Monterey, California: A man acting erratically barricades himself in a truck, claiming to be armed with knives. He yells at an officer and threatens to kill him. Additional officers respond, talk to the man for at least an hour, buy him a Gatorade, and eventually get him to surrender without anyone getting hurt.
“Back in the old days, the plan would probably be to drag the guy out, and if he has a knife, hopefully he doesn’t stab you and things escalate,” said Lt. Mike Bruno of the Monterey Police Department.
For Bruno, the “old days” is pre-ICAT — before he and his department were trained in the concept of de-escalation through the Integrating Communications, Assessment, and Tactics program. Before they were taught about using time, space, dialogue, and critical thinking — but not force — to defuse many situations. Before the call for reform in policing that has made many departments rethink the old way of doing things.
Cambridge, Massachusetts: A man possibly on drugs is swinging a branch the width of a baseball bat at people on a populated riverway. Police clear the area, keep a distance, don’t point weapons directly at him, and try engaging him in conversation. The man eventually drops the branch but won’t step away from it. When other tactics don’t work, police allow an EMT who said he had previously dealt with the man to give it a try. He calls him by name, and the man leaves the branch behind and immediately climbs into the ambulance.
“Prior to ICAT, I definitely would have had my weapon up and I probably would have been telling him to, ‘Drop the stick, drop the stick, drop the stick,’” said Cambridge Police Officer Cameron Deane. “That would have been the conversation instead of, ‘What’s going on today?’ trying to get to the root of the issue and looking for other alternatives.”
Officers also would have told the ambulance worker to stay back and not get involved because the man was dangerous.
Camden, New Jersey: Police officers are called to the scene where a man is wildly and violently slashing a 12-inch knife in the air. The man, who is not in a clear state of mind, had threatened people inside a restaurant and then took off down the sidewalk, his arm moving in wide arcs, slashing the knife over and over. Camden officers create a wide perimeter around the man and follow him in that position down five city blocks, giving commands along the way. When officers from other police departments arrive on the scene and run up with their guns drawn, Camden cops tell them to holster up. After the knife slips from the man’s hand, an officer tackles him, and he submits. No one is hurt.
“If not for [de-escalation] training, we would have used deadly force on that individual, and it would have been your prototypical ‘lawful but awful’ situation,” said former Camden Police Chief J. Scott Thomson. “It would have been one in which, absolutely, we would have met all of the elements for the legal justification for using deadly force, but clearly there was the feasibility to do something else, and we did it.”
It’s no wonder that the typical police response of gun-out-voice-booming tends to end poorly when officers are dealing with someone in a mental health crisis, said Nikki Smith-Kea, a Criminal Justice Manager at Arnold Ventures who has experience helping law enforcement agencies improve their responses to people with mental illnesses.
“In every situation where mental health is involved, where it’s an escalation of a situation, the typical law enforcement response will never work: The command voice, the shouting orders will never work in that situation,” Smith-Kea said. “De-escalation comes in handy because it is giving you a set of tools — keep a distance, use a low tone, identify yourself, treat them like a human being, calm them down. De-escalation is necessary, it’s important, it does save lives, it can save lives.”
Smith-Kea isn’t one to make such assertions without having the evidence to back them up. And that evidence arrived this fall, when a groundbreaking new study found that after the Louisville Metro Police Department in Kentucky took ICAT training, there was a significant drop in the number of use-of-force incidents and injuries to both citizens and police officers.
“This study provides an answer to a question we’ve long had,” Smith-Kea said. “I think this is coming at a critical, critical time in this reform discussion, because this is another tool added to [officers’] tool belts — a more effective, better way of responding to individuals in crisis.”
Anecdotes and Evidence
Over the course of the yearlong study, Louisville Sgt. Justin Witt saw first-hand the positive effects de-escalation training had on officers.
“We’re seeing stories and watching body-camera footage where officers are giving more time for subjects to comply, they’re understanding that if no one is in immediate risk of injury we can slow down and we can be on these situations all day; we can spend more time making sure that at the end of this situation, we have a good outcome for ourselves and for the person we’re dealing with,” Witt said. “We’ve seen officers talk suicidal subjects off bridges, we’ve seen them use cover to speak to people who are armed with knives and baseball bats and give the person time to cooperate and put those weapons down. There are tons of anecdotal stories where officers have used the skills and effectively saved people’s lives.”
There are components of ICAT that are helpful even when dealing with situations that don’t involve a person in crisis, Deane said.
