COVID-19 daily, 11th May 2020

This will be the last edition of this newsletter, as I'm moving on to other projects. Thanks for subscribing!

Some resources:
• When can we test everyone? - this site will remain updated
• covidcharts.com - fast, US-specific charts
• endcoronavirus.org uses a similar framing of pre- and post-growth countries; also by US state
• rt.live - US viral reproduction rates
• covid-19.direct - good US county-level data when you pick by state
• ourworldindata.org - an excellent resource for any world trend, but particularly good coronavirus info

Best wishes to you and your families.

Country plots

#masks4all and our footnote [4].

Suggestions? You can reply to this email.

permalink of today's copy

World stats

Totals

4,101,704 cases — 2,410,005 active — 282,714 deaths

Lockdown / stay-at-home orders

Country counts

Countries with 100+ cases: 150
Countries with 1,000+ cases: 93
Countries with 10,000+ cases: 44

Growth countries: 164
Post-growth countries: 24

Antibody studies

* implied percentage of population infected
** infection fatality rate, not just cases, but all infected

these studies are useful in aggregate, each study has flaws

Top countries by growth

* for comparison only [1]

Top countries by cases

* for comparison only [1]

US breakout

Totals

1,329,260 cases — 1,033,565 active — 79,526 deaths

Case trend (week over week, day by day):

Are we bending the curve: YES??

1.18x ⟶ 1.18x ⟶ 1.16x ⟶ 1.16x ⟶ 1.15x

Mortality trend (week over week, day by day) [5]:

Are we improving treatment: STASIS?

1.20x ⟶ 1.20x ⟶ 1.19x ⟶ 1.19x ⟶ 1.17x

Testing:

9.0M tests, +3% since yesterday, 14.7% positive [2]

Whole US testing (330M) possible in ~99 days, mid August, at average 4% daily increase in capacity.

For comparison:

	# of tests	per million people*
US	8,987,524	27,189
S. Korea	668,492	13,041
Italy	1,702,283	28,145
Iceland	54,213	159,105

* unknown how many are re-tests

Top states:

Infections in early & exponential-growth countries

projected weeks:

* change in growth, i.e. 2nd derivative

Infections in post-exponential countries (24)

Andorra, Australia, Austria, China, Croatia, Cyprus, Estonia, Greece, Hong Kong, Iceland, Israel, Jordan, Lebanon, Luxembourg, Mauritius, New Zealand, Norway, S. Korea, Slovakia, Slovenia, Switzerland, Taiwan, Thailand, Vietnam

projected weeks:

* change in growth, i.e. 2nd derivative

[1] The real case fatality rate (CFR) can only be determined after the fact; however, it can be narrowed when new cases and case resolutions reach parity, and the growth rate becomes linear.

This has happened in S. Korea (though noise persists); as of March 24th, we can narrow S. Korea CFR to 1.33% - 3.31%, favoring the low side.

[2] High positivity rate probably means under-testing. As of March 22:

• S. Korea: 2.7%
• UK: 7%
• Australia: 1%

[3] Tomas Pueyo: The Hammer and the Dance -- Skip to section 4.

[4] #masks4all:

• Short & sweet: Jeremy Howard on quick masks from t-shirt
• Longer rationale (Howard): We need #masks4all
• Non-video comprehensive summary, sections 5 and 6 for building and maintaining
• Do-it-yourself face masks (Hong Kong)
• DIY (India)
• Comparison of common materials
• 2006 CDC on household options

[5] If we discover a treatment cocktail, we would expect to see the death rate start to fall prior to the case load. If it doesn't, the death rate will continue rising beyond case falloff by 3-4 weeks.