Why Does This Exist?

In cities around the country, 911 dispatchers face the same limitation every shift: three buttons. Police. Fire. EMS. A woman calls because her brother is having a psychotic episode in the backyard. Police. A business owner calls about a man slumped in the doorway, intoxicated and unresponsive. Police. A resident reports a neighbor screaming at no one on the sidewalk. Police. A school reports a student threatening self-harm. Police. Needles on the playground. Police.

The University of Chicago’s Health Lab found that over 75% of 911 calls dispatched to police across America do not involve the type of serious crime or public safety threat that requires an armed response.¹ In Atlanta, researchers determined the city’s crisis response program could have been the appropriate responder to more than 600,000 calls over four years to which 911 operators instead dispatched police.²

The Community Safety Department exists because no single program solves this mismatch at scale. A mobile crisis team handles behavioral health emergencies. A violence intervention team handles retaliatory conflict. An outreach team handles chronic homelessness. A clean team handles biohazard removal. Each of these programs addresses a real piece of the problem. But when they live in separate agencies with separate budgets, separate chains of command, and no shared dispatch infrastructure, the gaps between them become cracks that people fall through.

The CSD is an institutional answer to an institutional problem: the 911 system was built around three departments, and the calls that don’t fit those departments need a fourth.

The Mismatch in Three Dimensions

The case for a CSD rests on three interlocking failures in the current system. Each one, standing alone, justifies a program. Together, they justify a department.

The Skills Mismatch

An officer responding to a psychiatric crisis typically has three options: arrest the person, transport them to an emergency room, or leave. None of these options treat the underlying condition. None prevent the next call. Sacramento County Sheriff Jim Cooper stated the problem plainly: “Being mentally ill is not a crime and we can’t be the answer. Law enforcement officers are not trained mental health professionals. We’re not psychiatrists and psychologists. We wear the badge, we carry the gun, we deal with crime, not mental health crises.”³

A Denver police officer, interviewed for an NYU study, described the stakes of that mismatch from the other side: “Why would you wanna put us in a situation where more than likely we may fail, and if we fail, our failure is epic, right? Our failure is front page news.”⁴

The skills gap extends beyond mental health. An officer responding to a man experiencing chronic homelessness cannot connect him to housing. An officer responding to needles on a playground is not trained in biohazard removal. An officer directing traffic at a fender bender is using arrest authority and a sidearm for a task that requires neither. In each case, the person calling 911 has a legitimate need, and the system sends someone whose training, tools, and authority are designed for a different problem.

The Time Mismatch

As the Denver officer told NYU researchers, the mismatch is felt on both sides.⁴ The calls that do not require police often require more time than police operations allow.

In Durham, police were losing more than 10,000 officer hours to mental health calls (earlier reports cited 5,500 hours before the department’s expansion).⁵ Those are hours officers could spend investigating robberies, solving homicides, or patrolling high-crime areas. In Oklahoma City, diverting mental health calls from police produced a 57% reduction in mental health dispatches in a single year.⁶ Albuquerque police leadership credits its CSD directly with enabling crime reduction by freeing officers to focus on felony cases.⁷ Austin Police Association President Michael Bullock testified before city council: “It’s time that we work towards getting law enforcement out of mental health. We have never claimed to be the experts, but yet we have been charged with the responsibility of responding to mental health crisis.”⁸

Durham’s department routinely spends thirty minutes or more on a single encounter, and over an hour on suicide-related calls.⁹ That extended engagement is part of the model’s design.

Minneapolis Deputy Chief Erick Fors: “Feedback from the rank and file has been very positive.”⁹ In St. Petersburg, when the crisis program launched alongside police, “No one came to me and fought the program,” said Megan McGee, the police captain who oversaw the partnership.¹⁰

The Revolving Door

The same individuals cycle through 911 repeatedly. A St. Petersburg police officer described the pattern: “I know I’m going to come back to that house in a month and it could be worse.”¹⁰

Guilford County, North Carolina, tracked a group of frequent callers generating repeated 911 calls. After deploying a small team of social workers to provide sustained engagement, their non-emergency 911 calls dropped from 344 to four in a single month.¹¹ Durham’s care navigation segment offers follow-up within 48 hours and ongoing case management.¹² St. Petersburg’s crisis program “never closes a client. They do as much follow-up as needed so that the client can get engaged with long-term services.”¹⁰

A Washington Post analysis of court records found that local governments representing the 25 largest police and sheriff’s departments paid out more than $3.2 billion in liability costs over a recent decade, with a substantial share involving encounters with individuals in behavioral health crisis.¹³

Why a Department, Not Just Programs

A city could address each of these mismatches with standalone programs: a mobile crisis team here, a clean team there, an outreach initiative somewhere else. Many cities have done exactly that. The CSD model goes further, consolidating those programs under a single institutional roof.

