Thread by @AtomsksSanakan, 1/EVarious southeast Asian nations suffered relatively few COVID-19 deaths per capita, especially [...]

1/E

Various southeast Asian nations suffered relatively few COVID-19 deaths per capita, especially in comparison to many "western" nations.

There's been a lot of speculation on why this is.
So this thread will examine some possible explanations.

https://archive.is/FkAho

2/E

There are at least 3 types of explanation for what's occurring in various southeast Asian countries:

1) insufficient testing that misses many infections and/or misses many COVID-19 deaths
2) lower number of infections
3) lower proportion of infected people die of COVID-19

3/E

For explanation 1:
It's unlikely their testing misses more deaths, since their excess deaths don't outpace their reported COVID-19 deaths more than in many 'western' countries.

https://www.nytimes.com/interactive/2020/04/21/world/coronavirus-missing-deaths.html
https://www.bbc.com/news/world-53073046
https://www.economist.com/graphic-detail/2020/07/15/tracking-covid-19-excess-deaths-across-countries

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7852240/

4/E

Insufficient PCR testing, or antigen testing, early on might still miss past infections. Antibody testing addresses that; this will be discussed more a bit later.

https://twitter.com/AtomsksSanakan/status/1347304761415135232

https://archive.is/sjUkT

https://twitter.com/AtomsksSanakan/status/1347304761415135232

5/E

For explanation #3:
IFR (infection fatality rate) is the proportion of SARS-CoV-2-infected people who die of COVID-19. So explanation #3 states IFR is lower in southeast Asian nations.

Some proposed reasons for why this could be:

https://twitter.com/AskeladdenTX/status/1309633580008644609

https://twitter.com/AskeladdenTX/status/1309633580008644609

6/E

The co-morbidities explanation seems less plausible, since IFR is driven more by age and sex than by co-morbidities.

IFR increases with age, as do co-morbidities, while the rate of various co-morbidities differs by sex.

https://link.springer.com/article/10.1007/s10654-020-00698-1

7/E

South Korea + Japan are useful test cases here, since they have relatively old populations with less obesity than in many "western" nations.

So explanation #3 would predict those 2 nations to have lower IFRs.

https://www.populationpyramid.net/world/2019/

https://ourworldindata.org/obesity

8/E

Yet South Korea's IFR is comparable to western nations in spring/summer, regardless of whether one uses antibody testing or PCR-based modeling to estimate the number of infections.

[~0.6% August IFR: https://arxiv.org/pdf/2101.11991.pdf]
https://twitter.com/AtomsksSanakan/status/1312562674656522240

https://link.springer.com/article/10.1007/s10654-020-00698-1

9/E

Japan's IFR is on the higher-end of what's seen in western nations.
[ https://twitter.com/AtomsksSanakan/status/1343866026879053824]

https://twitter.com/AtomsksSanakan/status/1327772908819779584
[with: https://wwwnc.cdc.gov/eid/article/27/2/20-4088_article]

And China's IFR is comparable to that of western nations.

https://www.thelancet.com/journals/lanmic/article/PIIS2666-5247(20)30099-9/fulltext
[with: https://www.thelancet.com/journals/lanwpc/article/PIIS2666-6065(21)00003-1/fulltext]

https://twitter.com/AtomsksSanakan/status/1343866026879053824

10/E

So explanation #3 is likely false.

That leaves explanation #2:
As per part 4/E, antibody testing on representative samples allows one to determine the number of infected people, without the limitations of PCR + antigen testing from explanation #1 https://twitter.com/AtomsksSanakan/status/1341185258964836353

https://twitter.com/AtomsksSanakan/status/1341185258964836353

11/E

Japan's antibody-based infection rates were ~1% or less

https://onlinelibrary.wiley.com/doi/full/10.1002/jgf2.408

https://www.mhlw.go.jp/stf/seisakunitsuite/bunya/0000121431_00132.html
https://www.fukushihoken.metro.tokyo.lg.jp/iryo/kansen/koutaikensa.html
http://www.pref.osaka.lg.jp/hodo/index.php?site=fumin&pageId=39934
https://www.pref.miyagi.jp/uploaded/attachment/823102.pdf
https://www.pref.aichi.jp/site/covid19-aichi/koutaikekka.html
https://www.pref.fukuoka.lg.jp/press-release/press-release-corona-koutai.html
https://www.pref.hiroshima.lg.jp/site/2019-ncov/message20201030.html

https://wwwnc.cdc.gov/eid/article/27/2/20-4088_article

12/E

Low antibody-based infection rates (i.e. seroprevalence) continued into autumn 2020 in Japan, as per 11/E, + in South Korea:
https://arxiv.org/abs/2101.11991

