Philadelphia has one of the largest outbreaks of coronavirus in the country, to judge by the number of infections that have been confirmed. But the city’s rate of infection — illnesses divided by people — compared with hard-hit cities like New York, Detroit, and New Orleans, has been significantly lower, according to data collected by Johns Hopkins University.

Just over 40 out of every 10,000 Philadelphians have tested positive for coronavirus, an infection rate that ranks 68th out of more than 3,000 U.S. counties included in the Hopkins database. But by the sheer numbers of infections, the city ranks 17th.

About one in 10,000 Philadelphia residents have died of coronavirus, the 140th-highest death rate among those counties.

Tom Farley, the city’s health commissioner, said he was grateful Philadelphia, though still struggling, hasn’t fared as badly as other cities.

“Our mortality rates are about one-tenth of the rates out of New York City,” he said. “That was an eye-opener. But there are other cities, I’m sure, that have not been hit nearly as hard as us.”

It’s unclear why Philadelphia has not had as catastrophic an outbreak as New York City.

“If you look at a map of the country, the whole Northeastern city corridor has been hit harder than much of the rest of the country — D.C., Philly, New York, New Haven, and Boston, with New York obviously hit the hardest," Farley said. "The other ones in that corridor all have pretty similar rates. We experienced a pretty similar epidemic and we did a shutdown around the same time. Small differences about when you shut down on the arrival of the virus can make a big difference. If we had delayed another week, things would have been a lot worse.”

Beyond simply tracking the number of cases and deaths and the rate of infection, though, Farley said it’s difficult to glean more information about how the virus is spreading and which communities are most affected, simply because cases spiked so quickly that the city had no time to follow up on individual cases.

The city has been able to track some cases by race, and those numbers suggest people of color may be disproportionately affected. On April 3, data on one-third of the city’s coronavirus patients — then just 1,675 confirmed cases — showed that 46% were African American, 37% were white, 10% were Latino, and 3% were Asian American. The city’s number of cases has since grown to more than 6,000.

The city is trying to match positive cases with other public health databases to “fill in that missing variable on race and ethnicity,” Farley said. “Hopefully we’ll have a decent proportion of them answered."

Data on the income level of people most affected is also scarce.

“This grew to so many cases, so fast, that it overwhelmed our ability to follow up on cases,” Farley said. “Initially, we were trying to complete case investigation forms. And that’s OK when you get 10 to 15 cases a day. It doesn’t work when you get 200 cases a day."