This story first published online at NC Health News.

In the final hours of August 2, Billy Bingham lay alone in his cell, silent except for the sounds of his short, shallow breaths. Thirty-four minutes after midnight, he was dead. Paramedics arrived at the Albemarle Correctional Institute, a state prison in central North Carolina, too late to take him to the hospital, and instead called the time of death and left. A prison official called his uncle, who would wake up to a voicemail sharing the news.

Bingham’s story is not unique. A North Carolina Health News investigation in partnership with VICE News found that in the first seven months of the pandemic, North Carolina failed to report all of the prisoners in its custody who died of COVID-19-related causes, according to death documents; through public information requests to the county register of deeds offices and the state medical examiner’s office, our investigation identified three prisoners with COVID-19-related deaths who were not included in the state prison agency’s death count. While North Carolina provides a case study in how some prisoners fall through the cracks in accounting, even when relevant documents are available, experts say other states could be similar.

“I think a lot of us are worried about underreporting,” said Lauren Brinkley-Rubinstein, a public health researcher at the UNC School of Medicine and co-founder of the COVID Prison Project. 

North Carolina, like most states, doesn’t publish the names of prisoners in its custody who have died of COVID-19; it announces them anonymously, in a practice that is ideally meant to protect incarcerated people’s medical information. But public health experts and family members alike have worried that the system creates a lack of accountability that enables prison agencies to underreport the true toll of the virus.

These undercounts come at high costs, and downplaying the severity of the virus within carceral facilities has left family members confused and fueled conspiracy theories along the way.

“People want to know what happened to their loved one,” said Jay Bingham, Billy’s brother. “I still don’t.”

The novel coronavirus pandemic continues to tear through prisons, jails, and detention centers across the country, where incarcerated people often live in close quarters and have little control over their ability to social distance. One in five prisoners in the United States has had COVID-19, according to a Marshall Project analysis released in December.

Bingham, who was serving a sentence for first-degree murder, may have never known he was one of them. Family members worry he was never informed.

Bingham tested positive on July 5, according to the state medical examiner’s investigation of his death, and was immediately taken to Central Prison in Raleigh, which contains a large medical center. But upon return to his original prison, Albemarle, on July 30, Bingham said he didn’t have the virus, according to his uncle, Eddie Bingham.

“He said they had tested him for COVID and told him he didn’t have it,” said Bingham. “But if you look at the death certificate, that was put on the death certificate.” 

The chief medical officer for North Carolina’s prison system, Dr. Les Campbell, said the agency is discussing with the state health department whether they should update their numbers following our inquiry about these death certificates.

“None of us like for someone to die in our care or custody, regardless of the reason,” said Campbell. “There’s no reason for us to try to hide or conceal what that reason is. We want to make sure we’re as objective as we can, and that’s how I look at this.” Campbell also said that all inmates are informed when they test positive for COVID-19, and receive a detailed sheet informing them of next steps.

Bingham’s medical investigation, obtained by a public records request from the North Carolina Department of Health and Human Services, states he returned to Albemarle prison on July 30; it makes no mention of whether he received a test that came back negative before his transfer. Regardless, a negative test is not used to determine whether someone’s death was COVID-19-related, according to Campbell. Bingham was placed in “sick ward segregation,” an isolated cell used to quarantine prisoners upon arrival.

His uncle Eddie alleged that Bingham, who was 61 and had lung disease and diabetes, told him he had simply gone to get treatment for trouble breathing. He would call Eddie three times that weekend, always to talk about his symptoms—how he thought he had a fever, how he was so weak he could barely walk to the bathroom. His illness consumed their conversations, Eddie said. But he insisted his nephew never mentioned previously testing positive for COVID-19.

Bingham may not have said he had COVID, but he made one thing clear—he was sure he was going to die. “He called me Sunday evening,” his uncle said. “And he said, ‘Eddie, I’m sick, I don’t think I’m gonna make it.’”

