COVID-19

[Jin+Guice]

Anyone hearing about the studies emerging in Cali (Stanford and maybe others) that, based on the presence of antibodies in random testing, supposedly suggest that the exposure rate has been 25X-80X the positive test result rate (meaning a denominator that's maybe almost two orders of magnitude too small has been used).

What kind of R0 does this imply assuming the disease 1) started spreading when we thought it did (around the time of first case detection in each local) and 2) started spreading in November?

If the disease has been spreading faster than originally assumed (or for longer than originally assumed), why didn't we see the spike in deaths until March-April?

How different is an estimated exposure rate of 25x-80x current positive test results from the estimate that 100x current deaths equals the actual number infected 17 days ago?


Thanks to those who responded to my two longer posts, you helped clarify some things for me, though I'm still not sure whether I think this is as bad as I thought originally or no where near as bad as I thought originally.

[jacob]

https://www.nytimes.com/interactive/202 ... eaths.html (contains graphs)

Looking at overdeaths relative to historical rates form all causes from ~Mar10-Apr10 for various countries makes it possible see the impact beyond insufficient testing (e.g. untested deaths outside the hospital) numbers. So basically, this is the increase in the funeral business this past month relative to BAU.

Summary:
New York City+298%
Spain+66%
Jakarta+36%
England&Wales+33%
Netherlands+33%
Istanbul+29%
France+28%
Belgium+25%
Switzerland+21%
Sweden+12%

Heart disease accounts for most deaths at 22% followed by cancer at 21%. Influenza and pneumonia is normally responsible for 1.9% of all deaths.

[HalfCent]

x

[Bankai]

@jacob: interesting, however, it's worth remembering that only some of these extra deaths are from COVID-19 with a substantial number of additional deaths being likely the result of the lockdown (see below).

Another interesting thing is that Sweden with no lockdown in place has by far the lowest rise in mortality from the countries you listed.

[George the original one]

LA County

It's frustrating that few, if any sites, publish the rates of new cases, new deaths. Most only publish the totals or daily numbers without a history.

So I dug around and found that California has the county numbers for April in a CSV file on the state site. Again, the new cases and new deaths were daily totals, but at least there was a daily history, so it was easy to derive the weekly rate. Weekly rate smooths out the daily bumps, which makes spotting trends easier.

Results:
1) LA County new cases peaked around April 8, have declined until April 15, but now are increasing at a rate above the earlier peak. Current new case rate is ~3800 new cases per week and can be compared to California overall at ~9500 per week. Note that California's overall rate has increased by ~1500 per week from the low 5 days ago, of which ~1000 per week is due to LA County's rise!
2) Deaths are on the decline for the past couple days, but only because deaths lag new cases. It seems that the lag is about 12 days for LA County. Current death rate is 257 per week. This is about half of California's overall death rate of ~500 per week.

[Jason]

"Excess" deaths? So like businesses, some deaths are essential and some non-essential? And exactly what are the essential deaths? (Don't tell me, I know.) You want to discuss death rates, that's fine. But utilizing qualitative language as a graph descriptor is a dead (see what I did there) giveaway of a narrative intent. Just give the numbers. We get it, old people and the comorbids are the ones predominantly dying. But damn, this movement towards "life years" gives me the heebie jeebies. Like Corona is assisted suicide. Slippery slope shit that everyone should be wary of. What happens if it was mainly killing babies and kids? It's ok because they don't have jobs and it cuts down on household budgets. Conditional morality is just that. Conditional. You want to use it. Fine. Just don't bitch when you're the one dying on that sword.

[George the original one]

I'm still betting that immunity works, if it doesn't we should go back to BAU anyways, because we're never getting rid of this thing unless the entire world bans all car and air traffic. Even if a country could stamp it out, they'd get reinfected on the next inbound flight. If immunity doesn't work we're all doomed and this is the new normal, might as well enjoy it and get back to living life normally again.

BAU is not normal when you have death stalking. Adaptive measures certainly work. Wouldn't you rather have South Korea's normal rather than Sweden's under these conditions?

[2Birds1Stone]

@Augustus, not only did they not close it, they limited the schedule such that the reduced number of riders was cramped into less frequent rides and ultimately reduced any benefit of less people having to physically go to work (at least for the commute portion of their day).

