Stanford study: Up to 81,000 Santa Clara County residents have been infected with COVID-19

The study was led by Dr. Eran Bendavid, associate professor of medicine at Stanford and senior fellow at the Woods Institute. Stanford photo.

Daily Post Staff Writer

Stanford researchers think about 2.5% to 4.2% of Santa Clara County residents, or 48,000 to 81,000 of the county’s 1.9 million people, have been infected with COVID-19.

Santa Clara County has reported 1,870 cases and 73 deaths. The county’s case numbers are understood to be an undercount of the real number of infections because of a national testing shortage.

The new statistics are part of a study led by Dr. Eran Bendavid, an associate professor of Primary Care and Population Health at Stanford Medicine.

Bendavid and his team collected blood samples from 3,330 county residents who volunteered to participate in the study. By analyzing the prevalence of COVID-19 antibodies in the blood samples the researchers came to the conclusion that the real number of infections is 50 to 85 times higher than what the county has reported.

Similar to San Mateo County estimate

Those numbers are similar to an estimation San Mateo County’s Health Officer Dr. Scott Morrow made earlier this week. Morrow said 2-3% of San Mateo County’s 768,808 residents have probably had the virus and recovered. He said the region’s stay-at-home order probably prevented the disease from spreading faster.

People who have already had COVID-19 might have protection against getting the disease again because of the antibodies they develop. No one knows yet whether that immunity is permanent.

The Stanford researchers used Facebook ads to recruit test subjects from different areas and backgrounds. They attempted to adjust results for race, Zip code and sex to make the sample better reflect the broader population.

More sick people signed up?

Natalie Dean, an assistant biostatistics professor at the University of Florida, said on Twitter that the study might have had a selection bias where people who thought they might be sick would be more likely to sign up.

The study results are part of a pre-print research paper, meaning the paper has been put online but has not been peer-reviewed by other medical experts.

Bendavid and Dr. Jay Bhattacharya, another Stanford professor and one of the paper’s co-authors, published an op-ed in the Wall Street Journal last month saying estimates which put the death toll for COVID-19 at 2-4% might be too high because infections have been undercounted.

They said more information about its spread in the population is needed to know how many people who contract the virus will die. They said the death toll of the country might be closer to 20,000 than 2 million.

So far 33,049 Americans have died from COVID-19, but the disease is still spreading.

Certain populations, such as older people and those with other medical conditions, are more at risk for dying from COVID-19.

In the absence of widespread testing, hospitalization numbers can also shed some light on how serious the situation is getting. California has seen less hospitalizations and deaths compared to other places like New York City.

Local hospitalizations

There are 187 people currently hospitalized for COVID-19 in Santa Clara County. There are 71 people hospitalized due to COVID-19 in San Mateo County.


  1. What about selection bias? The researchers invited people to be tested through Facebook ads. Wouldn’t that lead to an oversample of people who fear they’ve been exposed to Covid-19? And I don’t know how one would correct for that like you would if you have over samples or undersamples in other categories.

  2. I just saw on C-SPAN the Governor of New York bragging that his state appears to have “flattened the curve” of infection such that one infected person may only infect one other. At that rate, everyone will be infected eventually and many will die terrible deaths (in costly ICUs). Few live with respirators. So the government is buying time. For what? Capacity to hospitalize, for one thing. But unless we now concede that millions will die horrible deaths, there should be an all-out war to develop vaccines and treatments – not just for the virus but for thousands of others that could “jump” or be deliberately transferred from animals to the human race.

  3. Your story ignores an important piece of the puzzle — the true death rate of Covid-19. For weeks the scientists at WHO and the CDC have claimed the death rate was 3.4%. Then they revised it to 1.4%. This study says it’s 0.2%, about the same as the flu.

  4. Instead of a lockdown we need to have people mingling again so we can develop herd immunity. With a death rate below that of the flu, we should reopen the country and let people develop an immunity like we’ve done with previous flu outbreaks. Our leaders have made a bad decision with this lockdown.

Comments are closed.