Case counts for COVID-19 are commonly reported, but they can be misleading: reliable data suggest that the vast majority of infections in the United States have not been diagnosed, and that the true number of COVID-19 infections is much higher than the case count total indicates.
How did we reach this conclusion?
As of June 10, the US had reported 2 million total cases, and around 112,000 deaths.
This means that about 5.5% of reported cases in the US have resulted in death—or one in 17 people.
However, consensus estimates of the true number of people who die from COVID-19 are much lower— around .0.5% to 1%, or one in 100-200 people infected.
Based on the .8% infection fatality rate which is our best estimate in the US, we would expect 112,000 deaths to result from 14,000,000 COVID-19 cases (112,000/.008). This suggests that only 14% of US COVID-19 (2,000,000/14,000,000) cases have been detected.
There are many contributing factors that make the undercount of cases in the US likely, including low initial testing capacity and a large proportion (~35%) of asymptomatic cases.