[ZAFCorrection]

@Jason

The feds could switch paper types and someone somewhere would probably die from the decision without needing to squint too hard to see the causality (e.g. infected paper cut). People die from policy decisions, even the innocuous ones. It's the nature of outcomes with non-zero probability being applied across a bajillion people. Being an amoral bean counter about everything may be a slippery slope, but that's the only slope where any work would get done. Or maybe everyone is just living blissfully unaware of higher-order effects.

Maybe the solution is to multiply the number of people killed by an "obviousness" discount factor. Like if the government drone strikes some dude (x1), they better be pretty sure he was gonna kill at least 5 people (x0.2).

[Ego]

Interesting NY Times article from NY ER doc. On covid-induced pneumonia:

Goes on to recommend pulse oximeters and talk about breathing exercises that can help early-stage Covid-19 patients from progressing into needing ventilator.

Thanks for posting this. Covid is much easier to treat if caught early. I've been using my pulse oximeter ever day to see if there have been any changes. Also, loss of taste/smell has been present in 80% of the asymptomatic cases and some have what they are calling Covid toes.

[bryan]

Some auto insurance policies (Allstate, USAA, ...) giving refunds.. waiting to see what mine will be (I already pay very little..).

I always thought they were batshit crazy for letting the subways run. Those seem as bad as preschools and airplanes for disease transmission. COVID spreads by breathing in exhalant right? What percentage of the population uses subways? 10%? Seems like in a cramped box with recirculated air you'd get almost every single one of those people infected right?

Folks have been saying (not sure what the latest is) the virus is in "droplets" i.e. bigger/heavier than typical atmosphere, by some degree. However, yeah.. I've "heard" that not many cases have been found to be from "outside" contact; mostly "indoors" contact. Reminds me of CO2 meter reading differences between outside, small rooms w/ just you, rooms with more people.. makes you hypothesize that if you put 100k people into a stadium to watch a football match, most of any spread that might happen would be from the bathrooms or areas otherwise not well-ventilated. So yeah, I would expect a subway, bus (w/ closed windows), etc. to be pretty bad.

Maybe we should keep most bars or clubs closed, but open up the roof-top bars and beer gardens?

My GF's temporary contract got canceled two weeks early. Now she's eligible for $600 week + normal unemployment through July or something crazy like that. Normally she wouldn't have even qualified for unemployment. I'd like to get a taste of that type of deal myself!

Gotta admit, this is the kind of thing that makes me a bit furious... not with you or your girlfriend... but with the entire program.

Some perversion w/ that $600 for sure. Also in some of the other bits..

There are definitely businesses and charities out there that applied for, and got, the PPP loans who need it way less than others. I know one (that shall remain nameless) that will be getting >$500k and their income/revenue has actually slightly improved (tbd if that lasts..). Their plan is to keep it in a separate account and not touch it, as much as possible in the short-term (long-term they may put it toward growth).

[George the original one]

BAU is what is usual. If immunity doesn't work as @bigato is speculating, this is the new usual. This is what life is now, and it will never go away. You might as well enjoy it. If this is never going away, and immunity doesn't last, SK isn't immune either (every plane, boat, tourist, etc will bring in a new wave), everyone will get this, over and over again, for the rest of our lives. Unless you want to be a hermit.

The rate of infection is everything. South Korea's weekly infection rate is 137... gonna take a long time for everyone to get it in a country of 51 million and they're more or less BAU. 372,262 weeks to infect the entire population. At their peak, they had only 2,360 cases/week or 21,610 weeks to infect the entire population.

Compared to Sweden's voluntary lockdown's weekly infection rate of 3,829 for a country of 10 million. The entire population would be infected in only 2,611 weeks at that rate.

Finally USA with a leaky lockdown weekly infection rate of 203,707 for a country of 330 million. 1,619 weeks to get us all at that rate... yeah, we're worse than Sweden!

So pick your poison. South Korea seems very effective, Sweden not so much, and here in the USA, I'll stay a hermit.

Edit: Italy's peak rate translates to 1,526 weeks, so USA is pretty close to the Italians.

[Ego]

Many many pages ago, I calculated the average years of life expectancy lost to be 6 weeks for males and 8 weeks for females.

https://www.nber.org/papers/w26992#fromrss

Using daily state-level coronavirus data and a synthetic control research design, we find that California’s statewide SIPO reduced COVID-19 cases by 152,443 to 230,113 and COVID-19 deaths by 1,940 to 4,951 during the first three weeks following its enactment. Conservative back of the envelope calculations suggest that there were approximately 2 to 4 job losses per coronavirus case averted and 108 to 275 jobs losses per life saved during this short-run post-treatment period.