China's seroprevalence was highest in Wuhan in spring, and <1% elsewhere:

https://www.clinicalmicrobiologyandinfection.com/article/S1198-743X(20)30598-X/fulltext

https://www.thelancet.com/journals/lanwpc/article/PIIS2666-6065(21)00003-1/fulltext

13/E

Western nations had much higher infection rates, with their hardest-hit regions often having rates 1 or 2 orders of magnitude more than in the three southeast Asian nations noted before

https://www.istat.it/it/files//2020/08/ReportPrimiRisultatiIndagineSiero.pdf

[ https://www.mscbs.gob.es/gabinetePrensa/notaPrensa/pdf/15.12151220163348113.pdf]
https://portalcne.isciii.es/enecovid19/informes/informe_final.pdf

14/E

So explanation #2 is most likely correct.

https://www.ukbiobank.ac.uk/media/x0nd5sul/ukb_serologystudy_report_revised_6months_jan21.pdf

https://www.amjmed.com/article/S0002-9343(20)30909-8/fulltext
https://www.medrxiv.org/content/10.1101/2020.10.11.20210922v1.full?versioned=true
https://www.uoflnews.com/post/uofltoday/covid-19-rates-continue-to-increase-with-highest-rates-among-18-34-year-olds/

https://www.medrxiv.org/content/10.1101/2020.07.01.20143966v1.full

https://www.medrxiv.org/content/10.1101/2020.11.02.20221309v1

[ https://www.medrxiv.org/content/10.1101/2020.12.16.20248180v1.full-text
https://www.thelancet.com/journals/laninf/article/PIIS1473-3099(21)00054-2/fulltext]
https://www.corona-immunitas.ch/en/

15/E

It's unlikely that prior infection with other coronaviruses explains this (see part 5/E), since:

- innate immunity does not improve substantially
- T cells do not primarily limit infections, beyond impacting on B cells

https://twitter.com/profshanecrotty/status/1309170532965920769

https://twitter.com/AtomsksSanakan/status/1309311345947484166

https://twitter.com/profshanecrotty/status/1309170532965920769

16/E

Similarly, this likely is not explained by cross-reactive B cells, plasma cells, and/or antibodies from prior infection with other coronaviruses.

https://twitter.com/AtomsksSanakan/status/1312654302742183936
https://twitter.com/AtomsksSanakan/status/891040491214688257

https://www.nature.com/articles/s41591-020-1083-1
https://www.medrxiv.org/content/10.1101/2020.06.29.20142596v1

https://www.medrxiv.org/content/10.1101/2020.08.14.20173393v1

https://twitter.com/AtomsksSanakan/status/1312654302742183936

17/E

Many of those offering these alternative explanations do so for ideological reasons and/or in bad faith, because they want to avoid an obvious answer:

Policies + behaviors in southeast Asia limited infections

20:42 - 25:52 :

18/E

So instead of making excuses about how southeast Asian nations have special exposure to other coronaviruses, or better immune systems, or..., we in western nations should own up to the reality that their policies + practices worked.