Hours later, Bingham was dead.

When do North Carolina prison officials count a death as COVID-related?:

DPS uses a state health department criteria for defining if a person’s death is COVID-related: “COVID-19 deaths include people who have had a positive molecular (PCR) or antigen test for COVID-19, who died without fully recovering from COVID-19, and who had no alternative cause of death identified.” 

Here’s more info, gleaned from a conversation with Dr. Les Campbell, chief medical officer for prisons at DPS:

If a person has not “recovered.” Campbell said all prisons are isolated for at least 10 days after the date of taking a test that later came back positive. Saying someone has “returned to baseline” is “case-dependent,” Campbell said, and can be a complicated call.

If COVID-19 either contributed to, or was the direct cause of, the death, according to Campbell. “If we think COVID indeed hastened the death of one of the offenders, then I will call that a COVID-related death,” he said.

“We don’t use a negative test to exclude COVID as a cause of death,” Campbell said.

It’s not counted as a COVID death any time someone had tested positive “in proximity” to their death. “We don’t announce other tests in proximity to somebody’s death so I don’t know why we would do it necessarily in this one.”

It’s not NOT counted as a COVID death if a person’s death is “ambiguous,”Campbell said, “I certainly do not want to put stuff out in the public that we’re not fairly confident is correct.”

Bingham’s death certificate says he died of “pneumonia due to COVID-19 virus” on August 3. Two days later, Luther Wilson, 60, died of complications of COVID-19 coupled with end stage renal disease at Maury Correctional Institution in eastern North Carolina, according to his death certificate. On September 20, Daryl Washington, a 51-year-old man incarcerated at Central Prison in Raleigh, died of complications of endocarditis, with COVID-19 listed as a significant contributing factor to his death.

Deborah Radisch, former chief medical examiner for the state of North Carolina, reviewed all three death certificates at VICE News and NC Health News’s request and said all would be considered COVID-19 deaths under Centers for Disease Control and Prevention (CDC) criteria, which the Office of the Chief Medical Examiner follows. 

Robert Anderson, chief of the Mortality Statistics Branch for the National Center of Health Statistics at the CDC, also reviewed the documents. He said all would be coded as COVID-19 deaths by the federal government, but in tabulation, Washington’s would be classified more specifically as COVID-19 related.” It’s not clear why the Department of Public Safety (DPS), which oversees state prisons, didn’t include Bingham, Wilson, and Washington in their COVID-19-related death count, or announce them. The agency declined to comment on the findings.

A spokesperson for North Carolina’s prison agency said they couldn’t speak to why individual prisoners weren’t counted, citing state and federal privacy laws. He said “no one is trying to hide an offender’s death if it is COVID-related,” and that the agency follows the state health department’s definition for COVID-19 deaths.

That state criteria is as follows, he said: “COVID-19 deaths include people who have had a positive molecular (PCR) or antigen test for COVID-19, who died without fully recovering from COVID-19, and who had no alternative cause of death identified.

“Medical professionals may, at times, disagree with the impact of a COVID-19 infection on the cause of death,” said John Bull, DPS spokesperson, who noted the agency reviewed the death documents provided by NC Health News. “Also, new medical information on the impact of COVID-19 on human health continues to emerge.”

DPS’s count is for prisoners whose deaths are “confirmed COVID-19 related,” according to the public database on their website. This means the prison system counts both deaths that are directly due to COVID-19, and those where COVID-19 hastened the death, according to Campbell.

Investigations are also ongoing. A medical examiner investigation conducted by North Caroina’s office of the chief medical examiner additionally confirmed Bingham’s cause of death as COVID-19; those investigations are still pending for inmates Wilson and Daryl Washington due to a processing backlog. But Radisch said this doesn’t mean their COVID-19 diagnoses should be called into question. 

