Examining the DOJ’s National Standards for Law Enforcement De-Escalation Training

Part 1: What is law enforcement de-escalation training, and why did Congress agree to fund it?

Prior to 2009, only eight states required de-escalation training for police officers. Today, at least 18 do— and 77 of the 100 largest U.S. cities have written it into their use-of-force policies. For most of that period, nobody could say with confidence what the training should include, how long it should last, or whether it worked. On Aug. 5, 2026, the Department of Justice (DOJ) released its National Standards for Law Enforcement De-Escalation Training to answer those questions, as directed by a 2022 law that passed the Senate without a single objection. This analysis covers what de-escalation means, what the research shows, and how Congress came to fund it. Part 2 will examine the standards themselves.

What Is Law Enforcement De-Escalation Training?

De-escalation shows up in hundreds of department policies and more than a dozen state laws, yet it has no agreed-upon definition. Federal law uses language identical to that of the National Consensus Policy and Discussion Paper on Use of Force. Endorsed by 11 law enforcement organizations, the paper defines de-escalation as “[t]aking action or communicating verbally or non-verbally during a potential force encounter in an attempt to stabilize the situation and reduce the immediacy of the threat so that more time, options, and resources can be called upon to resolve the situation without the use of force or with a reduction in the force necessary.”

In plain terms, de-escalation means that when an officer has time and distance on their side, they should take advantage of it. This might look like keeping space between themselves and the individual they are dealing with, staying behind cover, keeping the conversation going, and waiting for backup instead of moving in and going hands on. Officers themselves describe it about the same way, pointing to respect, patience, honesty about what will happen, and giving the person some say in how the encounter ends.

De-escalation is not a promise to avoid force, and it does not apply when an officer faces an immediate deadly threat. Nor is it one uniform curriculum: More than 50 programs differing widely in length, content, and teaching method compete for agencies’ business. And the concept is not new—effective officers have done this for decades without having a term for it. What is new is the effort to teach it purposefully and measure its effectiveness.

How Did Law Enforcement De-Escalation Become a Mandate?

Pre-2014, “de-escalation” was mainly a hospital term for the tactics medical professionals use to keep agitated patients from hurting them. While law enforcement borrowed bits of the idea, including a communication method called “verbal judo” (taught since 1983) and Memphis, Tennessee’s Crisis Intervention Team model for mental health calls (built in 1988), it was hardly something every patrol officer was expected to learn.

The 2014 shooting of Michael Brown in Ferguson, Missouri, changed that. The President’s Task Force on 21st Century Policing recommended de-escalation training in 2015. The Police Executive Research Forum published 30 guiding principles on use of force the following year. And a national consensus policy by several major law enforcement groups debuted in 2017. Things moved much faster after the death of George Floyd in 2020, with Colorado, Florida, Illinois, Indiana, Iowa, Massachusetts, Nebraska, Oklahoma, Tennessee, Utah, and Virginia all considering or passing legislation on the subject.

Policing in the United States is primarily a state and local responsibility spread across more than 17,000 agencies, and academies still spend far more classroom hours on firearms and defensive tactics than on communication or crisis response. A national review of the evidence found that the fear of getting hurt is a common obstacle to de-escalation taking hold, and early critics argued that asking officers to wait before reacting would simply give a violent suspect the first move.

What Does Law Enforcement De-Escalation Research Say?

The task force had almost no evidence to go on when making its recommendations. A 2020 systematic review of every de-escalation training evaluation published between 1976 and 2016 found 64 of them, nearly all in nursing or psychiatry.

