According to the Centers for Disease Control and Prevention, published on Friday, the number of coronavirus infections in many parts of the United States is more than 10 times the reported level.
The analysis is part of a wide range of studies initiated by the CDC to evaluate how widely the virus has spread. Similar research, sponsored by universities, national governments, and the World Health Organization, is ongoing worldwide.
For example, a CDC study found that in South Florida, as of April 10, just under 2% of the population was exposed to the virus, but that proportion is likely to be higher at present, given the surge in infections in the state. Prevalence was highest in New York at almost 7% on April 1.
“This study emphasizes that many people are probably infected without knowing it, probably because they have a mild or asymptomatic infection,” said Dr. Fiona Havers, who led the CDC study. “But these people can still spread it among others.”
She emphasized the importance of handwashing, wearing fabric masks, and social distance to stop the spread of the virus among people without symptoms.
The figures show that even in areas heavily affected by the virus, the vast majority of people are not yet infected, says Scott Hensley, a virologist at the University of Pennsylvania who did not participate in the study.
“Many of us are sedentary ducks that are still sensitive to second waves,” he said.
The difference between the reported infections and the actual prevalence in the data was highest in Missouri, where as of April 26, about 2.65% of the population was infected with the virus, although many people may not have felt sick. This number is about 24 times higher than the recorded figure: almost 162,000 compared to 6,800, presumably infected by then.
The results confirm what some scientists warned for several months: that without more extensive testing, dozens of infected people go unnoticed and spread the virus.
“Our politicians may say that our testing is cool, but in reality our testing is inadequate,” Hensley said. “It is such research that we need now.”
Dr. Robert Redfield, director of the CDC, hinted at this trend on Thursday while talking to reporters.
“Our best estimate so far is that for every reported case, there were actually 10 other infections,” said Redfield.
The source of his claim at that time was unclear. The CDC later posted data on its website and on MedRxiv, a repository of scientific results that have not yet been reviewed by peer reviewers.
CDC researchers tested samples from 11,933 people in six regions of the United States during separate periods from March 23 to May 3: in Washington, in the Puget Sound region, where the first case of COVID-19 was diagnosed in the country, as well as in New York south. Florida, Missouri, Utah and Connecticut.
Samples were collected in commercial laboratories from people who came for routine examinations, such as cholesterol tests, and were evaluated for the presence of antibodies to the virus – which would indicate a previous infection even in the absence of symptoms.
Researchers then estimated the number of infections in each area. For example, in New York by April 1, 53,803 cases of the disease were reported, but the actual number of cases of infection was 12 times higher, almost 642,000 people.
In the CDC study, the prevalence of the city is 6.93%, well below the 21% estimated by a state survey in April. This number was based on people recruited in supermarkets, so the results would be biased towards people who would shop during the pandemic – young people or those who already had the virus and felt safe, experts say.
Havers also notes that when this study was conducted in New York on April 19-28, the spike in prevalence would correspond to a surge in infections in the city at that time. She said the CDC plans to repeat surveys in all regions to see how prevalence changes over time. Additional CDC studies will test how well this approach reflects real prevalence.
Saskia Popescu, an epidemiologist at the University of Arizona, said the CDC survey could also be distorted by people with chronic illnesses who visit commercial labs more often. However, it is more representative of the general population than other studies, because it includes everyone who came to the laboratory for different purposes, rather than restricting it to specific groups, such as healthcare providers, who felt sick with coronavirus .
“Most of the serological testing that we observed was really focused on this – people who thought they were being exposed or feeling sick at some point,” she said. “Ultimately, this approach is much more representative.”
She also praised the researchers for not drawing conclusions from the study on the immune status of the participants, because it is still unclear how the presence of antibodies affects protection against the virus.
The analysis also highlights the large differences between different parts of the country – and the importance of not only a sufficient number of tests, but also laboratory capabilities, Popescu said. In Arizona, she added, lag delays the test results for five to six days.
Hensley said he was worried that New York and other northeastern states might mistakenly believe that they had passed the danger and opened too soon.
“We need to turn south to see what disastrous events were there,” he said. “If we open up, like in Florida or Texas, you can almost bet that we will be in the same position as now.”
This article originally appeared in New York Times,
© 2020 The New York Times Company