Antibody study finds coronavirus infections may have been 10 times higher in Bay Area – San Francisco Chronicle

Nearly 10 times as many Bay Area residents had been infected with the coronavirus by the end of April than the official tally at the time, according to a new federal study that analyzed antibody tests to determine how widespread the virus was across a handful of United States hot spots.

The study underscores just how deficient testing for the virus was in the early weeks of the pandemic, when the vast majority of cases were never identified.

At the same time, it provides further evidence that aggressive shelter-in-place orders protected much of the Bay Area, where researchers estimate only about 1% of all residents had been infected by the time the study was done. That number is surely higher now with the outbreak surging again.

Scientists with the Centers for Disease Control and Prevention estimate that 65,000 people in the Bay Area had been infected with the coronavirus by April 27, based on results of antibody tests conducted on a sample of the population. During that period, only about 7,000 cases had been reported,

The CDC performed antibody tests on about 16,000 blood samples collected in 10 regions in the United States, including the Bay Area, in March, April and May. Results showed that the Bay Area had the lowest infection rate in the country during that time period, compared to 6.9% in New York City, which had the highest rate.

The study, results of which were published in JAMA Internal Medicine, is the largest look so far at the estimated infection rate in several of the countrys COVID-19 hot spots. As with previous studies, it shows how widespread the virus is in American communities and how much testing for active infections has fallen short, according to the authors and infectious disease experts.

This study is consistent with what everyones been thinking: The number of cases reported are a gross under-estimation of the actual number of cases occurring, said John Swartzberg, an infectious disease expert with UC Berkeley.

The 1% infection rate in the Bay Area certainly has climbed since the antibody tests were done at the end of April, infectious disease experts said. But the percentage of people who have been infected is probably still very small overall potentially 2% to 6%, based on the number of confirmed cases now compared to two months ago. The CDC has not yet done a follow-up study in the Bay Area but has plans to do so, the authors said.

The regions remarkably low infection rate, especially compared to places like New York City, demonstrates the success of sheltering in place for two months, infectious disease experts said. When the antibody tests were done, the Bay Area had been under stay-at-home orders for five weeks and case counts and hospitalizations were at their lowest.

In fact, though the Bay Area was one of the first parts of the country to experience community spread of disease, it had the lowest overall infection rate, according to the CDC study.

This is what we managed to do, said George Rutherford, an infectious disease expert with UCSF. Even with the current surge in cases in the Bay Area, he guesses the percentage of people who have been infected remains small.

Youre only talking about hundreds of cases a day in a region with millions of people in it, he said.

New York City had the highest infection rate, at close to 7% at the end of March; a follow-up study there found that the rate had jumped to about 23% by the first week of May. Washington state, which was also an early hot spot, had an infection rate of 1.1%.

About 5.8% of people had been infected in Louisiana at the time the antibody testing was done, and 4.9% in Connecticut, which was wrapped up in New Yorks outbreak. About 1.9% of people in South Florida had been infected by early April; that climbed to 2.9% about two weeks later, and infectious disease experts said the percentage is likely much higher now, as cases have swelled there in recent weeks.

The CDC study looked for coronavirus antibodies in blood samples taken from people during routine screening, such as cholesterol tests. Samples were analyzed from the Bay Area, New York City, Connecticut, South Florida, Louisiana, Missouri, Utah and western Washington State, plus the metro areas in Minneapolis and Philadelphia. The Bay Area samples were collected and tested April 23-27.

The CDC researchers extrapolated the positive results to the community at large to determine what percentage of all residents in the area likely had been infected.

The body produces antibodies to a pathogen when a person is infected, regardless of whether the individual experiences symptoms of disease. So antibody tests can identify people who were infected and may not have known it, or who were never tested for active illness because they did not have access to testing.

In the early months of the coronavirus pandemic, testing for active infection was in extremely short supply in the United States. Infectious disease experts have long said the number of confirmed cases represents only a fraction of the actual number of people infected. Generally, experts have estimated that the actual infections are about 10 times the confirmed cases.

In all sites, the estimated number of infections with the virus that causes COVID-19 ... was much higher than the reported cases. This may reflect the number of persons who had mild or no illness or who did not seek medical care or undergo testing, but who still may have contributed to ongoing virus transmission in the population, said the studys lead author, CDC researcher Fiona Havers, in an email.

Because people often do not know if they are infected, the public should continue to take steps to help prevent the spread of COVID-19, she said, including wearing face coverings, maintaining social distance and washing hands frequently.

An earlier study from Stanford looking at antibody results in Santa Clara County reported 2.4% to 4.2% of people had been infected, but that analysis has since been widely challenged. Another study, by UCSF, that looked at antibody prevalence in the Mission District found that about 6% of residents there had been infected as of early May. Almost all of the infections were found among Latino residents who were essential workers and unable to shelter in place.

The CDC study comes with some caveats, in particular that the people who were tested are not necessarily representative of the larger population, since they were individuals with access to health care who were undergoing routine screening.

But infectious disease experts said its still an important snapshot of how widespread the disease was in American communities. Such studies are critical to understanding key questions about the virus, including how many cases are asymptomatic. It will be important to repeat these studies at later dates to get a sense of how the spread of disease has changed over time, experts said.

These studies also demonstrate how much more testing needs to be done for active infections. Depending on the region, the number of people testing positive for antibodies was six to 24 times greater than the confirmed cases.

The good way to interpret this study is despite this huge pandemic, the actual number of people getting infected, even in places like New York City, was not that high. So thats kind of reassuring, said Lee Riley, an infectious disease expert at UC Berkeley. But the bad way of looking at this is there are still a lot more people who could potentially get infected in the future.

Its unclear yet whether antibodies prevent individuals from being reinfected, though most experts believe they provide at least some protection for a period of time.

The pandemic hasnt gotten everybody, Riley said. If the antibody is protective, there are still a lot of people who arent protected.

Erin Allday is a San Francisco Chronicle staff writer. Email: eallday@sfchronicle.com Twitter: @erinallday

See more here:

Antibody study finds coronavirus infections may have been 10 times higher in Bay Area - San Francisco Chronicle