The US has already reported the highest number of coronavirus cases in the world, with 3.9 million confirmed so far. Now, data sets released by the Centres for Disease Control and Prevention and the journal JAMA Internal Medicine yesterday suggest that the total count is likely about 10 times higher.
The data is based on blood samples from over 16,000 people who visited the doctor for routine checkups within two time periods: either in late March to early May, or May to early June. The CDC was able to perform antibody tests on the samples, which came from 10 different regions across the US, including New York City, San Francisco, Florida, Utah, and Washington.
The results suggested that, across the country, infections are between two and 13 times higher than what is currently being reported. Within some sites tested, that number may be even higher; in Missouri, there may be up to 24 times more infections than we know of.
Researchers believe that difference in reported and actual cases comes down to asymptomatic infections or symptomatic ones for which health care wasn't sought, or which weren't able be confirmed via testing. They also noted, that even if the higher estimates for unreported cases turn out to be accurate, the US is still far from the threshold for herd immunity. To reach that, it's likely that 70 to 80 percent of Americans would need to have the coronavirus antibody. For comparison, only about 24 percent of the population in New York City, the hardest-hit area in the US, tested positive for the antibody in early May.
In an accompanying editorial in JAMA, researchers noted that the data collection periods "overlapped with active stay-at-home orders, when most medical appointments and elective admissions were deferred," meaning that the people whose blood was ultimately tested "are likely not representative of a typical prepandemic cohort." In addition, the timing of data collection in different regions "coincides with different parts of the local epidemic curves, some of which were prior to peaks and some during peaks."
Even with these limitations, researchers noted that this is the first population-level COVID-19 study in the US looking at unreported COVID-19 cases, and the takeaways are important in different ways: one, that the current data may be vastly underestimating case counts; and two, that even if that data is corrected to account for unreported cases, herd immunity in the US is still a long way off.