usaar333

36 ms·

511.
by usaar333 7y ago
> But it depends on having a very large available pool of rapid testing Not necessarily. You can just aggressively quarantine - basically the strategy in China. > so you can't do it when you have thousands of cases per day At wha
512.
by usaar333 7y ago
https://www.wolframalpha.com/input/?i=china+population+%2F+s... Actually worse than China by that metric, though given credibility issues of Chinese data, let's just say about the same.
513.
by usaar333 7y ago
They had really good contact tracing. I'm not even sure how much we're attempting this in the US. "Lockdowns" without contact tracing is taking the path of "herd immunity for essential workers"
514.
by usaar333 7y ago
Once you drop the exceptional case of the NY metro area, not really. What makes Washington, Illinois, and Michigan so bad compared to Texas, Virginia or California?
515.
by usaar333 7y ago
1. Germany has nearly 3x the population so that difference is less stark than you are implying. 2. Maybe they haven't broken 200k tests because there isn't an obvious population to test when you've had an order of magnitude f
516.
by usaar333 7y ago
I'm not the first person to have thought about group variations. :) Here's an old paper (which also highlights how this important fact is omitted in a lot of analysis): http://web.eve.ucdavis.edu/sschreiber/re
517.
by usaar333 7y ago
The republican stronghold of Texas has one of the lowest covid deaths per capita in the United States, better than even California (which is doing relatively well). Deaths per capita at 16/million is lower than almost anywhere in West
518.
by usaar333 7y ago
> I can't help but wondering if the unlimited outside time under the guise of "exercise", but which people are clearly taking advantage of by going to hang out with friends in the park, has anything to do with it. Or perh
519.
by usaar333 7y ago
Correct. Testing is only a good measurement with massive testing, even randomized testing. Iceland's data is probably the closest to actual infections and many states probably look similar. ( https://www.covid.is/data )
520.
by usaar333 7y ago
Most of those increases are artifacts of testing parameters changing or reporting delays, not actual infection rates. WA data here is quite misleading.
521.
by usaar333 7y ago
Great visualizations - I'd love to see this done with hospitalizations though as case count overestimates r given increasing testing. (Granted hospitalizations are also breaking down recently due to recent disproportionate nursing home
522.
by usaar333 7y ago
> start spreading again when the masses break and go back to work I don't see why this must be true. R before the lockdown in northern CA and WA (heavy voluntary measures but people "working") was ~1.45. [source: https:&
523.
by usaar333 7y ago
Agreed that US as a whole can have a high IFR. Santa Clara county though is one of the healthiest places in the US. Life expectancy of 84 exceeds Italy in fact.
524.
by usaar333 7y ago
We do need to be careful talking about "IFR" as with a disease with such high age-dependence demographic skews heavily alter population IFR. Castiglione d'Adda has more people over 70 than Santa Clara County has over 65; popu
525.
by usaar333 7y ago
We have randomized PCR studies and ships where everyone got tested. Yes, there is error there (false negatives), but there's an upper bound to how much there can be.
526.
by usaar333 7y ago
It's impossible to believe. Currently, Santa Clara's crude CFR is 3.7%. NYC has similar demographics (and as bad nursing home hits?) -- 0.14% of the population has died from covid. That puts an upper bound of 26x (and that'
527.
by usaar333 7y ago
Unless you think FB users for some reason would be infected at significantly different rates, that shouldn't be much of a problem. I'm much more concerned about selection bias toward prior sick people; IIRC, Stanford offered to re
528.
by usaar333 7y ago
CDC flu surveillance data suffers from even more selection bias than this study. It's basically useless. Iceland PCR data is reliable but it is suggestive of an IFR of more like 0.3%. Except Iceland also managed to successfully isolate
529.
by usaar333 7y ago
Correct. For some things you could use baseline data, but there is zero baseline data to use here.
530.
by usaar333 7y ago
It is far more credible that Wuhan hit 3% infection rate (that's about double the PCR testing rate of the evacuees) than Santa Clara county, with an order of magnitude lower death count, having a 2% rate.
531.
by usaar333 7y ago
NYC has 30x the per capita death rate of Santa Clara county. Santa Clara county only being 15% as infected is not consistent.
532.
by usaar333 7y ago
Perhaps they can't get statistical significance. You'd ballpark a 0.5% infection rate in the Bay Area going by deaths and current best guesses for IFR. Your false positive rate might exceed that. These tests are better done in hig
533.
by usaar333 7y ago
No it's just a sign that they aren't comfortable running a full on controlled experiment yet. Only 2 deaths out of 113 severe in a week is far better than baseline. There's also a recently released monkey study showing huge
534.
by usaar333 7y ago
We also don't Shelter in place for flu. Bay Area school closure date is 1 day before SIP. Sweden and Iceland never closed schools for young kids. Data doesn't look any different.
535.
by usaar333 7y ago
Fair enough, but would there be a reason it would be different? And even if it were, is that actually relevant? Anyway, here's some IFR estimations for covid by Imperial college: https://www.thelancet.com/journals/
536.
by usaar333 7y ago
But can you not compare the ratios between age groups? That's all I'm doing here - showing deaths have a higher age correlation with covid than flu. - not making any statement about absolute CFR/IFR.
537.
by usaar333 7y ago
I don't think that's true. covid has stronger age dependence. Just look at the ratio of covid vs. influenza deaths on the CDC's website ( https://www.cdc.gov/nchs/nvss/vsrr/COVID19/index.h
538.
by usaar333 7y ago
whoops, sorry it's 567 infections in passengers with 13 deaths. 2.3% CFR. Expected IFR 3% from Imperial College (17)
539.
by usaar333 7y ago
I'm extrapolating from hospitalization rates for that reason. Otherwise, you'd get a 0.45% CFR in Iceland. Looking at the raw data, I don't think Korea ever had below 3% using deaths/recoveries. Iceland is at 0.75% righ
540.
by usaar333 7y ago
hence why I divided by the passenger, not total, count

More ›