The data suggests the fatality rate for pregnant women with Covid-19—0.25% —is about the same as women who aren’t pregnant. But it’s impossible to say, because the data doesn’t even tell half the story.

The 91,412 women are just 28% of the initial set of 326,335 women of reproductive age who tested positive for Covid-19 that were selected for the analysis. Of the remaining 71%, the CDC didn’t have information on whether the women were pregnant or not.

Only about a quarter of the total case reports for Covid-19 include any information on pregnancy status, according to the CDC. In some cases, the status is being updated, in others it might take a long time to know—if ever at all.

“We do not know the total number of cases and are seeing in these data the cases that represent those who are most likely to get tested and to have their status reported,” says Neel Shah, a gynecologist at Harvard Medical School who has long been working to tackle maternal mortality in the US. So the total number of pregnant women with Covid-19 remains unknown.

Something else seems even more egregious: For a study that relies on rates of hospitalization as an indication of the severity of Covid-19, cases, there’s no indication of why these pregnant women were actually in the hospital. Specifically, all the pregnant women recorded as hospitalized by the CDC data might have been admitted for other causes—including, crucially, labor and delivery, which lead to the hospitalization of nearly all pregnant women.

It will take a long time to get clarity, too. Because records often don’t accurately account for the cause of death and require extensive review, data on maternal mortality in the US currently takes years to be collected and formally analyzed.

📬 Sign up for the Daily Brief

Our free, fast, and fun briefing on the global economy, delivered every weekday morning.