Jabez Chakraborty was experiencing a mental health crisis. A family member called 911 requesting an ambulance, but NYPD officers were first to respond. The moment an officer came in the front door, Chakraborty grabbed a large kitchen knife, charged the officer, and was shot four times.

Could this tragic outcome have been avoided? The following adapted excerpt from my forthcoming book doesn’t answer that question, but it does describe a better approach to mental health crises.

Does the absolute risk of violence during mental health crises justify deadly force? Recall that first-episode psychosis is the most dangerous phase of mental illness. Psychosis is hard to hide, but psychosis alone doesn’t justify the use of deadly force. Deadly force is allowed only if “the officer has probable cause to believe that the suspect poses a significant threat of death or serious physical injury to the officer or others.”

A 2024 article reviewing 22 studies of first-episode psychosis found a 13.4 percent prevalence of “any violence.” That percentage may be “significant,” but the article’s definition of “any violence” included conduct that posed no risk of death or serious physical injury, like merely yelling at someone. Only 2.2 percent of violence resulted in “serious injury.” Deadly force is almost always an overreaction.

In some places, when a person in crisis has no weapon and poses no immediate threat, a response team comprised solely of mental health professionals is dispatched. For decades, we have known that this crisis response model is safer for everyone and has better outcomes.

The trailblazing Crisis Assistance Helping Out On The Streets (CAHOOTS) program has operated in Eugene and Springfield, Oregon, since 1989. A two-person team consisting of a medic and a crisis worker are dispatched in response to mental health calls. Police backup is needed in only a tiny fraction of cases (approximately 2 percent), and no CAHOOTS staff member has ever been seriously injured. This should not be surprising in light of the research finding that there is only a 2.2 percent chance of serious injury during even the most dangerous type of mental health crisis, first-episode psychosis.

According to the organization behind CAHOOTS, the program’s success depends on the fact that 75 percent of responders have lived experience of incarceration, substance use, neurodivergence, houselessness, and other forms of oppression. Lived experiences like these foster understanding and trust. Some experts believe that all mental health crisis response teams should be led by peers.

In San Francisco, there is a peer on every three-member response team. In addition to having a deeper understanding of mental illness, peers are more likely to live in the same neighborhoods as, and share other demographic characteristics with, the people they serve.

Notwithstanding the well-known success of CAHOOTS, only in recent years have a significant number of jurisdictions begun to implement similar approaches to crisis response. In many areas, [including New York City,] there aren’t nearly enough specialized teams to meet demand. Recently approved federal funding for programs like this should help, but federal funding is precarious and that money won’t be enough without substantial investments at the state and local levels.

Ironically, CAHOOTS has saved Oregon millions of taxpayer dollars because the program is less expensive than a police response and much less likely to result in jail or prison. Savings in a single year have been estimated at $8.5 million. Here is a cheap and easy way to reduce killings by police, but few places have adopted it.

It is impossible to identify with certainty any specific individual whose life was saved by a non–law enforcement crisis response. We can never know for sure what would have happened if the police had arrived first. That said, the following true story illustrates how peer-led crisis response can dramatically reduce the chances of an encounter turning deadly.

Vania Mendoza, a peer supervisor with San Francisco’s crisis response team, told me this story. It begins with a striking juxtaposition: at a street corner atop one of those beautiful hills in San Francisco, a young Black man named Tyler [not his real name] is yelling loudly at pedestrians and swinging around a large, serrated knife. Someone calls 911. Because Tyler is wielding a knife, dispatch sends law enforcement, not a crisis team. The police will arrive just a few minutes later.

But by an almost unbelievable stroke of good luck, Vania’s crisis response team happens to be driving on that block at that moment, in between its own assignments. When Tyler, who is a regular client of the team, recognizes their vehicle, he throws down the knife even before the vehicle stops, comes over, and starts venting to the team about his stresses and struggles.

They give him some snacks. Vania explains that people with mental illnesses can get “hangry,” just like everyone else. In about two minutes, he’s “de-escalated himself.”

Tyler is calm now, but he’s not out of the woods just yet. Three police cars with lights blaring arrive nearly simultaneously. An officer explains that they got a call about a knife. Somewhat hilariously, Tyler now feels comfortable enough to start directing the scene, telling the cops that they can leave: “We’re good here. I’m fine. There was a problem, but it’s not a problem anymore.”

Understandably, the cops still want the knife. Vania points to the knife a dozen or so feet away, under another vehicle. Now Tyler wants to be helpful, so he starts walking over to the knife to get it for the police. Alarmed, Vania sees what’s happening and distracts Tyler by restarting their conversation.

The cops retrieve the knife, and the crisis response team gives Tyler a ride back to his neighborhood. It’s quite possible that the crisis response team saved Tyler’s life once, or maybe twice. Either way, this story illustrates how effective a peer-led, non–law enforcement team can be in defusing a potentially dangerous situation.

No injury, no arrest, no hospitalization.

The cited material above is an adapted excerpt from Through the Fire: How People with Mental Illness Are Empowering Each Other. Copyright © 2026 by Fredrick E. Vars. Used by permission. All rights reserved.

© Psychology Today