If necessary, state certifiers will amend death certificates for cases where COVID-19 is found to not play a role–such as a stabbing or car crash–but it’s unlikely for natural deaths such as these, according to Radisch. “What you have to understand it that bottom line, they would still be coded and included in COVID deaths,” she said. “The only reason these natural cause deaths are being investigated by the state medical examiner is because they’re prisoners–all in-custody deaths have to be investigated by state law.”

In lieu of releasing prisoners’ names–citing HIPAA and state law that deems all prison records confidential–North Carolina’s state prison agency instead publishes a de-identified press release each time an incarcerated person is deemed to have died of COVID-19-related causes. They provide the prisoner’s rough age range, their last prison location, and their date of death. The death is then added to their COVID-19 dashboard online, where the official number of prisoners who have died currently sits at 44.

Using full names and offender data of every prisoner who died in 2020 obtained from the state agency by a freedom of information request, VICE News and NC Health News cross-referenced these press releases with each prisoner’s date of death, age, and location to identify incarcerated people who had likely died of the virus and whose COVID-19-related deaths were announced to the public. Reporters then obtained the death certificates of every prisoner who had died between February 29 and September 20, 2020, from COVID-19 or otherwise, from the register of deeds offices in the counties in which they died through public information requests, as well as medical examiner investigations, and autopsies when available. In North Carolina, incarcerated people’s death certificates must be completed by county medical examiners.

Men take recreation in the yard of Central Prison in Raleigh. Photo credit: Rose Hoban

The certificates then go to the state vital records office, where they receive final review and are coded into a vital statistics database before being sent to the CDC’s National Center for Health Statistics, which codes each cause of death that occurred within the U.S. using the International Classification of Diseases, a global diagnostic tool maintained by the WHO.

The death certificates confirmed the deaths due to COVID-19 that North Carolina’s prison system announced, through September 20, when the deadliest months for North Carolina prisoners were yet to come. They also revealed that there were three other incarcerated people who had died due to the virus and had not been made known to the public. While we investigated deaths that occurred through late September, 33 of the official virus-related prison deaths occurred after this date, as COVID-19 raged inside the state’s 55 prisons in the final months of 2020, leaving open the possibility that more prisoners have been uncounted.

“I think it just speaks to a reluctance on the part of the state agency and state leadership to really be accountable and transparent about what is happening to folks who are incarcerated during this pandemic,” said Leah Kang, attorney at the American Civil Liberties Union of North Carolina, one of the plaintiffs in an ongoing lawsuit against the state over its ability to protect prisoners in their custody from the novel coronavirus.

Throughout the pandemic, criminal justice experts and advocates have feared the public doesn’t actually know how many prisoners have died of COVID-19.

“Every state is undercounting,” said Michele Deitch, attorney and senior lecturer at the University of Texas Law School who specializes in prison and jail oversight and directs the COVID, Corrections, and Oversight Project. “I’m 100 percent sure it’s happening. Wherever those numbers are reported, they should be clear how they’re being counted. We can’t assume that everyone’s counting the same way.”

It’s a problem that likely extends across the country, other experts said. North Carolina has had relatively few prisoners die of the virus compared to other states, such as California and Michigan. “To anyone who has worked in the carceral system, it’s blindingly obvious that prisons and jails both operate with impunity a lot of the time, and prize their opacity,” said Aaron Littman, teaching fellow at the UCLA School of Law and deputy director of the COVID Behind Bars Data Project. “We have been concerned from the outset that there would be a variety of kinds of underreporting.”

Littman said there are different ways incarcerated people who die due to COVID-19-related causes may be undercounted this year. “There’s people who are tested and confirmed to have had and died of COVID-19, who are not being reported as such,” he said. “That’s sort of the clearest case.”

That doesn’t include incarcerated people who were not tested before they died, particularly in the earlier months of the pandemic when states didn’t have the capacity to test everyone. Nor does it include people who were released from custody but died of COVID-19 they may have contracted while inside a prison or jail.