2.5 - 6.5 jobs were lost for each additional DAY of life for a man who didn't die because we locked down California.

[Campitor]

Prior to the quarantine my company barred all in-person gatherings but hadn't given the general order to work remotely for those who could. I asked management why in-person meetings were considered a risk but riding a crowded train was not. I also asked when they would issue the general order to work from home. It took them 2 weeks to grant telecommuting privileges after barring the in-person gatherings.

I completely understand that upper management had some legal and logistical considerations to review but the level of logical inconsistency was upsetting to many people. It's similar to the warnings to avoid using masks despite their efficacy against infection. Anyone with a bit of common sense realized that recommendation was horseshit.

Recently we've discussed when onsite work would resume. The current assessment is that it will be "safe" once the quarantines are lifted. I replied that the quarantines are about avoiding surges that impact the healthcare system; the virus will remain potentially lethal even after the quarantine is lifted.

The silent pause was noticeable and I could tell the likelihood of COVID's continued virulence wasn't even considered; not everyone will die in April/May - we just get to die in intermittent phases throughout the year.

The lack of critical thinking is either intentional (suppressing information to calm the sheeple) or natural (low level thinking required during high level threats).

I hope everyone is safe and doing well.

[jacob]

2.5 - 6.5 jobs were lost for each additional DAY of life for a man who didn't die because we locked down California.

As the prime minister of Albania said [paraphrase], "We can bring back jobs. We can not bring back the dead."

Also, to compare apples to apples, we should compare days of living lost with days of working lost. The lock down has lasted 20-40 days now, so it's more like 50-260 days of work lost for each day of life gained.

[George the original one]

One to file in the unexpected and unintended consequences...

People accidentally poisoned by household cleaners spike since stay-home order

The Washington Poison Center reports a 23 percent jump in the number of people accidentally getting poisoned by household cleaners and disinfectants.

https://katu.com/news/coronavirus/peopl ... home-order

[fiby41]

Use of Chinese rapid antibody testing kits suspended after complaints of faulty results.

Mumbai might have been NYC had the local trains not been stopped. They come under western, central, harbour divisions which comes under the ministry of railways which is a portfolio of the central government.
Buses, which comes under electric supply and transportation undertaking coming under municipal corporation that comes under state government, are running but you've to sit one seat removed. 721 containment zones sealed. Only if we were a dictatorship, we could solder their doors from outside Wuhan-style... Meanwhile in Asia's largest slum schools are being turned into quarantine for close contacts and sports complex for positives. Stadiums being used for this purpose in other cities. Burial of infected within municipal limit prohibited. Dedicated hospital built last month on the outskirts has started receiving occupants.

[Jason]

The Washington Poison Center reports a 23 percent jump in the number of people accidentally getting poisoned by household cleaners and disinfectants.

lol@accidental. How about a bunch of spouses fed up with smelling each other's farts for a month on end. Keep this up and there will be more deaths by Clorox than Corona.

[Jin+Guice]

A lot of these numbers are suspect.

Using the NYC deaths/ total population is an interesting and clever way to determine the "minimum death rate," but how sensitive is it to NYC reporting presumptive cases and possible other data flaws?

Looking at extra deaths for a region is also interesting but (as @Bankai) pointed out, how many of those deaths are due to people not seeking medical cares? OTOH, how many fewer deaths are we seeing from things like traffic accidents and the flu?

Are job losses being determined by unemployment numbers? The unemployment rules are super lax right now and the feds were encouraging tons of people apply. If you're furloughed but have a job waiting for you, you are eligible for unemployment temporarily (this is basically true for me). OTOH, I think a lot of companies are paying employees to retain them but not necessarily working them, if they can afford to (this is true for all of my co-workers).

I know everyone is just working with what they've got, but the ambiguity surrounding the numbers makes almost all analysis inconclusive (not encouraging y'all to stop though).

[BeyondtheWrap]

@Jin+Guice: NYC reports confirmed and presumptive deaths separately, so you are free to consider only the confirmed if you prefer.

Confirmed: 9,562
Probable: 4,865

Confirmed means they tested positive. Probable means they did not test positive but their death certificate lists Covid-19 as cause of death.