Deal with it.

https://twitter.com/AtomsksSanakan/status/1325442309668921351

https://twitter.com/AtomsksSanakan/status/1325442309668921351

18/ESo instead of making excuses about how southeast Asian nations have special exposure to other coronaviruses, or better immune systems, or..., we in western nations should own up to the reality that their policies + practices worked.Deal with it. https://twitter.com/AtomsksSanakan/status/1325442309668921351

19/E

And h/t to @KawasakiKR11 for help in finding seroprevalence studies for Japan in part 11/E, since I only read and write in English. https://twitter.com/AtomsksSanakan/status/1364470684949299203

https://twitter.com/AtomsksSanakan/status/1364470684949299203

20/E

Sources on infection-limiting strategies in southeast Asia:

(act early:
https://link.springer.com/article/10.1007/s10640-020-00466-5
https://academic.oup.com/cid/article/71/12/3174/5866094)

https://twitter.com/AtomsksSanakan/status/1364599742257364995

https://web.archive.org/web/20210224184118/https://www.mhlw.go.jp/content/10900000/000635891.pdf
https://journals.sagepub.com/doi/full/10.1177/0275074020943707
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7567433/
https://www.thelancet.com/article/S0140-6736(20)32007-9/fulltext

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7189844/

21/E

Examples of what I critiqued from those using a Tokyo study w/ non-representative sampling [ https://www.twitter.com/AtomsksSanakan/status/1365177417023774720]:

https://www.twitter.com/carlheneghan/status/1309558504307544064
https://www.twitter.com/MonicaGandhi9/status/1335747616148127745
https://www.twitter.com/dwallacewells/status/1308855749821554689
https://www.twitter.com/AlexBerenson/status/1308798932097658881
https://www.twitter.com/MLevitt_NP2013/status/1308979101215096834

Nic Lewis:
https://judithcurry.com/2020/10/14/t-cell-cross-reactivity-and-the-herd-immunity-threshold/

https://www.twitter.com/AtomsksSanakan/status/1365177417023774720

22/E

Further context on why prior infection with other coronaviruses does not adequately explain low COVID-19 deaths per capita in southeast Asian countries, especially in light of low seroprevalence:

https://twitter.com/AtomsksSanakan/status/1364660714342932487
https://twitter.com/AtomsksSanakan/status/1364664008830246913

https://twitter.com/AtomsksSanakan/status/1365870324861067266

https://twitter.com/AtomsksSanakan/status/1364660714342932487

23/E

Below John Ioannidis defends a failed version of explanation #3 in February 2021. He under-estimated IFR in southeast Asia, as covered in parts 8/E + 9/E, along with other threads.

https://twitter.com/AtomsksSanakan/status/1369420295187144709
https://twitter.com/AtomsksSanakan/status/1341286778884550656

3:03 and 16:05 :

https://twitter.com/AtomsksSanakan/status/1369420295187144709

24/E

Re: "Ioannidis defends a failed version of explanation #3 [...]. He under-estimated IFR in southeast Asia"

Update on Asian seroprevalence studies:
https://www.thelancet.com/journals/lancet/article/PIIS0140-6736(21)00238-5/fulltext
https://www.mhlw.go.jp/content/000761671.pdf
https://twitter.com/KawasakiKR11/status/1379530595148181507

He's still doing it:
https://twitter.com/AtomsksSanakan/status/1376029923903672322

25/E

Ioannidis, March 2021:
"I'm not talking about [New Zealand, Vietnam, Taiwan] that did a wonderful job and did exactly what I was saying a year ago. Test, test, test. That's what the WHO was saying. And then isolate, then contact trace"

31:35 - 31:47

26/E

Basic summary of the thread:

Southeast Asian countries successfully limited COVID-19 deaths predominately by pursuing policies + behaviors that prevented infections.
It's not a matter of them having better immune systems, special pre-exposure to other viruses, etc.

27/E

Based on this thread, one would predict that Malaysia's NHMS results (likely to be released in May/June) would show <2% of people infected, with an IFR of ≥0.4%.

http://iku.moh.gov.my/nhms
https://twitter.com/AtomsksSanakan/status/1360114634662019073

https://ourworldindata.org/explorers/coronavirus-data-explorer?zoomToSelection=true&pickerSort=asc&pickerMetric=location&Metric=Confirmed+deaths&Interval=Cumulative&Relative+to+Population=true&Align+outbreaks=false&country=KOR~JPN~CHN~VNM~SGP~PHL~EuropeanUnion~AUS~NZL~THA~MYS