Study: 55% of Police Killings Don't Appear in Statistics

Police killings in the US represent an "urgent public health crisis," the size of which is not accurately reflected in the official statistics that guide health policy, according to a new study in the Lancet medical journal. Researchers say they compared data from National Vital Statistics System to open-source databases and found that an estimated 30,800 Americans died from police violence between 1980 and 2019, but more than 17,000 of the deaths, around 55%, were either unreported or misclassified. More:

• Racial disparities. The study found that Black Americans were 3.5 times more likely to die from police violence than white Americans, and around 60% of their deaths were misclassified in the NVSS database, the Guardian reports. "Inaccurately reporting or misclassifying these deaths further obscures the larger issue of systemic racism that is embedded in many US institutions, including law enforcement," says co-lead author Fablina Shahara.

• Reasons for the undercount. The researchers say that while clerical errors likely account for some of the discrepancy, coroners and medical examiners who work closely with police departments are likely to feel "substantial conflicts of interest" that make them reluctant to record police violence as a cause of death, reports USA Today. The researchers note that vital statistics systems in the US and other countries are usually considered reliable, but conflicts of interests arise when reporting police violence, "since the same state responsible for violence is also responsible for reporting it."
• Oklahoma was the worst in two ways. The study found that Oklahoma had the highest rate of underreporting fatal police violence, followed by Wyoming, Alabama, Louisiana and Nebraska. Oklahoma also had the highest death rate from police violence. Massachusetts, Connecticut, and Minnesota had the lowest rates of police killings.

• Men 20 times more likely to die from police violence. The researchers say men were around 20 more times likely than women to be killed by police, with an estimated 30,600 deaths in men and 1,420 deaths in women during the period studied, though they noted that data from death certificates only lists two genders, concealing the extent of police violence against transgender women.
• Calls for a uniform nationwide system. There are currently no national standards on identifying deaths caused by law enforcement violence. Former Washington, DC chief medical examiner, an expert on investigating deaths in custody, tells the New York Times that death certificates should have a box that can be checked to indicate law enforcement involvement. "If it’s a function of training, a function of bias, a function of institutional and structural racism—all the things we can assume—we can identify that once we have a uniform system," he says.

(More police stories.)