[GoodLife]

The unadjusted prevalence of antibodies to SARS-CoV-2 in Santa Clara County was 1.5% (exact binomial 95CI 1.11-1.97%), and the population-weighted prevalence was 2.81% (95CI 2.24-3.37%). Under the three scenarios for test performance characteristics, the population prevalence of COVID-19 in Santa Clara ranged from 2.49% (95CI 1.80-3.17%) to 4.16% (2.58-5.70%). These prevalence estimates represent a range between 48,000 and 81,000 people infected in Santa Clara County by early April, 50-85 fold more than the number of confirmed cases.

The population prevalence of SARS-CoV-2 antibodies in Santa Clara County implies that the infection is much more widespread than indicated by the number of confirmed cases. Population prevalence estimates can now be used to calibrate epidemic and mortality projections.

https://www.medrxiv.org/content/10.1...463v1.full.pdf

[blueash]

This is a well done study. The finding that the incidence is 50 to 85 times greater than known from positive COVID PCR reports is tempered by a few caveats.

The study did not ask whether the participants believe they were sick with COVID but had not been tested. When you ask people to volunteer for this kind of study you might, might, get volunteers who suspect they were infected and are looking for validation. This would get you a higher number of positives in what you are hoping is a randomly selected representation of your population.

A factor that would give you a lower number would be that only healthy people can make it to your screening site. If I'm home in bed with fever and cough but have not been COVID tested, you're not getting my sample. This becomes a bigger issue if there are many ill people at home.

They are using a not approved device to detect antibodies. They did run some known samples thru the device. It did a good job with not identifying persons as having COVID if they were negative. [They used old blood samples from before the virus appeared thus they had to be COVID negative.] The machine manufacturer did the same, with good results.

On the other hand, the test was good but not great at correctly identifying patients who had been diagnosed with PCR testing as having been ill with COVID. It missed about 2 out of 10 calling those people negative. Several things could cause this:

1. The positive PCR could be wrong. They never had COVID thus shouldn't have antibodies.
2. They had COVID but it is too soon for antibodies to appear. This is unlikely as IgM appears in a couple days.
3. The patient did not make enough antibodies. Normal people vary in level of antibody production. It does not reliably depend on how sick you were with the disease. But any test has a cut off level of detection. It calls you negative if your antibody measure is below a selected value.

The authors did adjustments to correct for unbalanced bias in their population tested. It was excessively white and female for the balance in the county. All in all, a good contribution to having data about population spread in the absence of COVID PCR test positivity.

Keep in mind as we are nearing relaxation on the measures to slow community spread. This county is one of the highest impacted in California. This testing only suggests immunity in 2 to 4 percent of the population. We are nowhere near herd immunity which is usually set at 80% immune.

[chet2020]

Originally Posted by blueash

This is a well done study. The finding that the incidence is 50 to 85 times greater than known from positive COVID PCR reports is tempered by a few caveats.

The study did not ask whether the participants believe they were sick with COVID but had not been tested. When you ask people to volunteer for this kind of study you might, might, get volunteers who suspect they were infected and are looking for validation. This would get you a higher number of positives in what you are hoping is a randomly selected representation of your population.

A factor that would give you a lower number would be that only healthy people can make it to your screening site. If I'm home in bed with fever and cough but have not been COVID tested, you're not getting my sample. This becomes a bigger issue if there are many ill people at home.

They are using a not approved device to detect antibodies. They did run some known samples thru the device. It did a good job with not identifying persons as having COVID if they were negative. [They used old blood samples from before the virus appeared thus they had to be COVID negative.] The machine manufacturer did the same, with good results.

On the other hand, the test was good but not great at correctly identifying patients who had been diagnosed with PCR testing as having been ill with COVID. It missed about 2 out of 10 calling those people negative. Several things could cause this:

1. The positive PCR could be wrong. They never had COVID thus shouldn't have antibodies.
2. They had COVID but it is too soon for antibodies to appear. This is unlikely as IgM appears in a couple days.
3. The patient did not make enough antibodies. Normal people vary in level of antibody production. It does not reliably depend on how sick you were with the disease. But any test has a cut off level of detection. It calls you negative if your antibody measure is below a selected value.

The authors did adjustments to correct for unbalanced bias in their population tested. It was excessively white and female for the balance in the county. All in all, a good contribution to having data about population spread in the absence of COVID PCR test positivity.

Keep in mind as we are nearing relaxation on the measures to slow community spread. This county is one of the highest impacted in California. This testing only suggests immunity in 2 to 4 percent of the population. We are nowhere near herd immunity which is usually set at 80% immune.

I've been waiting for this Stanford study data. I agree with your criticisms, but finally, some serology test data. We need much more. I've been using the Gangelt Germany data for my own extrapolations for way too long. Bummer about the exposure rate. I was hoping for higher. However, if you extrapolate the Santa Clara data to the NYC area using death rate to exposure ratios, a rough calculation gives 65%+ exposure for Queens NY, for example. Overall though, yes, we aren't even close to herd immunity.

[Altavia]

Antibody tests suggest that coronavirus infections vastly exceed official counts