The Cambridge police officer recalled a situation that happened this summer after George Floyd’s death at the hands of Minnesota police. Protests had made their way to Cambridge, and the city was having an issue with fireworks being set off all day and all night. Residents at one housing development were upset, and police decided to step in to shut it down.
Three officers went on the initial response and were immediately outnumbered: A hostile crowd of 25 had gathered, and the officers recognized three or four of them as men who had been involved in past gun crimes or were known to carry guns. Instead of immediately shouting orders to disperse, one of the officers decided to use his ICAT training and spin the Critical Decision-Making Model: The three officers returned to their cruisers — even as the crowd taunted them — and looked at their options. They called for backup and then acted on their plan to not approach the biggest aggressors in the crowd but talk instead to those on the fringe and persuade them to leave. The plan worked. Eventually the crowd scattered; no one was hurt, and no one was arrested.
“We as police, we’re confident in our skills and we’re confident in our abilities, and we sometimes think, ‘To end this, I need to show who’s in charge’ and I feel like sometimes showing who’s in charge means taking a second to take that extra breath before you actually do something,” Deane said. “They fought smarter, not harder.”
Anecdotes, however, only go so far in the world of research, which is why the study on the Louisville Metro Police Department and ICAT is so important.
And that’s where Robin Engel comes in.
The researcher and professor of criminal justice at the University of Cincinnati had worked for years to find evidence to support whether de-escalation training was effective. It was a critical piece of the puzzle during Engel’s work reforming the UC Police Division following the fatal 2015 officer shooting of Samuel Dubose, an unarmed Black man. But there was no evidence to be found because no one had performed rigorous research on the topic.
So with funding from Arnold Ventures, Engel’s research center — the International Association of Chiefs of Police/University of Cincinnati Center for Police Research and Policy — took on the comprehensive study, which included a modified version of a randomized control trial, the gold standard in research. And she got buy-in from Louisville police as well as the organization that runs ICAT, the Police Executive Research Forum (PERF).
More than 1,200 LMPD officers were trained in ICAT between February and November 2019, and through surveys and the randomized control trial, Engel and her team assessed its effectiveness.
The surveys revealed positive changes in officer attitudes. For example, sizeable percentages of officers said the training was useful to them and thought it would improve their interactions with all citizens and in particular with people in crisis.
But it was the behavioral assessment that was the most significant — and critical to building an evidence base on de-escalation: It showed 28.1 percent fewer use-of-force incidents by police officers, 26.3 percent fewer injuries to citizens, and 36 percent fewer injuries to officers.
Engel admits to being surprised by the results — particularly because it’s often hard to detect significant change in research studies.
“That we were able to find immediate changes in officer behavior and that it was linked with a pretty strong research design gave me confidence that there was great opportunity with de-escalation training,” she said.
There’s still much more to learn. Now that Engel’s center has a rich data set from LMPD, it is looking deeper into the training’s effects, including what happens to racial disparities and what types of officer characteristics lead them to be more likely to use de-escalation tactics.
“While I had entered the study thinking we need more evidence, I left the study thinking we need a lot of agencies to test all different types of de-escalation training so that we can make things more precise and really have an impact,” Engel said. “Because this could save lives.”
The Community’s Role
Though there were a lot of positives to the study, research also showed that there were some problems in the decision-making model that didn’t correlate to the street, so the Louisville department is continuing to adapt. They streamlined the model so officers can recall in stressful situations what’s expected, and they reinforced training for front-line supervisors to make sure they are reminding officers on the street to slow down, communicate, and always have the principles of ICAT and the department in mind.
“We wouldn’t have known without the study what the deficiencies were,” Witt said. “A lot of times we look to those successes, but there’s as much to learn from the failures.”
Another recommendation Engel made in her report was for Louisville to bring the community into the fold by showing them the de-escalation training and the work being done within the police department, which could help rebuild the fractured relationship between the police and public.
“It breaks down barriers on both sides, so the police who may see citizens who are critical of their work as ‘they don’t understand,’ well give them an opportunity to understand, bring them in and let them be a part of the training,” Engel said. “And for the citizens who are there, it does give them a different perspective, and they’re able then to talk to others about the challenges that officers face and the ways they are changing the training.
“Then some of our strongest critics become our biggest advocates,” Engel said.
Turning to the community for input was something Engel did at UCPD and found to be very effective.
Following the Dubose shooting, the university put a team in place to focus on police reform and community relations and brought in S. Gregory Baker to guide that effort. Baker, who had worked for the city of Cincinnati for more than 30 years, was the public safety director following another tragedy: the 2001 fatal police shooting of unarmed Black teenager Timothy Thomas and the riots and investigation that followed.
