New research suggests that the US already had 6.4 million cases of coronavirus by mid-April - more than were officially reported as of Thursday.
Researchers at the University of California, Berkeley, calculated that the real number of infections by April 18 was nine-times higher than the 712,000 reported by the CDC, which they believe was a massive undercount due to insufficient or inaccurate testing.
According to Johns Hopkins University's tally, 6.39 million US cases were confirmed by September 10. The current death tolls stands at more than 191,000.
Testing in the US has ramped up considerably - but if as many people might have had coronavirus by spring as are confirmed now, the real case number remains a mystery, and is likely much higher.
It is a grim perspective on the early spread of coronavirus - but the upside is that if cases were as high as the Berkeley team estimates by April, and as few as 20 percent of the population needs antibodies to reach herd immunity, the US may be closer than previously thought.
The numbers of estimated versus confirmed coronavirus cases were closest, as of April, in states like New York and Washington (green) that were hit hard early on, while the disparity was greater in places like California and Texas (dark blue), where infections spiked later
UC Berkeley scientists calculated that the real number of US coronavirus infections in the US was anywhere between three- and 20-times higher than the Centers for Disease Control and Prevention's (CDC) tally of confirmed cases as of April 18.
The disparity ranged broadly from state-to-state and, as a consequence, so did their mathematical model's estimate of the real national case number.
On April 18, the US ran a total of 146,156 coronavirus tests, according to data from the COVID Tracking Project.
Today, 614,042 tests were performed in the US - four times more than were being done in April.
President Donald Trump boasted on April 27 that the US had tested 'more than every country combined,' for coronavirus.
But the new study blames 'incomplete' testing for 86 percent of the undercount of coronavirus tests up until April 18.
Granted, a portion of that inadequate testing took place in the earlier months of the pandemic's grip on the US, when the CDC had sent out faulty tests, private sector kits were just beginning to trickle onto the market, and fury over lacking testing in the US abounded.
The UC Berkeley authors attributed the remaining 14 percent of disparity between their April case estimate and the CDC's to inaccurate tests.