The tests that followed have mostly pointed the same direction. In Louisville, a randomized controlled trial of a two-day course called “Integrating Communications, Assessment, and Tactics” (ICAT) found that use-of-force incidents fell 28 percent, while citizen injuries fell 26 percent and officer injuries fell 36 percent. The DOJ’s own research clearinghouse now rates the course as effective. In Tempe, Arizona, a randomized trial of a course the department built itself found that trained officers were 58 percent less likely to injure the people they used force on and no more likely to be hurt themselves. In Chicago, a randomized trial of a 16-hour course called “Situational Decision-Making” found that trained officers used force 23 percent less often and missed 49 percent fewer days due to injury, with no drop in gun recoveries or traffic stops. Officer injuries decreased or stayed flat in every trial that measured it. The worry that restraint endangers officers was reasonable before anyone checked; however, the data do not support it.

This research comes with a caveat. Only six curricula have been tested rigorously for their effect on use of force, and only two of them—ICAT and Situational Decision-Making—produced reductions. Several others improved how officers talked to people and how they felt about de-escalation without changing how often they used force. That is a strong argument for training and an equally strong one for being particular about which training program is used.

What Is the Law Enforcement De-escalation Training Act?

Two big policing bills were introduced in 2020: the Republican JUSTICE Act and the Democratic George Floyd Justice in Policing Act. Neither one passed, nor did a 2021 reintroduction of the Democratic bill. What did pass was narrower. When Sens. John Cornyn (R-Texas) and Sheldon Whitehouse (D-R.I.) introduced the Law Enforcement De-escalation Training Act (S.4003), more than 20 senators from both parties signed on. It cleared the Senate by unanimous consent in August 2022 before becoming law later that year. Supporters ranged from the National Fraternal Order of Police and the National Sheriffs’ Association to the National Alliance on Mental Illness.

The law tells the DOJ’s Office of Community Oriented Policing Services (COPS) to develop or identify training on four subjects:

• De-escalation and alternatives to force
• Responding safely to people in a mental health or suicidal crisis or living with a disability
• Working on Crisis Intervention Teams
• Connecting people to community services, including the 988 Suicide & Crisis Lifeline

It also requires the DOJ to write national standards, certify the programs that meet those standards, and authorize $124 million over four years (roughly $34 million to build the standards and $90 million in state grants). Every certified course must include pre- and post-tests as well as follow-up checks on whether officers use the skills on the job. The National Institute of Justice is tasked with studying implementation, and the Government Accountability Office must review the grants.

The funding rule is the provision that matters most. Once the standards are out, the DOJ must ensure that every dollar for de-escalation and crisis response training goes to a certified program—meaning the standards will shape what departments actually buy. Getting here took almost four years: Congress first appropriated the funds in fiscal year 2024, which allowed the DOJ to begin sending grants to states through the Safer Outcomes program. COPS put roughly $2 million toward the standards and certification system, which funded the development of 14 National Standards released in August of this year.

Why Did Congress Decide to Fund the Law Enforcement De-Escalation Training Act?

S.4003 passed the House 264 to 162 on Dec. 14, 2022, and was signed into law two weeks later. Members who voted no most likely objected to creating another federal grant program rather than to better training. Skeptics of new grant programs tend to make three points:

• The national debt has more than doubled over the past decade.
• Washington money tends to pull decisions away from local governments.
• Dollars often land in already well-funded places while the departments that need them most never see them.

This law answers those objections better than most. First, it requires nothing of anyone. Fewer than one in five of the roughly 700,000 sworn officers in the country work for the federal government, and the law leaves the other four under state and local control, attaching conditions only to federal money a department is free to turn down. It also pays for a practice backed by randomized-trial evidence and requires results to be measured so that the next Congress can decide whether the money was well spent. Finally, it targets a cost taxpayers bear either way: As Minneapolis’ $27 million payment to George Floyd’s family showed, a single unjustified use of deadly force can cost a city tens of millions of dollars in settlements—not to mention the human cost and damage to public trust.

Police officers carry the most direct coercive power within the government, and a government serious about using that power only when necessary owes its officers the training to tell the difference. Well-run de-escalation training means fewer injuries on both sides of an encounter and more officers free to work serious and violent crime. Part 2 will look at the National Standards for Law Enforcement De-escalation Training, examining what the DOJ wrote, how well it matches the research, and where it falls short.