Inaccurate counts have consequences for families, but also for public health responses, according to Littman. “This information is important for politicians, judges, sheriffs, and governors – people who have decision-making authority – to know what the true scope of the impact of COVID-19 in prisons and jails is,” he said. “It helps them weigh decisions about reducing population and protecting medically vulnerable people. There’s a debate going on about who should be prioritized for the vaccine – and the fact that people are dying at a high rate in prison is a strong argument in favor of prioritizing them.

“But that argument is reliant on data,” he added. “All of those decisions turn on what we know about what’s happening.”

When prisoner cause-of-death determinations are made, exactly who gets counted—and why—is often opaque. Currently, the vast majority of state prison agencies, including the 34 out of 39 states who responded to public information requests, do not make the names of prisoners who have died of the virus available to the public, citing HIPAA concerns or state privacy laws. The Bureau of Prisons, which oversees federal prisons across the country, does share these names.

So does Texas, the country’s deadliest state for prisoners during the pandemic. “For us, it’s a transparency issue. Not all elements of HIPAA expire after death, but the name of an offender does,” said Desel of the Texas Department of Criminal Justice.

To ensure accuracy in reporting, Texas only publishes these names after the death undergoes an investigation conducted by the state’s Office of the Inspector General and the prison agency’s correctional health care partners. Investigators review all death certificates, existing autopsies, and medical examinations before deciding whether the death was COVID-related.

Deitch of UT Austin said this practice of publishing names could help with transparency, by creating greater opportunity for families, journalists, and public health officials who fill out death documents to hold prison agencies accountable.

“Texas has been hard-hit, but they’ve been relatively good on transparency,” she said. “The invisibility of the data in other places is really troubling,” said Deitch. “It doesn’t even give this person dignity in death, by acknowledging that they had ever lived or died in this facility.”

Still, the ultimate determination for a prisoner’s cause of death is made by the Texas Department of Criminal Justice, meaning the public still learns information at the whim of the prison agency’s judgment. Deitch said a stronger solution would be to democratize the data, like having an independent oversight body collect and report raw data about the prison system to the public.

“This is not just a COVID issue,” Deitch said, noting researchers have long advocated for more openness around in-custody deaths of all kinds. “This is an all-data issue—we need more transparency when it comes to the data about what’s going on in prisons and jails. Deaths are one piece of it.”

Since 2008, the American Bar Association has called on the federal government and the states to create and fund independent “monitoring entities” to oversee correctional facilities in their jurisdiction. These bodies would access all aspects of a prison or jail’s operations and conditions, including inmate death records. They would publish findings and documents online, particularly when they identify any potential problems.

The entrance to the hospital facility located at Central Regional Prison in Raleigh. Photo credit: Rose Hoban

In 2018, Washington was the first to create such an oversight body. Known as the Office of Corrections Ombuds, it has provided independent reports on state prison conditions and data throughout the pandemic.

“You want to get as much information about any death available in one single spot,” said Deitch, who co-chairs ABA’s Subcommittee on Correctional Oversight. “As long as the circumstances of the deaths are listed there, it’s not really a matter of how the prison is counting it. They can have rules about the way they count, whatever. But you as a researcher could look at the data that’s in that database directly and decide.

“And it also allows you to look at that data and go back to the prison system and ask, ‘Why not this one?’” she added. “That’s how you have accountability in government.” The sheer presence of these oversight bodies may also lead to more accurate internal reporting by prison agencies, Deitch said.

Like Washington state, the federal Department of Justice’s Office of Inspector General conducts independent investigations of the Bureau of Prisons (BOP). The Office publishes its own database about COVID-19 in BOP-run facilities.

Our investigation found that inside North Carolina’s one federally-run prison, Butner Correctional Complex, which has had more prisoners die of COVID-19 than any other federal facility in the nation, the BOP did not undercount deaths during this time period as the state system did.

