people who are trying to issue forecasts, including the sainted Dr. Fauci, come across to me as fools. He may get lucky. But if you held a gun to my head and made me forecast, then I would give a wider range, particularly at the upper end.
There’s an obvious institutional interest here in estimating the worst case scenario. It both scares people into compliance and makes them look good if the death rate ends up being lower.
And, this is a situation where, ultimately, I think folks have a metaphorical gun pointed at their heads, so you can’t afford to not forecast either. It’s necessary because the longer we stay locked down, the more people will die unnecessarily based on that.
That’s the biggest problem I see that needs to be pointed out to the folks who say that opening the economy too early is just about killing off grandma to keep the profits flowing. It’s not just a matter of money. At this (minimum) level of economic activity, there are going to end up being lots of otherwise preventable deaths from all sorts of things.
– People put off getting their brakes fixed and get killed in a fatal car crash.
– People getting depressed and killing themselves
– Financial stress leading to physical stress
– People putting off routine medical appoints that would have detected and cancers and other diseases while they were still treatable.
The list is really pretty endless once you start thinking it out, and it needs to be simply and flatly stated there’s no avoiding a tradeoff. Lockdown is currently the life-maximizing solution, but in the long-run lockdown is a bigger life-taker.
The models out there all suck, but an improvement would be addressing this fact.
Beyond that, the models need to be transparent as to what they include and don’t.
IF people are locked down, there will be fewer car wreck deaths. Also, if the healthcare system is overloaded, then more people die from non-covid causes because an overloaded emergency room delivers poor care to everyone.
Also, if you try to open the economy too soon, many people will refuse to go to work and some who fill forced tow work will shoot their idiot boss rather than endanger themselves.
There is no clever or easy solution. The only solution now is for everyone to work hard to stop the spread. Yet, everyday we read news articles about people being stupid such as the University of Texas students who insisted on going on spring break and are now sick or Liberty University.
“There’s an obvious institutional interest here in estimating the worst case scenario. It both scares people into compliance and makes them look good if the death rate ends up being lower.”
The evidence from the governments behavior so far is that they consistently downplay and underestimate whenever possible. Is there a reason why you expect this to change?
“the government” is not a discrete body.
I think you’re right that folks like the FDA and CDC seem to consistently downplay. On the other hand, if take the pronouncements of public facing folks, like the estimates of 200,000 dead as kind of an upper bound, so that, when their turn out to be only 50,000 dead, it looks like they did a relatively good job.
The trajectory in the US and Europe seems to be: initially “downplay and underestimate”, then when that turns out to be wrong, swing strongly in the opposite direction.
My analysis suggests that the number is going to double 5.497 times between now and the end of the year. Not 5.496. Not 5.498. 5.497. And my number has 3 decimal places of precision. So you can feel pretty good about this estimate.
We reach equilibrium in a petty chaotic fashion. New York might be much farther along in the peak with a wider population of immune.
Consider the following unit of spread:
One bridge player playing one day in a national tournament will spread the virus to 10%.
Then the problem is boiled down to finding the number of equivalents bridge tournaments in NYC. For example, there were 150 people entered to play bridge. 150 people are two car loads on a transit car, each car taking a one hour ride.But your super spreaders are also doormen, taxi drivers, and family members, so we have to double the count. The super spreader needs two days to do the same spread a bridge player did at the tournament, New York is packed, so was the tournament. So I estimate the tournament twice as packed as everyday NYC life.
In New York the spread rate is high, hat constant is high The first super spreader died on Mar 3, four weeks ago. Pick that constant, you will see that at least 30% of NYC has the virus.
The CDC reports that the prevalence of obesity in the US was 42.4% in 2017~2018.
Some of these people are struggling to breathe as it is.
So I spent the month of March commuting daily from an apartment in Queens to the Perlmutter Cancer Center on E. 34th St. in Manhattan. For the first three weeks I took the subway every weekday back and forth. (During the fourth week I drove my car.) Upon arrival I was surrounded by a bevy of elderly, immuno-compromised cancer patients, and attended to (often at close quarters) by employees who also took the subway every day to work. Sometimes I wore a mask and gloves, and sometimes I didn’t. (The facility required me to remove the gloves upon arrival.)
By all rights I should have caught the virus by now. But it’s now two weeks since I last rode the subway and nearly a week since I last visited the Center–I remain symptom-free. So I conclude that the disease is not as contagious as the panic-mongers (e.g., ABC News) suggest. For that reason alone I tend to distrust the high-end fatality estimates.
Further, there is little on this blog (or many others) that mentions possible treatments for the virus. In particular, I do think one can be optimistic about Trump’s suggested hydroxy-chloroquine, if only because it’s already available in large quantities. Of course it’s not yet “proven” (in the Fauci-an sense) that it works, but the anecdotal evidence seems overwhelming. The drug won’t prevent the spread of the virus, but it could substantially lower the death rate.
There are a number of other drugs that may also work, though they’re not as easily procured. Either way, extrapolating “exponential growth” to the end of time is as fallacious as assuming linear growth, or a simple-minded, Hannity-like model.
It won’t happen that way. Things will change. People will adapt. The curve will flatten. It isn’t as bad as it looks now. The S&P 500 will not go to zero.