But Covid cases could be undercounted by a factor of 30, an early survey of the surge in New York City indicates. “It would appear official case counts are under-estimating the true burden of infection by about 30-fold, which is a huge surprise,” said Denis Nash, an author of the study and a distinguished professor of epidemiology at the City University of New York School of Public Health.

About one in five – 22% – of adult New Yorkers likely had Covid between 23 April and 8 May, according to the preprint study, which has not been peer-reviewed or published. That would mean 1.5 million adults in the city had Covid in a single two-week period – far higher than official counts during that time.

While the study focused on New York, these findings may be true throughout the rest of the country, Nash said. In fact, New Yorkers likely have better access to testing than most of the country, which means undercounting could be even worse elsewhere.

“It’s very worrisome. To me, it means that our ability to really understand and get ahead of the virus is undermined,” Nash said.