How to Count the People Police Kill

Harvard researchers argue that health agencies can tally deaths without law enforcement cooperation.

Ferguson Police Shooting Grand Jury Probe to Begin Tomorrow

Police officers holding riot shields and wooden batons stand and watch protesters during a night time demonstration in Ferguson, Missouri, on Aug. 19, 2014.

Photographer: Luke Sharrett/Bloomberg

Last year 14,249 people were murdered in the U.S., according to federal crime statistics. Those numbers reveal that 51 law enforcement officers were killed in the line of duty, and an additional 45 died accidentally.

What the official reports can’t tell us is how many people were killed by the police.

That question has drawn intense interest following the 2014 shooting death of Michael Brown by Ferguson, Mo., police. Since then, a drumbeat of highly publicized deaths at the hands of law enforcement, and furious debate over whether some were justified, has only inflamed the issue. Newspapers including the Guardian and the Washington Post have attempted to track public reports, with one total for this year exceeding 1,000, but no reliable accounting exists. The Department of Justice is considering how to improve national data on the use of force by police, but that's just in the planning stage.

Now public health researchers at Harvard are saying that federal agencies already have the power to make an official tally, with no need for legislation or the cooperation of local police departments. They propose making law enforcement-related deaths a "notifiable" condition to be reported weekly to the Centers for Disease Control, alongside afflictions such as tuberculosis and syphilis.

"Just as epidemic outbreaks can threaten the public's health, so too can police violence and impunity imperil communities' social and economic well-being, especially if civil unrest ensues,” the authors contend in an article published on Tuesday in the online journal PLOS Medicine. The number of Americans killed by police, according to the Guardian tally, exceeded the number of cases “for several diseases of considerable concern,” including measles, mumps, and malaria.

Such a tabulation could serve as a tool without necessarily focusing on whether a death was legally defensible. The reasoning for it is similar to arguments that doctors make to lift a ban on CDC and National Institutes of Health funding for gun violence research—that it's a public health issue.

“In public health, we count dead people,” said Nancy Krieger, professor of social epidemiology at the Harvard T.H. Chan School of Public Health and lead author of the article. “We count dead people in order to understand mortality rates” and to monitor changing trends.

The authors propose to track both civilian and officer deaths in law enforcement encounters. The CDC and the Council of State and Territorial Epidemiologists would start by recommending that such deaths be considered reportable, spurring state and local health agencies to collect the data. The cases could be noted by hospital staff or emergency medical technicians, but they could also be reported by anyone aware of the deaths, Krieger said. Whether this would result in comprehensive statistics, or run afoul of Congress's restrictions on gun violence research, is an open question.

In an earlier report published in the Harvard Public Health Review, Krieger and her colleagues used data from national mortality files to examine trends in “deaths due to legal intervention” over several decades. This data, compiled from death certificates and made available only years after the incidents occurred, show "excess risk for black men," she said.

Trends in police killings vary over time and by city.
Trends in police killings vary over time and by city.
Source: Krieger et. al, PLoS Medicine

The data also show that disparities between how often blacks and whites were killed by law enforcement have varied substantially over time and from place to place. From a public health perspective, Krieger said, such a variation means "you know it’s preventable."

The idea of placing reporting responsibility in the hands of health authorities isn't welcomed by James Pasco, executive director of the National Fraternal Order of Police. He says state and local health agencies aren't equipped to collect the data.

"The public health system of the U.S. is a shambles," Pasco said. His group favors mandatory reporting, both of violence against police and of deaths in custody, through the Justice Department. He added that violence against police is under-reported because some local departments don't report assaults or deaths of their members.

Society shouldn’t require the cooperation of police to understand the scope of the problem, Krieger said.

“Why are these police data only? Why are these not public health data? These are mortality records,” she said. “It seemed to me this should be a kind of data [collection] we should be able to do in public health, and be able to do it in real time."

Before it's here, it's on the Bloomberg Terminal. LEARN MORE