Over Half of US Police Killings Mislabeled or Unreported, Study Finds | American Police
More than half of all murders involving police in the United States go unreported, with the majority of victims being black, according to a new study published in the Lancet, a peer-reviewed journal.
Research conducted at the Institute for Health Metrics and Evaluation at the University of Washington School of Medicine found that in the United States between 1980 and 2018, more than 55% of deaths, more than 17,000 in total, due to police violence were either misclassified or unreported.
The study also found that black Americans are more likely than any other group to die from police violence and are 3.5 times more likely to be killed by police than white Americans.
“The recent high-profile murders of blacks by the police have drawn worldwide attention to this urgent public health crisis, but the magnitude of this problem cannot be fully understood without reliable data,” said Fablina Sharara, researcher at the University of Washington School. of medicine and co-lead author of the study.
To fully understand the non-reporting of murders involving the police, the researchers compared data from the National Vital Statistics System (NVSS), a government database for tracking the American population, with non-government and open source databases. that follow police brutality. Open source databases consolidate information from reports and requests for public recordings, capturing a wider range of fatal incidents involving police.
âOpen source data is a more reliable and comprehensive resource to help inform policies that can prevent police violence and save lives,â said Sharara.
In total, the NVSS database misclassified nearly 60% of all fatal police encounters involving black Americans. The NVSS also missed about 50% of all Hispanic deaths involving police, 56% of all non-Hispanic white deaths involving police, and 33% of deaths involving non-Hispanic people of other races.
âThe inaccurate reporting or misclassification of these deaths further obscures the larger problem of systemic racism that is ingrained in many American institutions, including law enforcement,â said Sharara.
The document found that men die of police violence at higher rates than women, with 30,600 deaths involving police recorded among men and 1,420 among women between 1980 and 2019.
The researchers also noted the great conflict of interest inherent in tracking police deaths. Coroners are often integrated into police services and can be deterred from determining that deaths are caused by police violence.
âThe same government responsible for this violence is also responsible for reporting it,â Sharara said.
Previous studies have analyzed the underreporting of fatal police incidents and how black Americans disproportionately die as a result of police violence, but previous research has been conducted over much shorter time periods.
The new study published by Lancet is the longest study period to date, although the researchers recognized that future studies are needed to fully examine the impact of police violence in the United States, as the data collected does not ‘does not include police killed by civilians, police violence in US territories. or abroad, and used death certificates that did not identify non-cisgender people, in particular masking police violence against trans people.
Overall, an increased use of open source data collection is needed to document and understand disparities in police brutality by race, ethnicity and gender, the researchers said, allowing for more targeted changes in police and public safety protocols. The authors also recognized that more needs to be done to address violence involving the police.
âAs a community, we need to do more. Efforts to prevent police violence and combat systemic racism in the United States, including body cameras that record police interactions with civilians as well as de-escalation training and implicit bias training for police officers, for example, have been largely ineffective, âsaid the co-manager. author Eve Wool.