Harvard Researchers: Make Police Killings A Matter Of Public Health

In this Oct. 20, 2014, frame from dash-cam video provided by the Chicago Police Department, Laquan McDonald, right, walks down the street moments before being shot by officer Jason Van Dyke in Chicago. Van Dyke has now been charged with murder. (Chicago Police Department via AP)

In this Oct. 20, 2014, frame from dash-cam video provided by the Chicago Police Department, Laquan McDonald, right, walks down the street moments before being shot by officer Jason Van Dyke in Chicago. Van Dyke has now been charged with murder. (Chicago Police Department via AP)

By Richard Knox

Every week, it seems, a new police killing enters the news stream, sparking outrage, breeding cynicism, fraying still further the social compact between police and communities.

The issue reached a new peak just this week, when the Justice Department announced a probe of the Chicago Police Department, the nation’s second-largest, to determine if there has been “systematic misconduct.” The investigation comes in the wake of social unrest and the recent firing of the police commissioner, after two police killings there.

In fact, police killings happen in America far more often than once a week.

The best available data come from news organizations, such as a website launched earlier this year, ironically enough, by the British newspaper The Guardian. They show that U.S. civilians die at the hands of police nearly three times a day. So far this year, 1,055 Americans have been killed by police, by The Guardian count. The Washington Post has tallied up 913 people “shot dead by police this year.”

About 120 law enforcement officers were killed in the line of duty last year, according to the National Law Enforcement Officers Memorial Fund.

Run through the Guardian’s website of civilian police-related deaths, called “The Counted,” and you’ll see that many of these everyday police killings involve suspects who were armed and menacing. The 14 people killed in the past week include the San Bernardino shooters and others who reportedly were threatening police officers. These are not the kind of cases that generate Black Lives Matter protests, although they shouldn’t necessarily be classified as justifiable use-of-force without careful investigation, either.

“No act of Congress is needed,” they write. “No police departments need to be involved. Public health agencies can do the job.”

The typical investigation focuses on the circumstances and actions in a specific case. But larger forces may be driving the phenomenon as well, forces that don’t get identified in case-by-case investigations. And that’s just the point of a new proposal, out Tuesday, that makes a strong case for collecting data on law enforcement-related deaths a matter of public health.

The authors, from the Harvard School of Public Health, assert that these killings — both by and of police — should be “notifiable” to public health agencies, just like homicides, suicides, many infectious disease deaths, work-related fatalities and injuries, and death by poisoning, fire and spinal cord injuries. That means they should as a matter of law be reported to health departments; currently police-related deaths are reportedly voluntarily (or not).

The Harvard researchers write, in the journal PLoS/Medicine, that death and injury due to police encounters are “a matter of public health, not just criminal justice, as is the occupational health of law-enforcement officials.”

“Deaths are part of our bailiwick,” lead author Nancy Krieger says.

She argues that only by compiling data on a national basis (but with details specific to local jurisdictions) can public health scientists identify time trends, racial-ethnic and geographical disparities, and other relevant indicators. And only then can they put these events in context with, say, the racial makeup of communities and police forces.

Such data now are fragmentary and delayed. Using what’s available, the researchers charted arrest-related deaths in eight U.S. cities at the top of The Guardian’s rankings, along with some recent hotspots such as Ferguson, Missouri.

“We show enormous variability over time among the eight cities,” Krieger says.

Take, for instance, the critical issue of black-white differences in who dies as the result of a police encounter.

Not surprisingly, blacks are strongly over-represented. But interestingly, the black-white ratio has declined over time, from nearly 8-to-1 in 1965 to 3-to-1 in 2005.

Even more interesting is the variation from one city to another. In New York City, the five-year average ratio as of 2005 was 19 black police-related deaths for every white killed. In Cleveland the ratio was about 6-to-1 for the same time period. And black-white ratios bounced around quite a bit in given localities from one five-year time period to another.

“Anyone in public health knows that when you see that kind of high variability, it shows you that there’s something that’s preventable,” Krieger says. “It’s being shaped by special contexts and it doesn’t have to be a fixed ratio.”

That kind of insight opens the door to analysis of what forces are shaping the rate in City X versus City Y — and then perhaps to preventive interventions, whether focused training of police personnel, work on improving police-community relations or tackling an off-the-charts density of illicit firearms on the street through buyback programs or better regulation.

Krieger likens it to the current reframing of the opioid use epidemic, from a War on Drugs criminal justice approach to a public health perspective. “It doesn’t say there aren’t criminal justice issues,” she says. “But there’s an expansion of views and possibilities for intervention outside the criminal justice approach.”

In the wake of Michael Brown’s death in Ferguson; Eric Garner’s “I can’t breathe” choke-hold death in New York City; Freddie Gray’s death-in-custody in Baltimore; 17-year-old Laquan McDonald’s death by 16 shots in Chicago; unarmed Walter Scott’s fatal shot-in-the-back after being stopped for a broken taillight in North Charleston, South Carolina, and others, there seems to be an awakening to the need for more systematic data collection on such deaths.

In October, U.S. Attorney General Loretta Lynch said the Justice Department would launch a pilot project to collect data on arrest-related deaths.

The Harvard researchers claim their proposal is the first to inject public health into the mix. And they say it could be done fairly simply — hypothetically, at least.

“No act of Congress is needed,” they write. “No police departments need to be involved. Public health agencies can do the job.”

Specifically, public health experts could work with a group called the Council of State and Territorial Epidemiologists to set up a uniform reporting system that could be implemented by public health agencies in each state, usually under existing regulatory authority, Krieger says.

“It’s an interesting idea. I’m hoping it will stimulate discussion,” says Dr. Albert DeMaria, immediate past president of the council and Massachusetts’ state epidemiologist. “In theory it’s feasible. States could do it on their own. In actuality, there’ll be different reactions. It’s certainly not a slam-dunk.”

That’s immediately apparent from a brief sampling of opinion in the police and criminal justice fields.

“I think it’s misguided,” says Bill Johnson, executive director of the National Association of Police Organizations, which represents rank-and-file groups such as the Boston Police Patrolmen’s Association. “The best way to reduce the number of deaths by police is to follow the instructions of the officer in any kind of confrontation. I don’t have a lot of hope that academics from Harvard would publicize that as an easy and quick way to reduce deaths by police.”

James Alan Fox, a professor of criminology at Northeastern University, is skeptical that collecting data on police-related deaths by public health agencies will do much good.

“I would agree that more data can only help,” he adds, “but we’ve got to be careful. These are not high numbers, so they’re very volatile.” That is, they can vary greatly from year to year and place to place, leading to wrong conclusions.

“Once you’re aware of that and don’t start making conclusions on the basis of highly volatile data, you’re fine,” he says. “But declaring it a public health issue isn’t going to make things better.”

Please follow our community rules when engaging in comment discussion on this site.