“That’s because of this work that’s being done with OIG,” maintained Deitch.

States vary widely in how they determine and report that a prisoner’s death is COVID-19-related.

“There is no unanimity, or even uniformity in how state criminal justice or corrections agencies are handling any element of this,” said Jeremy Desel, spokesperson for the Texas Department of Criminal Justice, which oversees state prisons. “It’s very much a blaze-your-own trail.”

Some states, like Minnesota and Virginia, report a death any time a prisoner dies while COVID-positive, according to information requests made out to each state prison agency. Some do so after deciding the virus “contributed to” or was “partially responsible” for the person’s decline, like Arizona; other states only do so if it is deemed the root cause of the death. States like Georgia and Arkansas were more vague in their explanation of this process, with the determination simply “made by an attending physician” inside the prison system.

Others check their internal determinations alongside death certificates. Wisconsin, for example, waits to report a prisoner’s death until they receive a death certificate from an outside medical examiner or coroner, and announces whenever COVID-19 is listed as either the cause or a significant condition contributing to the death, according to Department of Corrections spokesperson John Beard. Illinois announces a COVID-related death any time a prisoner is positive for the virus when they die, and adjusts their death counts accordingly after receiving death certificates completed by a county coroner, according to Department of Corrections spokesperson Lindsay Hess.

But at least 20 states, including North Carolina, do not vet their decisions alongside external death documents. This decision about whether a prisoner’s death is COVID-19-related is made internally, by the prison system’s chief medical officer, according to Bull. The prison doctor reviews all medical records from within the prison system and outside medical facilities to see if “COVID-19 infection played a role in the offender’s death.” Only 11 states, of the 32 respondents to nationwide public information requests, said they adjust COVID-19 counts upon receipt of outside death records such as death certificates, medical examiner investigations, or autopsies.

Exactly what broke down between the North Carolina prison system’s criteria and the one medical examiners use to fill out death certificates and investigations is unclear. Regardless, it has led to a discrepancy between what families saw and what the public was told.

Campbell, the chief medical officer for North Carolina’s prison system, was not the head prison doctor at the time these deaths occurred. He assumed the role on October 8, and said he could not comment on individual cases regardless.

He did note, however, that when a prisoner’s cause of death is ambiguous—where it’s unclear whether a person died of a preexisting condition or recent COVID-19 infection—the prison opts not to report it, even if there’s a chance the virus may have played a role.

“In those cases, it’s uncertain,” he said. “We want to be as open as we can, but we also want to provide the most factual information we can to the public. I think it’s unfair to the family to put something out there that we’re not extremely confident is indeed the case.”

After being presented with the three death certificates, Campbell conceded that it may make sense to review their reporting of ambiguous cases alongside medical examiner documents to prevent future discrepancies.

“So I would say those are cases where we’ll rely on the medical examiner’s report to kind of help us make that determination,” he added. “We’re fairly confident in our assessment, but certainly things can change. What we think may be the cause of death, without doing an actual post mortem exam, is possibly going to change when the medical examiner reviews the case, and that’s the whole purpose of having that exam done.” To date, the agency has not relied on medical examiner reports or death certificates to update prisoners’ COVID-related death counts throughout the pandemic, Campbell clarified.

DPS receives death certificates and medical examiner investigations after they’re completed, according to Bull. This process can take weeks or months. The existence of Bingham’s finished investigation, which lists COVID as the root cause of death, further suggests North Carolina prison officials previously haven’t moved to modify death counts even when they are in possession of such documents.

“At every level of staffing at DPS, people should be focused on how they can accurately assess just how lethal it is for this extremely vulnerable population,” said Kang, the American Civil Liberties Union lawyer. “The state has a constitutional duty to keep the people in their custody alive during this pandemic—it has failed in that responsibility for at least 30-ish people who have died, and it’s clear that there are likely more. It is just a question of how many more.”

