It’s been almost three weeks since I crunched the Covid-19 numbers for the US, California, and Monterey County, and that’s been enough for clear trends to become apparent.
- In none of the locales does the growth continue to be exponential.
- Nationwide, the daily new cases and new deaths have both dropped significantly from their peaks of several weeks ago. That’s mostly because of New York, New Jersey, and Massachusetts. New York changed its Covid-19 death criteria about 5 weeks ago, which made the reported peak higher than it would have otherwise been. Elsewhere, there is growth or flatness to the curves.
- In California, we have been on a slowly rising ramp for some time, with the number of new cases about 2000 per day, and a plateau in deaths, with about 75 per day.
- The Monterey county statistics are a little harder to understand because the numbers are small and there’s more noise in the system, but the number of cases per day is flat to rising.
- Testing continues to grow slowly both nationwide, in the state, and locally. IT’s going to be hard to parse the national numbers now that the CDC has decided to lump viral testing and serologic antibody testing into the same big bucket. I hope the state and local authorities don’t decide to do that.
Now that we’re out of the exponential growth phase of the epidemic, I’ve replaced most of my semi-logrithmic charts with linear ones.
Here are the cases and deaths nationally.
The dots are the daily totals, measured at noon PDT. The gray line is a 7-day moving average.
The 6000+ deaths in the middle of April needs some explication. That was the day that New York changed the rules for deciding what was a Covid-19 death to include cases where the death was probably due to the disease, but with no viral testing done. They made that retroactive, and that meant that it looked like a lot of people dies that day who had actually dies previously.
The death rate has been falling, but remains high:
This drop is probably mostly due to more testing and the easing of overcrowded facilities in New York.
The reason the number of cases continues to trend upwards while the number of deaths is flat is probably better testing. However, this is not what an epidemic in control looks like.
The California death rate is not as bad as the national rate:
That may be do to more testing, or it might be due to better care. Probably both. In any event, unless there are huge numbers of Covid-19 cases that we don’t know about, this is a deadly disease. By the way, it looks like on the order of half the Covid-19 victims are completely asymptomatic, and even with ready access to (non random) testing, are unlikely to ever be in the denominator of the above curve.
We have some fairly good historical data on California testing from the CDPH:
There have been about 1.5 million (viral, I believe) tests in California since this started, and the number of tests per day has gradually risen to about 50,000.
The percentage of the California tests that are positive is now under 5%. Repeated tests on the same person are counted as different tests in these stats. I have no idea how many of the tests are of health care workers, and how many are of patients.
Monterey County is frustrating to keep track of, because they keep revising their numbers. Here is this morning’s report.
With all these numbers, it’s impossible for me to guess how much of this is behind us. I have seen credible reports saying that R0 (number of people infected by an average Covid-19 victim) is about 2.6. It looks like with al the social distancing that we’ve managed we’ve driven the R down to about 1 or maybe a little worse, depending on whether you look at national or state and local numbers. Above 1, the number of cases per day continues to grow until herd immunity slows that growth. Herd immunity in this case is achieved with about 60% infections, assuming that recovered victims have complete, lasting immunity, which is no slam dunk. If the virus doesn’t mutate its way around it, a vaccine is the other solution to the problem. But this looks like a long slog to me.
We need massive random serelogic testing to figure out how many people have actually had Covid-19. That will allow us to calculate the fatality rate. We need better treatment options. It looks like ventilators are not the magic bullet that people thought they were at the beginning of this — New York’s fatality rate for ventilated patients as well north of 80%. Some antiviral drugs look promising, and there are a lot of smart people working on this, albeit without much guidance on a national level, at least as far as I can see. We will have to find out how to combine effective social distancing with keeping the economy running and feeding and housing people. For that, we need to know a lot more than we do about transmission mechanisms.
A lot of moving parts, and a lot of unknowns.