Unarmed 911 Teams Now Operate in Nearly 100 US Cities

Five years after the police killing of George Floyd, programs that dispatch unarmed civilians—rather than police officers—to certain non-violent 911 calls have spread to nearly 100 American cities, according to NPR. In 2020, only six such programs existed in the United States. Today, they answer hundreds of thousands of calls a year, including mental health crises, trespassing complaints, and welfare checks, while reporting one of the strongest safety records of any recent public-safety reform.

The expansion reflects a broad recalculation among city governments: that a call like a man refusing to leave a parking lot with his belongings does not always require an armed officer. But as the movement matures, it faces a harder question—how many 911 calls these teams can safely absorb, and whether they can address the poverty and homelessness that drive so many of the crises they are sent to resolve.

How Durham Became a Model

Durham, North Carolina’s Holistic Empathetic Assistance Response Team (HEART), launched in 2022, has responded to more than 25,000 911 calls and now operates citywide. The program pairs a peer support specialist such as Rob Crisp with a mental health clinician like Allie Craft, and dispatches them to call types the city identified as rarely ending in violence—including certain mental health crises, trespassing, panhandling, intoxication, and welfare checks.

“I can relate to not having anything,” Crisp told NPR, describing how he calmed a tense man during a recent wellness check. “I was like, ‘We’re not trying to bother you, but hey, we do have some supplies while we are here. … Hey, I don’t have a gun. I don’t have a badge. I’m here because I feel for you, brother.’”

Ryan Smith, director of the Durham Community Safety Department, framed the program’s measure of success pragmatically. “Honestly, I would say if you can send a social worker to a call that you could send a law enforcement officer and you got the same results, then it’s already a success,” he said.

A Growing Body of Evidence

The research so far points to measurable benefits. A preliminary study from Duke University researchers found that Durham’s program increased 911 calls—a potential sign of growing public trust. A 2026 study by RTI International found the program associated with fewer arrests, with most police officers finding it helpful and police rarely called for backup.

Nationwide, the safety data is striking. Across programs, responders handled calls without needing emergency police backup in 99.8% to 99.96% of cases, felt safe on about 99% of calls, and managed calls without referring to police in roughly 97% of cases, according to NPR’s reporting. In Durham specifically, responders needed emergency police backup on just 0.04% of calls in 2024 and referred about 2.6% of calls to police.

Other cities show similar promise. A 2022 Stanford University study of Denver’s STAR program found evidence it reduced reports of certain crimes by more than a third. A 2025 study found that an Oregon program reduced the likelihood a 911 call ended in an arrest and increased access to medical care.

Where the Model Falls Short

Yet in some cities, fatal police shootings have exposed the limits of alternative response. In New York City, the 2024 fatal police shooting of Win Rozario, a 19-year-old experiencing a mental health crisis, raised questions about whether the city’s B-HEARD program was answering as many calls as it could. A 2025 audit by the Office of the New York City Comptroller found that the majority of the city lacked access to the program, citing limited hours, geography, and eligible call types.

In Baltimore, an investigation by The Baltimore Sun after the 2025 police killing of Pytorcarcha Brooks found that the number of 911 calls diverted to mental health services had fallen by more than 50% in recent years—even as total mental health calls surged. The city is now considering modeling a program after Durham’s, with officials describing it as a potential “third branch” of the city’s public safety apparatus.

Funding fragility is another risk. CAHOOTS, the nation’s first community responder program, launched in Eugene, Oregon, in 1989 but lost its Eugene service in April 2025 because White Bird Clinic said it could not financially sustain a full-service mobile crisis service. Structural obstacles are also emerging: a 2026 report from Seattle’s Inspector General found that the city’s police guild contract “throttles” its CARE crisis-response teams, restricting call types and letting police override dispatchers.

The Debate Over How Big These Programs Should Be

Measuring success by call volume is tempting but incomplete, researchers warn. Durham HEART handles roughly 28 calls per 1,000 residents, ranking it in the top 10 U.S. programs by call volume. “It is the easiest data point to get and to compare programs,” said Amos Irwin of the Law Enforcement Action Partnership, a criminal justice reform nonprofit. “It’s a shortcut. It doesn’t tell you the quality and whether they manage to make a long-term difference in the lives of the people that they served.”

Because almost all these programs are only a few years old, it is difficult to say what a reasonable ceiling is. “Many of us are still trying to figure that number out, but we anticipate it’s far more than what any city is handling at this point,” said Mariela Ruiz-Angel, director of community safety initiatives at Georgetown Law’s Center for Innovations in Community Safety.

A recurring limitation, Irwin notes, is that programs are often authorized to respond only to clear-cut mental health crises. “No one calls and says somebody is in a mental health crisis,” he said. “They call and say there’s somebody panhandling really aggressively here. There is somebody loitering around my store, refusing to leave.”

The Root-Cause Question

Even as programs resolve immediate crises, many of the people they encounter lack basic needs. Durham is betting on a more ambitious path: a Stabilization Services division launched in 2024 to connect people to housing, medical care, behavioral health care, and food.

“The aim of the 911 response is to de-escalate the situation and leave everybody in a somewhat stable place. That does not mean that whoever’s left there has a connection to the right resources,” said Anise Vance, who leads the initiative. “Those services have to include some of the basic things in life.”

Ruiz-Angel, who founded Albuquerque’s community response program, said every such program eventually faces a choice: stay in a narrow lane managing immediate crises, or try to tackle their deep-seated causes. “They all come down to poverty issues, and there’s just not enough resources at this point,” she said.

Political resistance has also shaped the programs’ evolution. Patrice Andrews, Durham’s former police chief who helped launch HEART, recalled officers’ early concerns—both about protecting unarmed responders and about the perception that the effort was “this ultra-uber-liberal idea being pushed down our throats.” Over time, she said, police came to see the team as a fourth and equal wing of 911 response, aided by sharing the same dispatch radio channel. “They’re not police officers, but they are a part of the public safety net,” she said.

What to Watch

The next phase of the movement will test whether its growth can outpace its fragility. As more cities—including Baltimore—move to build Durham-style programs, the open questions are whether they can secure durable funding and the political space to handle a wider range of calls. The ceiling on how many calls unarmed teams can safely absorb remains unknown, and the deeper questions about housing and poverty that generate so many crises remain far from solved.