When a mobile crisis team lives in the public health department, a violence intervention initiative lives in the mayor’s office, and a clean team lives in a nonprofit contract, there is no shared dispatch infrastructure, no unified training standards, no single budget line a council member can evaluate, and no institutional entity accountable for the overall question of whether the right responder reached the right call.

Albuquerque’s experience illustrates the alternative. Under one department, the community safety responder who responds to a “down and out” call and discovers the person also has a substance use disorder and is experiencing homelessness can access the department’s outreach division, needle pickup capacity, and violence intervention resources through the same chain of command.¹⁴ The responder does not need to navigate a referral across three separate agencies with three separate eligibility criteria.

Durham Director Ryan Smith described the department’s functional areas as “pieces of a larger idea.”¹⁵ One handles the immediate field response. Another handles calls involving weapons or threats. A third resolves calls by phone from inside the 911 center. A fourth follows up and connects people to lasting services. In July 2025, the department reorganized these into two direct-service divisions (911 Crisis Response Teams and Stabilization Services), but the four functional areas continue. Each addresses a different piece of the problem, and they share a director, a training culture, and an institutional identity.

National polling found 62% of voters prefer the unified department model over separate programs.¹⁶

The Calls That Fall Between Departments

The CSD addresses calls that do not cleanly belong to police, fire, or EMS.

Harris County Commissioner Rodney Ellis described one such call: a young woman “took a bottle of pills out of her mother’s medicine cabinet and locked herself in her bedroom.” Her mother called 911 but feared “what would happen when the dispatcher sent a sheriff’s deputy to her home,” worrying “that the presence of an armed law enforcement officer inadvertently could escalate an already volatile situation.”¹⁷

Polling reflects the depth of this problem: 84% of voters agree that CSDs “meet unmet needs because some people who need help are too scared to call 911 because they’re scared of the police.”¹⁶ And 81% agree these departments “reduce the likelihood that a mental health crisis will result in injury or death.”¹⁶

The CSD does not replace police for calls that require police. It provides a fourth option for the calls that do not.

What We Don’t Know

The case for CSDs rests primarily on the operational experience of a handful of cities, most of them mid-sized, most of them in politically progressive environments. Almost all the evidence is reported by the programs themselves rather than independently verified. No independent randomized evaluation has tested the CSD as an institutional form, as distinct from the individual programs it houses. Some of the individual programs housed within CSDs or similar structures have stronger independent evidence at the program level, but whether the department structure itself produces better outcomes than the same programs housed in separate agencies is an open question.

The cities that have built CSDs are not representative of all American jurisdictions. Albuquerque, Durham, Portland, Seattle, and Cambridge are all mid-to-large cities with political environments favorable to the concept. Whether the model translates to rural counties, conservative suburbs, or cities with different political dynamics is unknown. The evidence base comes from contexts that may not resemble every community considering this approach.

---

The Bottom Line

Community Safety Departments exist because 911 was built around three response options, and American cities produce calls each year that do not fit any of them.¹ Individual programs address pieces of the problem. A department consolidates them under one institutional roof. Polling shows 62% of voters prefer that unified model.¹⁶ The operational data from a small number of cities shows reduced crime, reduced arrests, and police workload relief.⁵ ⁶ ⁷ ¹¹ Independent evidence on the department model specifically, as distinct from its component programs, does not yet exist.

