The WSJ's takeaways:
- More Than 15% of Americans Have Had Covid-19, CDC Estimates
- Some 53 million people in the U.S. likely had had Covid-19 by the end of September, according to a modeling estimate from researchers at the Centers for Disease Control and Prevention. Roughly 6.9 million infections had been confirmed within that time period, suggesting that roughly one in every eight cases was identified.
The CDC estimates are also in line with studies that have looked at blood samples for disease-fighting antibodies as a sign of a person’s past infection and arrived at similar conclusions: Many more people have had Covid-19 in the U.S. than have been reported, but the majority of people is still at risk.
“This indicates that approximately 84% of the U.S. population has not yet been infected, and thus most of the country remains at risk, despite already high rates of hospitalization,” the authors wrote.
...
- The researchers estimated that roughly 16% of the U.S. population had been infected with the virus through September, ranging from 9% to 31% across different locations. More cases went unreported among younger people, especially children, the report says. Case reporting has improved over time, particularly among those younger cohorts, as the capacity to test for the virus has grown. The biggest leaps in the ability to find cases occurred in the spring and summer.
In June, CDC Director Robert Redfield said the country was capturing roughly 10% of cases through May, and the current report estimates that 13% of cases were captured through the end of September.
Y'all might find this podcast/interview on disease modelling, some of it specifically relating to COVID interesting.
It would be nice to know what their modeling says about the last two months, when cases have skyrocketed. It would seem like if the undercount is similar, we'd be approaching 100M cases by now, with certain isolated regions probably approaching herd immunity. That's pretty wild.