Death certificates undercount police-involved fatalities, study says

Death certificates in the United States undercounted more than half of people killed by police in 2015, with Oklahoma failing to note any of its officer-involved fatalities on the official records, according to a recent study.

Touting their work as the most accurate count to date on the number of people killed by police in the U.S., researchers at Harvard University estimated that 1,166 law-enforcementrelated deaths occurred in 2015, including 37 in the Sooner State.

“With better death certificate data on officer-involved fatalities, we could answer some important questions,” said Justin Feldman, doctoral student at the Harvard T.H. Chan School of Public Health and lead author of the study. “How does Oklahoma compare to other states in terms of death rates? How big are racial and economic inequalities in the risk of being killed by police?”

Experts suggest that without a system that consistently tracks officer-involved deaths, policymakers are hamstrung in bringing about needed reforms.

At the same time, Americans are left to track the data through media reports or else mine the information from opaque government websites that frequently undercount fatalities where “legal intervention” is involved.

Feldman and his colleagues compared what they considered two incomplete lists of law enforcement-related deaths. One list was The Guardian newspaper’s dataset called “The Counted.” The other was the U.S. Centers for Disease Control and Prevention’s National Vital Statistics System, which is based on state death certificate data.

In looking at the degree of overlap, the researchers concluded that the NVSS misclassified more than 55 percent of all police killings. The fatalities were most likely to be misclassified if the death was not from a gunshot wound or if it occurred in a lowincome county.

In 2015, the Sooner State ranked No. 4 per capita and No. 6 in total people killed by police in the Unites States, according to The Guardian.

A government database that kept accurate records would hold more weight with the public than those created by media websites, Feldman said.

“We might also link death certificates to other medical records and see that many of these cases involve people with mental illness,” he said. “That could again lead to better training for officers in crisis intervention, and encourage police departments to collaborate with mental health professionals who are better prepared to use non-violent de-escalation tactics.”

Keeping track

The state Medical Examiner’s Office did not respond to requests for comment, nor to confirm or dispute Harvard’s findings in Oklahoma.

The Oklahoma State Bureau of Investigation only tracks fatal officerinvolved shootings. OSBI no longer tracks Taserrelated deaths, medicalrelated deaths or deaths from law enforcement vehicle chases.

OSBI recorded 33 officer-involved fatalities in 2015.

Through Nov. 15 this year, the bureau recorded 23 such deaths. About 2:40 a.m. that morning in the 1400 block of SW 20, Oklahoma City police shot and killed Dustin Pigeon, who was threatening suicide and disobeyed orders to drop the lighter fluid he had doused himself with and attempted to ignite, authorities said.

“We learn about most fatal officer-involved shooting deaths through the media,” OSBI Spokeswoman Jessica Brown said. “Once we identify a qualifying death, we add it to a spreadsheet. We request the (Medical Examiner) report from the ME’s office at the end of the year.”

OSBI does not receive information from local law enforcement agencies, and the Medical Examiner’s Office is not involved in the data collection process, Brown said.

“The information is no longer forwarded to any federal agency,” she said. “In the past, our unit received federal funding from the Bureau of Justice Statistics to collect arrestrelated deaths information, but they no longer fund data collection.”

Oklahoma extreme

According to the Harvard study, police-related deaths were misclassified at a rate of more than 60 percent among a few groups — blacks, people in low-income counties, people under 18 and those killed by something other than a firearm.

“I was surprised by how much it varied by state and county,” Feldman said. “Oklahoma is an extreme in terms of not reporting.”

Conversely, the federal government tracks police deaths at the hands of civilians and makes the information easily accessible to the public. The information is published online through the FBI’s Law Enforcement Officers Killed and Assaulted Program.

“The goal of the program is to provide relevant, high quality, potentially lifesaving information to law enforcement agencies focusing on why an incident occurred as opposed to what occurred during the incident, with the hope of preventing future incidents,” the FBI says.

As of Nov. 16 this year, 39 law enforcement officers were feloniously killed, according to the FBI. Forty-eight law enforcement officers had been reported accidentally killed.

“Getting the full picture on the other side is piecemeal, a little here and there,” said Christopher Hill, who teaches criminal justice at the University of Oklahoma. “My perception at least is that there’s not been one of these shootings in Oklahoma that has been a bellwether that has rallied attention around the state.

“And from my reading on officer-involved shooting fatalities, the data are a little bit challenging to come by.”

Seen as a public health issue, some call for a uniform collection of data across the country.

“The system needs to be overhauled,” said Samuel Walker, emeritus professor of criminal justice at the University of Nebraska at Omaha. “We have very detailed data on death related to smoking, cancer, every disease imaginable. Traffic data on deaths without seatbelts. This is an extremely important policy issue and we have the right to know what the facts are.”

Complete and accurate data would enable the public to compare police departments, and look into differences between policies and oversight, Walker said, adding that a renewed push for such information has come about because of the many high-profile cases of officer-involved deaths around the nation.

“We are really beginning to find out the deficiencies,” he said. “I just think there’s been a deliberate indifference to police shootings and unwillingness to face up to the facts.”

The study was published Oct. 10 online in PLoS Medicine.