As part of his new role, Baker worked with Engel to establish and manage a Community Advisory Council. It was critical, Baker said, to bring the community in to offer input and guidance regarding UCPD’s reform efforts — but not just be figureheads, “individuals who were just being trotted out to persuade the community that the right thing was being done. We wanted some real validity, some teeth, into that effort because we had to live in this community, but also for the efficacy of the effort.”
Engel made sure the council played a key role and was not just a public-relations move. Members weighed in on every aspect of reforming UCPD — from the different types of training officers would receive to the language used in policy. And they even sat in on de-escalation training, which Baker said was helpful to understanding why police do what they do.
“Most times when a shooting occurs, community members want to know, ‘Well why did you have to kill them? Why couldn’t you just shoot them in the leg?’” Baker said. “The majority of the African American community is very law-abiding, and if an individual goes out of that norm, they want the issue to be addressed — but not necessarily by killing or incarcerating that individual. So when you start to expose community members to why police do certain things, that changes some of their perceptions in understanding what’s going on. Any time you can make those connections, you are going to improve people’s understanding and therefore improve their trust and confidence in the police.”
Officials at other police departments that have enacted de-escalation training agree that the community is essential to the process.
When Thomson made headlines in 2013 by scrapping Camden’s police department and starting over with new policies and new hires, those new policies were signed off by community leaders, the police union, and the ACLU.
“Traditionally, use-of-force policies are written by cops for cops, and the mentality is the public doesn’t know and therefore they shouldn’t have a say,” Thomson said. “It’s kind of ironic because if you use force illegally, guess who’s sitting on your jury making a decision on whether you followed the law or not? An everyday citizen.”
Back in Louisville, the police department has already reached out to some in the community, bringing in members of Citizens of Louisville Organized and United Together (CLOUT) — a vocal critic of the police department — to evaluate and provide feedback on what officers have done and what they could do better. The group also sat in on ICAT training.
“We were really encouraged by it,” said Chris Finzer, chairman of CLOUT’s mental health and addiction issue committee. “The police officers as a group seemed interested in it and they seemed to accept the concepts behind it.”
But CLOUT made it clear they wanted to make sure the tactics that were being taught were reflected in policy and that officers would be held accountable if they didn’t follow that policy. Witt said those policies are now on the books.
While Finzer considers the de-escalation training worthwhile and a good tool, he said it’s not the whole answer. He worries that there is a culture imbedded in LMPD that a police officer should be able to go home at night “by any means possible,” citing the fatal Breonna Taylor shooting, which occurred a month after the study ended.
“The culture,” Finzer said, “has got to be changed.”
Cincinnati’s Baker agreed. “For the integrity of the profession and the betterment of police-community relations, until police leadership decides to police themselves and have a desire or a goal to change police culture, to move from the warrior to the guardian, to include empathy in policing, I think this reform stuff including de-escalation will only go so far.”
The Culture Change
Engel’s study and ICAT advocates agree with Finzer and Baker that de-escalation training can’t be a one-off.
“We can’t do the training one time and expect it to stick,” Witt said. “We don’t want an agency that is training in silos; we want to have these concepts blend with everything we do.”
And that essentially means changing the culture of policing and returning to PERF’s guiding principle that the sanctity of life should be at the heart of everything an agency does — and that everyone, not just police, should be able to go home at the end of the day.
“ICAT could not just be a check-the-box training or a paper exercise,” said Thomson of Camden. “De-escalation had to be the culture of the organization.”
That culture shift, which included making de-escalation a part of policy and holding officers accountable, is responsible for Camden’s dramatic turnaround, Thomson said. In 2013, the Camden Police Department had 95 allegations of excessive force. In 2018, it had three — a 97 percent drop.
Coming out of this summer’s protests, there have been calls to defund the police or reimagine public safety so there’s a reinvestment in alternative resources and responses. But Walter Katz, Vice President of Criminal Justice for Arnold Ventures, says that regardless of whether the policing footprint shrinks, there will always be a need for police in our society — and that means it’s critical to ensure reform efforts are effective.
“Since we are a society with more than 300 million handguns, there is good reason to still have police,” Katz said. “But if we’re going to have police, how do we want them to function, what are their policies supposed to be, how are they supposed to be trained and hired, held accountable? All of these are vital questions that people have been asking for quite a while but have taken on a real new urgency. And being able to point to reforms and say the research tells us that this reform works — that de-escalation works and that the public benefits — is an important piece to add to the police reform conversation.”
Read Part I: ‘You Can Call Me V’: The Rise and Promise of De-escalation in Policing