By mid-August, jails, prisons, and other detention centers accounted for all of the top 10 Covid-19 clusters in the country. This week, the number of Covid-19 deaths among inmates and correctional officers passed 1,000, with more than 160,000 infected.
But as Covid-19 cases among incarcerated people continue to climb, the racial makeup of those cases has remained obscured, despite the fact that the groups most affected by Covid-19 — people of color, and in particular, Black people — have also been disproportionately incarcerated. Only four states are proactively reporting any demographic data on their Covid-19 testing or case counts among incarcerated people: Vermont, Tennessee, Massachusetts , and Washington.
Without information from far more states, many public health experts, policymakers, and community organizations say that they have an incomplete understanding of Covid-19’s racial disparities behind bars, which limits the effectiveness of their response.
“Reporting case and testing numbers by race is especially important to ensure that there is an equitable response, especially in the context of the criminal legal system that disproportionately targets and incarcerates people of color,” said Lauren-Brinkley-Rubenstein, a professor of social health at the University of North Carolina-Chapel Hill and a co-founder of the Covid Prison Project.
The lack of data is cause for concern, experts said, given that federal health data show that outside of prisons, there are significant racial disparities in Covid-19 infections, hospitalizations, and deaths.
Still, many prison systems have withheld the demographic breakdown of their case data. Some have said they simply aren’t stratifying their data that way, while others, including California, have argued that doing so would violate health privacy laws.
California’s correctional facilities have been particularly hard-hit by the pandemic. Half of the top 10 Covid-19 clusters were in the state, which has seen at least 9,580 cases and 54 deaths among incarcerated individuals.
In a bid to understand the demographic makeup of those cases, STAT filed a public records request with the California Correctional Health Care Services (CCHCS) for race and ethnicity data on Covid-19 cases, hospitalizations, and deaths among inmates. The agency twice denied that request, arguing that de-identifying health information to make it shareable would require that identifiers — “including all geographic subdivisions smaller than a State” — are removed. But geographic subdivisions smaller than states, like Los Angeles County, are releasing city- and county-level racial data without violating patient privacy.
“This information should be made public and is not covered by HIPAA or other privacy laws,” said Michael Bien, a lawyer who represents inmates in two ongoing class-action lawsuits against the California Department of Corrections and Rehabilitation. “It is important public health information and criminal justice information that is necessary for public health, the media, scientists, scholars and citizens.”
After multiple appeals, the correctional department released partial records that provided a racial breakdown. Of the 8,766 positive tests, 27% of those infected were white, 22% were Black, 44% were Latino, and 7% were another or unknown race. This compares to a prison population that was 21% white, 28% Black, 44% Hispanic, and 6.5% other, as of December 2019.
The department did not release demographic data on inmate deaths. But Bien, who has access to California correctional health records as a result of a whistleblower lawsuit, shared that data with STAT. Of the 54 inmates who have died, 25 were Hispanic, 15 were white, 8 were Black, 3 were Pacific Islander, 1 was Vietnamese-American, 1 was Native-American, and 1 was of unknown race. Nearly half of inmates who died also had serious mental illnesses.
“The data … shows a disproportionate impact on certain marginalized groups in the prison system,” Bien said.
At first glance, the statistics seem to suggest the demographics of Covid-19 cases reflect the demographics of CDCR prisons. But experts said the picture is far more complicated.
Kathryn Nowotny, a statistician at the University of Miami who studies correctional health, cautioned that the relatively muted disparities in the preliminary data likely stem from the fact that Black individuals are so overrepresented in prisons that they have become a random sample of the overall Black population, while incarcerated white people tend to have more health problems than white people in the general population.
“If we’re pulling from the sickest portions of the white population, and more randomly from the black population, then… you wouldn’t see the same health disparities as in prison,” said Nowotny, who co-founded the Covid Prison Project. “So it’s really understanding population flows and who we incarcerate in society and how we punish people, who gets arrested for what sentence, who gets arrested for what, those are all social processes that influence these patterns that we see in health.
John Eason, a University of Wisconsin-Madison sociologist who studies the effect of prisons on rural communities, argued that “it doesn’t matter who it is” that’s getting worse-hit by Covid-19 behind bars, given how many Black individuals are incarcerated. “If we don’t find a way to decarcerate, Black people are going to lose.”
In early August, Sens. Elizabeth Warren (D-Mass.) and Cory Booker (D-N.J.) introduced a bill that would require federal, state, and local correctional facilities to collect and publish data on Covid-19 in prisons, and to break that data down by race and ethnicity, gender, and disability, among other categories.
Many correctional facilities have started reporting more health data on inmates during the pandemic, launching online dashboards with tests, cases, hospitalizations, and deaths.
“In some ways, the coronavirus information is the most innovative and timely national health data that we’ve ever had for prisons,” said Nowotny.
But as the pandemic has surged to new levels, many correctional facilities have refused to release demographic data. And even among those states that are proactively doing so, the approaches to data collection vary so widely that it’s difficult to glean broader insight.
Washington state discloses the racial breakdown of its total number of positive cases — though the state doesn’t count Hispanic people in their own category. Those numbers show that positive cases mostly follow the overall population count, with slight variations. Black individuals, for example, make up roughly 18% of the prison population, but only 14% of positive cases. Vermont has twice tested all those incarcerated. About 86% of its inmates are white, while white inmates accounted for 80% of positive tests. Just over 8% of its population was Black, but 16% of inmates testing positive were Black. Massachusetts, meanwhile, tracks the daily number of tests for Black and White individuals, as well as Hispanic people in a separate category for ethnicity.
Tennessee has also conducted universal testing, though it only reports the breakdown of inmates tested, not the results. Tennessee uses more specific racial categories than other states. But of the nearly 24,000 tests the state has conducted, 74% were categorized as race “unavailable.”
Experts say that categorizing inmates of different races into an “unknown/other” group further disenfranchises those populations. And, they add, the broader lack of data also makes it difficult to plan an equitable response to Covid-19, including considerations into who is released early.
The data provided by Bien show that a number of the deaths in custody might have been prevented if more incarcerated individuals had been released early. The correctional department’s internal risk assessment system showed the majority of those who died were at low risk of reoffending. Nearly one-quarter were eligible for parole in 2020 or 2021.
CDCR has released more than 10,000 individuals since the start of the pandemic, and plans to release up to 8,000 more by September. White people make up 21% of the prison population, but 30% of the 4,053 individuals identified thus far for release, while Black and Hispanic prisoners are underrepresented in those eligible for parole.
Experts said this shows the racial biases inherent to the criminal justice system, which research has shown disproportionately arrests and sentences people of color, and in particular Black people.
The concern is “that the release policies are going to favor white people,” said Nowotny, “because they tend to be given more lenient sentences for the same crime.”