Note on updates: As more supporting data for my theory comes in, I continue to update this post. You may wish to bookmark it so you have a convenient reference.
Last update 7/222/2020.
Or so I’ve been maintaining. I based that on the Diamond Princess case where 17.1% of the ship population tested positive, but more than half were asymptomatic. Only 7 people died, 1.1% of the cases.
Back when Worldometer was still reporting mild vs. serious/critical cases numbers, only about 5% were serious. (Added: They’re still reporting it on the world page: 3% “Serious or Critical”.)
On April 22, there were 717,008 active US cases, but only 58,173 hospitalizations (8.1% of active cases), and 15,341 in ICU (2.1%). For the vast majority of people it’s no big deal.* Apparently most of those who test positive are asymptomatic (see Diamond Princess) or have the equivalent of a mild cold.
I also noted how quickly known/tested cases were popping up well away from known, large COVID-19 clusters.
To me, all that implied that the virus was already widespread well before the first known cases. Before lockdowns started.
I have some data to support that, now that researchers are doing antibody testing (checking to see if a person was exposed and developed an immune reaction).
- Massachusetts: 30% of people with no symptoms tested positive. Admittedly, it was a small sample: 200.
- Santa Clara County, California: Between 2.5% and 4.2%. “These prevalence estimates represent a range between 48,000 and 81,000 people infected in Santa Clara County by early April, 50 (to) 85-fold more than the number of confirmed cases.” They sampled more than 3,000 people.
- Los Angeles County: 4.1%, “suggesting the rate of infection may be 40 times higher than the number of confirmed cases.” 863 samples.
- And then there’s New York state. They tested around 3,000 people in 19 counties. 13.9%.
- Added, 4/26/2020: Miami-Dade: 6% test positive for antibodies; “about 165,000 estimated infections”.
Show of hands: Who believes that all those people went through the roughly 4-5 week process of contracting COVID-19, getting “sick” (asymptomatic or mild), developing antibodies, and got better all since lockdowns started? New York only locked down March 20.
I strongly suspect that the lockdowns did not even “flatten the curve.” They happened too late.
Fortunately, there’s a way to test whether or not I’m correct. Georgia’s Governor Kemp is lifting most lockdown restrictions effective tomorrow. If I am wrong, and the curve was flattened, then in about two weeks — call it May 8 — the state should see a large surge in new cases and hospitalizations. (The delay is the incubation period.)
If I am correct, then we may see a minor bump in a continuing downward trend. Watch here. You can ridicule me and say, “I toldja so,” or congratulate me, as the case may be.
Results: And the winner is… No one. Georgia changed testing and reporting methods, so you can no longer tell how many new daily cases there are.
More supporting data:
Added: One more data point; USS Theodore Roosevelt. The Navy tested 100% of the crew. A few tests are pending, but of the 4,938 completed:
- 4,098 negative
- 840 positive
- 9 sailors had been hospitalized
- 1 sailor in ICU
- 1 death
17% tested positive (hey, like Diamond Princess), and 1% of the infected needed hospitalization. 0.1% died. Although I’ll note that active duty Navy personnel are younger and healthier, in general, than the large number of retirees (i.e.- more vulnerable due to age) on the Diamond Princess, so fewer deaths on the Roosevelt are unsurprising.
Added, 5/1/2020: New data point supporting my theory: No COVID-19 Spike from Wisconsin’s In-Person Voting
Added, 5/1/2020: And another. That’s out of 890 workers so far. More than 40% of the workforce at Tyson Foods pork-processing plant in Indiana has tested positive for coronavirus, NBC News reports.
Added, 5/16/2010: Yet another data point supporting my theory that SARS-CoV-2 was already widespread much earlier than has been generally recognized.
Antibody testing places earliest Ohio coronavirus case in early January
The earliest coronavirus cases in Ohio now date back to January, indicating COVID-19 might have been in the state and spreading here earlier than initially thought.
Six people have reported feeling ill in January – as early as Jan. 7 – according to Ohio Department of Health data released Sunday. Few details about the patients were available Monday.
That’s two weeks earlier than what had been believed to be the first US case on January 21, and more than eight weeks prior to what had been thought to be Ohio’s first case.
Data point, 5/28/2020: Half the population may already have significant resistance to COVID-19, even if not exposed. “Importantly, we detected SARS-CoV-2−reactive CD4 + T cells in ~40-60% of unexposed individuals, suggesting cross-reactive T cell recognition between circulating ‘common cold’ coronaviruses and SARS-CoV-2.”
This doesn’t surprise me a bit. SARS-CoV-2 is just another coronavirus. Corona viruses are responsible, along with rhinoviruses, for the common cold. This would go far in explaining why most people testing positive for COVID-19 are asymptomatic (or very mild cases), while a few go bad quickly. The ones that go bad may be in the half or so that don’t already have these T cells.
Data point, 5/28/2020: COVID-19 apparently in the US in December 2019: A woman got sick 2 days after Christmas (meaning she was exposed a week or two prior). Classic COVID-19 symptoms, but no one had even heard of it yet. She hadn’t traveled. Months later, a COVID-19 antibody test was positive. This indicates the virus was already spreading in the US in mid-December last year.
Added, 6/19/2020: And the hits just keep coming:
Coronavirus was in northern Italy in December, officials reveal after studying wastewater
SARS-Cov-2 RNA (ribonucleic acid) was found in samples collected in Milan and Turin on Dec. 18, 2019, and in Bologna on Jan. 29, 2020.
Added, 6/23/2020: More than 8.7 million Americans with COVID-19 went undiagnosed in March, study says
Now, new research says the number of COVID-19 cases in March may have been 80 times greater than what original estimates revealed, amounting to more than 8.7 million new cases in the U.S. that health officials and the public never knew existed.
The researchers blame testing issues, high false-negative rates and asymptomatic spreaders for the “under-counting of the true prevalence of SARS-CoV-2,” according to the study.
The only surprise to me is that it was that much more widespread than even I estimated back in March. Admittedly, I didn’t find out until later that we had December 2019 and early January 2020 cases, already spread across the country. I expect to learn of more as antibody testing expands, and more old waste water samples are tested.
Datapoint, 6/25/202: Worldometer reports 2,487,638 total US cases, but…
US health officials believe 20 million Americans have had the coronavirus
U.S. officials believe as many as 20 million Americans have contracted the coronavirus, suggesting millions had the virus and never knew it.
Twenty million infections would mean about 6% of the nation’s 331 million people have been infected, leaving a majority of the population still susceptible to the virus.
I think that estimate is way too low. In April, New York state antibody testing showed 13.9% positive; a rate more than twice as high as what the feds are claiming now, which would have extrapolated to 2,703,550 cases in that state alone, two months ago. About the same time a small Massachusetts study found 30%. Currently, Georgia — with a much lower known infection rate than those states — is running a mere 5.9%.
Datapoint, 7/22/2020: By April 1, New York City had over a half a million people with ChinCOVID antibodies. Statewide, they only knew of 85,050 cases.
* Yes, I know it’s really a big deal for that comparatively small number of vulnerable folks whose cases do go bad. Those vulnerable people should minimize human contact, just as they should in flu season. For the exact same reason.