High-profile killings including those of Michael Brown, Charleena Lyles, Tamir Rice, Stephon Clark and many others have brought the issue sustained national attention in recent years, but a lack of official data had prevented accurate estimates about the extent of the problem.
The new study relied on data from Fatal Encounters, a journalist-led effort, as well as the National Vital Statistics System, to analyze the period 2013-2018.
It found African American men and women, American Indian men and women, and Latino men all faced higher lifetime risks than their white peers.
"We think that there's ample evidence that police are a threat to public health in the United States," Frank Edwards at Rutgers University, the lead author of the paper published in the Proceedings of the National Academy of Sciences, told AFP.
"They're a lot more violent in communities of color than they are in white communities, and we need to take that seriously when we talk about police reform."
Black men had the highest fatality risk, with the researchers estimating 1 fatality from police use of force for every 1,000 male births.
"Those numbers are high. The chances of you being killed by police in this country is higher than winning a lot of scratch off lottery games," said Edwards.
The group was also 2.5 times more likely to be killed than white men, while black women were 1.4 times more likely to be killed than white women.
American Indian men were approximately 1.5 times more likely to be killed than white men, and American Indian women were about 1.6 times more likely to be killed than white women.
Latina women were however 17.5 percent less likely to be killed than white women, though Latino men were about 1.4 times more likely to be killed than white men.
Asian and Pacific Islanders had the lowest fatality risks.
The risks were even more pronounced for young men: more than 1.5 percent of all deaths of black men between the ages of 20 and 24 were caused by police, dropping slightly to 1.3 percent between 25 and 29.
Police violence is therefore one of the leading causes of death for these demographics, just behind cancer.
Edwards said that one of the study's main limitation was its reliance on media reporting, which likely resulted in a dramatic under reporting of numbers.
"Cases have to be reported in the news to be recorded in the data systems we're using, and there's almost certainly cases that were not reported in the news," he said, adding there was "an obligation" for authorities to step in.