When it comes to African Americans and COVID-19, reports swung from one extreme to the other: first, rumors that black people don’t get the coronavirus, then news that the virus is killing African Americans in droves.

The initial misconceptions showed how living in a racist society had warped a sense of black resistance into the myth of black immunity. Those narratives had the subtext of racist “blacks are resistant to pain and illness” stereotypes. But we are not superhuman. If anything, the lower life expectancy, higher rates of diabetes, heart disease, and comorbidities make it clear that African Americans are at greater risk for developing and dying from COVID-19.

Then ​that elevated risk bore out in the news. Across the country, African Americans were becoming infected and dying at obscenely higher rates than other racial groups. Suddenly, in St. Louis, the only people dying from “the Rona” were African Americans. Black Philadelphians accounted for more than half the deaths in the city. In New Jersey, blacks have so far made up more than 20% of deaths while being only 14% of the population. Some medical experts explained that underlying health issues were one reason for the greater proportions of blacks becoming infected by and succumbing to the virus, ​alongside representing a high proportion of unprotected essential workers.

But the truth about black vulnerability to the coronavirus is somewhere in the middle.

As Europe has proven, there is a lot of variation in counting COVID-19 infections and deaths that can skew the data, especially in early reports. England’s public health organization used tests but not death certificates to count ​coronavirus deaths, so the Office for National Statistics stepped in to do so. Spain’s initial counts included postmortem examinations but stopped doing so as its health system experienced greater strain. In Belgium and France, the count did not initially include people outside of hospitals, like those in nursing or retirement homes.

When New York City included the deaths of people who had not been tested but were presumed to be positive, the number of victims increased by more than 3,700. Yet officials in California, Washington, Louisiana, and Chicago have only been counting deaths confirmed by testing. While half of all European deaths have involved people in nursing homes, it is unclear how many Americans in institutional living have been affected.

Epidemiologists use the number of “excess deaths” to better understand the impact of a disease on a population. Excess deaths are defined as how many more people have died each month in comparison with historical averages. It includes people who exhibited coronavirus symptoms but were not tested or those who could not receive care due to health facilities being overwhelmed. In Europe, updated statistics show a vast initial undercount in deaths that are likely attributable to the disease.

In the United States, the risk of not counting excess deaths has led to mass underestimates of COVID-19’s impact. Although blacks are likely overrepresented in the pandemic’s spread due to the health impacts of living in a racist society — ​such as poor access to care and everyday aggressions — the rates may be overstated due to the lack of widespread testing, as well as data that has excluded nursing homes and institutionalized disabled populations.

In this context, we have likely undercounted the number of whites who have taken ill or died from COVID-19. Even as blacks face disproportionate mortality rates, whites are still the majority in this country and are the majority of those dying from this disease. As of April 23, over 10,000 African Americans and 17,000 white Americans had died of this illness. The risk of undercounts can provide a false sense that the disease is “not really as bad as they say,” or that it only impacts black and other minority communities. This may lead people to not take the necessary precautions for their well-being. As a professor living in one of the hardest-hit states, New Jersey, I can tell you that this disease is affecting my students and their families across colors.

We need a massive increase in testing in the United States, a national standard for counting COVID-19 infections and deaths, including outside of hospitals, and information on excess U.S. deaths broken down by race, age, and other social indicators. This will help us have a better understanding of how this crisis is affecting our society.

Until then, we can keep staying at home and washing our hands.

Chinyere Osuji is an assistant professor of sociology at Rutgers University-Camden. She is the author of “Boundaries of Love: Interracial Marriage and the Meaning of Race” and is writing a book about racism in nursing. @chinyereosuji