Emergency lights on a us police carShare on Pinterest
A new study finds that Black people in the U.S. are the most likely racial group due to die at the hands of the police. Oliver Helbig/Getty Images
  • An assessment of deaths attributable to police violence in the United States concludes that half of all such killings are misclassified or misreported.
  • The analysis finds that Black people are the most likely of all racial and ethnic groups to die as a result of police violence.
  • The study authors conclude that militarized policing and systemic racism are largely to blame for the high numbers of such deaths.

The morning after the death of George Floyd, a 46-year-old Black man, the Minneapolis Police Department issued a press release with the title “Man Dies After Medical Incident During Police Interaction.”

An autopsy report highlighted Floyd’s medical history of heart disease, substance use disorder, and sickle cell trait. However, a court subsequently convicted a police officer of murdering Floyd by kneeling on his neck for more than 9 minutes.

The public learned the truth about Floyd’s death. However, a new study by the Institute for Health Metrics and Evaluation (IHME) has found that official statistics incorrectly classify or fail to report many thousands of deaths due to police violence.

The study, which appears in the journal The Lancet, estimates that there were 30,800 such deaths between 1980 and 2018, whereas the official figures report only 13,700.

In other words, 55.5% of all deaths from police violence in the U.S. may go unreported.

The analysis also shows that lethal police violence disproportionately affects Black people.

Once the researchers had accounted for age, the study revealed that the mortality rate due to police violence was 0.69 per 100,000 Black people, compared with 0.35 for Hispanic people of any race and 0.2 for white people.

The study shows that these racial disparities have remained largely unchanged since 1980.

“All of these statistics really outline the systemic racism that is driving police violence in the U.S.,” said co-first study author Eve Wool, M.P.H., of the IHME, in a video statement. She added:

“[P]olice violence is a public health issue, violence is a public health issue, and systemic racism is a public health issue.”

According to the IHME, its new analysis is the most accurate and comprehensive assessment of deaths attributable to police violence in the U.S. to date.

“So many different groups, from policymakers to advocates to commissions, have been calling for an approach to decrease the use of force,” said Prof. Alexes Harris, a sociologist at the University of Washington in Seattle. Prof. Harris was not involved in the research.

In a video statement, she said that many people had been “hiding behind” the lack of numbers on racial disparities in police violence. “This report now calls us to the carpet and asks us what [we are] going to do to address this violence,” she added.

During their study, the researchers compared the number of deaths published by the National Vital Statistics System, which compiles data from death certificates, with deaths reported in nongovernmental databases.

Journalists and independent research teams maintain these open-source databases, which are based on news coverage and public records of deaths following police violence.

The researchers used data from three such databases. These were:

The study authors note that physicians usually fill out the cause of death on death certificates. However, a medical examiner or coroner — who may or may not also be a physician — does this when there is a suspicion of crime or foul play, including police violence.

They point out that there are “substantial conflicts of interest” in the system for investigating deaths, including the fact that many medical examiners and coroners are embedded within police departments.

These conflicts of interest could deter these officials from noting police involvement on the death certificate, the researchers write.

“[F]or every George Floyd, hundreds of other Americans’ deaths after violent exchanges with police go unheeded, unacknowledged, and uncounted,” says The Lancet in an accompanying editorial. According to the author:

“Ensuring the veracity of data collection requires moving it out of the remit of law enforcement, which has been self-interested, voluntary, and incomplete.”

The editorial adds that reporting the cause of death can itself be “an issue of complicity in racist policing.” As an example, it notes that in the past, physicians have often given “sickle cell trait” as the cause of death for Black people in police custody.

Correspondence published in a previous issue of the journal highlighted the cases of 45 Black people whose deaths were falsely attributed to this condition. Unlike sickle cell anemia, however, sickle cell trait is almost completely benign.

The authors of the recent study also blame “militarized” policing in the U.S. for its high police-related death toll.

They note that countries such as the United Kingdom and Norway, which do not routinely arm their police officers, have very few deaths due to police violence.

The researchers conclude:

“To respond to this public health crisis, the [U.S.] must replace militarized policing with evidenced-based support for communities, prioritize the safety of the public, and value Black lives.”

The study authors note some limitations of their analysis. For example, they write that it did not address nonfatal injuries.

“This topic is crucial to understanding the full burden of police violence and should be examined in future studies,” they write.

Medical News Today asked senior study author Prof. Mohsen Naghavi, also of the IHME, about the possibility of inadvertent reporting bias in open-source databases such as The Counted.

He responded that Harvard School of Public Health and the U.S. Bureau of Justice Statistics have validated the accuracy of this kind of open-source data.

“While there may be bias present in any data source,” he added, “we believe that our methodology accounted for bias as best as possible with the available data.”