Campbell said North Carolina’s prison system may adopt this policy in the future.

“We haven’t had this happen before, so we’re gonna have to have that discussion [with DHHS] now,” he said. “I want to do the right thing, families deserve that right to know. We want to make sure we’re giving them factual information, so that’s what we’re gonna do.”

A person could be sick for weeks, or even months, before they die by the cascade of events triggered by COVID-19, such as pneumonia, according to Dr. Michael Osterholm, an epidemiologist at the University of Minnesota. But their passing is still considered a COVID-19-related death. “That’s why we always say deaths are a ‘lagging indicator,’ because they often occur two or three weeks afterwards,” he said. This can be confusing for families of the deceased, and in some cases, can lead to disbelief.

In this case, omitting reports of prisoners with virus-related deaths from the official COVID-19 count can contribute to conspiracy theories. For Wilson’s family, the discrepancy between his death certificate and prison officials’ account fueled disbelief in the severity of the novel coronavirus, as well as the credibility of medical professionals.

“They write that stuff on everybody’s [death certificate] now, cause they get like $14,000 from the government if it’s COVID,” said Roger Wilson, Luther’s brother and his emergency contact in the prison. “Any health—any hospital, anything like that. That’s why there’s so many cases. This is exactly what I’ve been hearing.” All of the family members NC Health News spoke to at first cited this conspiracy theory as a possible explanation for their loved one’s uncounted COVID-19 diagnosis, except for Washington’s family, which did not reply to requests for comment.

The conspiracy theory that doctors are paid more if they list COVID-19 as someone’s cause of death—thus leading health professionals across the country to exaggerate the number of people who have died of the virus—proliferated early on in the pandemic and has persisted. The claim is baseless and there is no evidence to support it. Contraindicating it is the fact that doctors and medical examiners can face legal consequences for falsifying death certificates. Instead, public health experts and recent studies suggest COVID-related deaths are likely undercounted.

Wilson said he’d been informed by a warden’s assistant at the Maury prison that his brother had died of “natural causes,” but given no further explanation.“I didn’t get into all that with the warden assistant,” he said.

Both Eddie and Jay Bingham, Billy Bingham’s brother, said they called the Albemarle prison after learning of his death. Both alleged they were told their brother hadn’t had the coronavirus by prison officials.

“I contacted the prison camp, and they said he didn’t have COVID-19, because they checked him when he came back and he didn’t have the symptoms,” said Jay Bingham.

“And then later, once the doctor says it’s COVID on the death certificate, there’s no denying what the documentation says,” Jay Bingham added. “I don’t know if they just weren’t sure, or didn’t want me to know at that time – but my approximation was that they were trying to be low key to not blow it up, because of the concern of the inmates and families, and what kind of repercussions would come as a result of that, you know?”

Bingham’s family is trying to move forward, but it’s difficult when they still don’t understand the circumstances of his death, Jay Bingham said.  “That was my brother, you know,” he said. “I know my brother was very tough and it would have took a lot to kill him, I can tell you that.”

He wishes Billy had tried to call him that weekend and still wonders what his brother knew in his final moments, what he was thinking. “If I had got to talk to him, he could have told me so much,” Jay said. “I would have been the one to get him to elaborate, to get him to talk about what all was going on in that prison camp – how they were handling things, and his whole diagnosis.”

Most of all, as prisoners across the country continue to die of the virus, he wants clarity.

“Did he die of pneumonia strictly, or was it COVID-related? I still don’t know,” he said. “The families of the inmates—some of these people will never know or find out what the deal is.”

North Carolina Health News is an independent, non-partisan, not-for-profit, statewide news organization dedicated to covering all things health care in North Carolina. Visit NCHN at ​www.northcarolinahealthnews.org​.

Comment on this story at [email protected]. 

Support independent local journalism. Join the INDY Press Club to help us keep fearless watchdog reporting and essential arts and culture coverage viable in the Triangle.