Asymptomatic Deep-Dive

Read on, you’ll be surprised

This NYT article revealed that high Ct PCR was identifying mostly non-infectious positives (trace viral RNA)

The upside to supersensitive PCR is that it also tells us a lot about asymptomatics

1 https://www.nytimes.com/2020/08/29/health/coronavirus-testing.html
Case 5: Tyson Foods, Missouri: 85% asymptomatic of 291+

Note - Ct means cycle threshold or how many cycles of PCR you run to amplify RNA. Low 30’s is normal and picks up normal infections. 40 is high and can amplify trace amounts of RNA

5 https://apnews.com/d3646ba688a79eba78db3bcabee80b24
Case 8: Meat Processing Plants: 9,000+ employees across the plants were mass tested -> 1/3 PCR+ -> 95% Asymptomatic

Source: Direct contact with top management

8
Case 10: Karachi, Pakistan:

“Aga Khan University's researchers have discovered in a study that 95% of those who had tested positive for Covid-19 through blood tests, which register the presence of antibodies to fight the disease, were asymptomatic.”

10 https://tribune.com.pk/story/2262980/1 
These aren’t scientific studies so they have limitations

Also need to factor in ages, health, who is tested

But this official survery of care homes in the UK shows that:

“80.9% of residents who tested positive were asymptomatic (5,455 out of 6,747)”

13 https://www.gov.uk/government/publications/vivaldi-1-coronavirus-covid-19-care-homes-study-report/vivaldi-1-covid-19-care-homes-study-report
There are also studies/reports that show that asymptomatics are far lowerer

Yes, but there are two major confounders which come into play with this type of study

WAPO article:
https://www.washingtonpost.com/health/2020/08/08/asymptomatic-coronavirus-covid/
14
Confounders:
1. AB testing misses many asymptomatics who serorevert quite fast
2. Ct count for PCR - the more cycles, the more likely you are to pick up trace RNA from low level imfections (asymps). Lower Ct would pick up far fewer

Methodology also important - who u testing?

15
Some of these studies are quite old and when I first saw them I could not make sense of them

When enough stuff wasn’t making sense I then realized that maybe the testing is letting us down

AB studies greatly underestimate number of infected/recovered

Moment of awakening:

16 https://twitter.com/gummibear737/status/1297495211308965888
So seroprevalence is mostly a fail, although if you notice in tweet 10 (Karachi), they found 95% asymps with AB testing - must have been recent infections

High cycle PCR on the other hand is able to amplify trace amounts of RNA and detect previous infections

17
CDC still estimates asymptomatics at 35% because they are going by AB testing studies

Mass testing with high Ct PCR shows that they are much larger

So much larger that it changes the denominator for calculating IFR

18
In SIR modeling used to predict epidemiological evolution, R stands for recovered

This type of model can help calculate herd immunity

But for some reason, everybody is using seroprevalence as the correct measure of recovered

I disagree -> an asymptomatic is recovered

19
In fact, to me the high numbers of asymptomatics is how we know that pre-existing immunity is real

How else could we explain so many people brushing off infection with the “novel” coronavirus

For most people, it’s just another coronavirus...no big deal

20
So ultimately, it’s all about trying to identify an accurate denominator

If we know how many have been infected, we can accurately calculate IFR

We can also do a better job at identifying herd immunity

People say NYC can’t be at herd immunity with only 20% seroprevalence

21
Unlike some EU countries which are able to find new cases (although most are asymp), NY can’t

They are testing like crazy...can’t find cases

1% = false positive threshold

Maybe a stable 20% seroprevalence is indicative of a much higher number of recovered?

See this thread

22 https://twitter.com/gummibear737/status/1305967263405821954
I think the data on asymptomatics sheds an important light on Covid-19
-Not nearly as deadly as thought
-Significant pre-existing immunity
-We are much closer to herd immunity than anybody thinks
-Lockdowns needlessly put vulnerables at risk (protect vulnerable only solution)

23
If this deep dive was easy for you to understand, then have a look at this deep dive into estimating NYC’s real number of recovered

24 https://twitter.com/gummibear737/status/1305215126379917314
That’s it

Sorry it was so long

If you learned something useful please RT first tweet in this thread

Keep Calm and Carry On
You can follow @gummibear737.
Tip: mention @twtextapp on a Twitter thread with the keyword “unroll” to get a link to it.

Latest Threads Unrolled: