A new report from ProPublica found that black teenagers were 21 times more likely to be shot and killed by police officers than white teenagers.

“The 1,217 deadly police shootings from 2010 to 2012 captured in the federal data show that blacks, age 15 to 19, were killed at a rate of 31.17 per million, while just 1.47 per million white males in that age range died at the hands of police,” according to the report.

To show just how great the disparity between the deaths of black and white teenagers is, ProPublica calculated how many more white teenagers would have to be killed to make the risk equal between the two groups. “The number is jarring – 185, more than one per week.”

Data on police shootings is woefully underreported, mostly because the data is provided to the FBI’s by the police stations themselves.

“Vast numbers of the country’s 17,000 police departments don’t file fatal police shooting reports at all, and many have filed reports for some years but not others,” reported ProPublica. For example, Florida’s police departments haven’t filed any reports since 1997 and New York City hasn’t filed any since 2007.

In terms of who the police are killing, ProPublica found that “41 teens 14 years or younger were killed by police from 1980 to 2012.” Of those, 27 were black, eight were white, four were Hispanic, and one was Asian.

The data also showed that it’s not just white cops doing the shootings. Black officers were responsible for just over 10 percent of all fatal police shootings, but 78 percent of those they killed were black. White officers did, however kill “91 percent of the whites who died at hands of police. And they were responsible for 68 percent of the people of color killed.”

Colin Loftin, co-director of the Violence Research Group and professor at the University of Albany, said that the FBI data is a minimum count of homicides by police, and it’s impossible to get an accurate measure of what puts people at risk without more and better records. He did say that what the data that is available shows is that race is “certainly relevant.”

“No question, there are all kinds of racial disparities across our criminal justice system,” Loftin said. “This is one example.”

The circumstances surrounding the killings is also underreported, but still proves there is a racial bias. The data shows that there were 151 instances in which police reports stated that the teens shot had been “feeling or resisting arrest at the time of the encounter.” Of those killed in these circumstances, 67 percent were black.

“The disparity was even starker in the last couple of years: of the 15 teens shot fleeing arrest from 2010 to 2012, 14 were black,” the report found.

Sometimes the police don’t actually list the circumstances surrounding a fatal shooting. ProPublica found numerous instances where the circumstances were listed as “undetermined.” In these cases, 77 percent of those killed were black.

Given that the reporting on police shootings are so underreported, it’s safe to assume that an analysis all of the fatal shootings that actually took place would show an even wider gap between the races of those killed. Until the police are forced to report all shootings, and do so accurately and thoroughly, this pattern is likely to continue, and unfortunately worsen.