---

Source Appendix

1. University of Chicago Health Lab / Transform911, “Blueprint for Change, Chapter 7: Ensuring the Right Response at the Right Time.” “More than 75% of 911 calls for service dispatched to the police are not related to public safety threats that obviously merit a law enforcement response.” https://www.transform911.org/blueprint/chapter-7-response/

2. CSG Justice Center, “Atlanta, GA — Expanding First Response Program Highlights,” updated December 2024. Applied Research Services, Inc. analysis: “identified 600,000, or approximately 18 percent, of all incoming 911 calls as suitable for diversion.” https://csgjusticecenter.org/publications/expanding-first-response/program-highlights/atlanta-ga/

3. Sacramento County Sheriff Jim Cooper, press conference, February 4, 2025. Reported in Folsom Times, CBS Sacramento, Sacramento Observer, and KCRA. “Being mentally ill is not a crime and we can’t be the answer.” https://www.cbsnews.com/sacramento/news/sacramento-sheriff-explains-new-policy-to-turn-away-certain-calls-for-help/

4. Policing Project at NYU School of Law, “Transforming Denver’s First Response Model,” 2023. Qualitative research on dispatchers and officers. Jessica Gillooly (Suffolk University) lead researcher on dispatcher decision-making. https://www.policingproject.org/transforming-denvers-first-response-model-report

5. City of Durham / ICMA, “City of Durham Wins ICMA 2025 Community Health & Safety Award.” 10,000+ officer hours saved, 32,000+ calls. Durham: https://www.durhamnc.gov/m/newsflash/home/detail/3977. ICMA: https://icma.org/page/2025-community-health-safety-award-city-durham. Earlier 5,500-hour figure from Assembly NC profile (see source 15).

6. KOSU, Sierra Pfeifer, “Oklahoma City plans to expand mental health crisis response with state opioid settlement money,” February 6, 2026. 57% reduction in mental health dispatches, program expansion. https://www.kosu.org/health/2026-02-06/oklahoma-city-plans-to-expand-mental-health-crisis-response-with-state-opioid-settlement-money

7. CSG Justice Center, “Albuquerque, NM — Expanding First Response Program Highlights,” April 2025. Police Chief Medina and Commander Barnard context. https://csgjusticecenter.org/publications/expanding-first-response/program-highlights/albuquerque-nm/

8. Austin Police Association President Michael Bullock, testimony before Austin City Council, January 30, 2025. Reported in Community Impact, Ben Thompson, February 11, 2025. https://communityimpact.com/austin/south-central-austin/government/2025/02/11/austins-responses-to-mental-health-calls-under-review-with-aim-to-reduce-police-involvement/

9. Minneapolis Deputy Chief Erick Fors, testimony at Minneapolis Public Safety Committee meeting, April 20, 2022. Reported in Southwest Connector, Jan Willms, June 21, 2022. https://swconnector.com/stories/alternative-to-police,6234

10. Results for America, “Non-police emergency response: St. Petersburg, FL” case study. 11,000+ contacts, 97% diversion rate, Megan McGee quotes, officer quotes. https://catalog.results4america.org/case-studies/non-police-emergency-response-st-petersburg-fl. See also SPPD CALL program page: https://police.stpete.org/call/index.html. Gabber Newspaper coverage: https://thegabber.com/st-pete-may-send-social-workers-instead-of-police-to-more-calls/

11. NACo, “North Carolina county decreases 911 non-emergency calls by connecting residents to services they need,” November 2024. 440 calls → 4 in 30 days (NACo count); $352,000 staff cost, $4.5M projected annual savings, Justin Hargett EMS Manager. https://www.naco.org/news/north-carolina-county-decreases-911-non-emergency-calls-connecting-residents-services-they. WFMY TV reports 344→4 figure: https://www.wfmynews2.com/article/news/local/guilford-countys-adult-care-team-non-emergency-ems-calls/83-e01ad2e6-3ad7-4354-b521-6f0ffb1850cb

12. CSG Justice Center, “Durham, NC — Expanding First Response Program Highlights,” updated December 2024. Care Navigation division, 48-hour follow-up, four-division structure. https://csgjusticecenter.org/publications/expanding-first-response/program-highlights/durham-nc/

13. Keith L. Alexander, Steven Rich, and Hannah Thacker, “The hidden costs of police misconduct,” The Washington Post, March 9, 2022. “Nearly 40,000 payments at 25 of the nation’s largest police and sheriff’s departments within the past decade, documenting more than $3.2 billion spent to settle claims.” https://www.washingtonpost.com/investigations/interactive/2022/police-misconduct-repeated-settlements/

14. City of Albuquerque, “Albuquerque Community Safety Department Marks Four Years of Impact and Innovation,” press release, September 2025. Multi-division operations under single department. https://www.cabq.gov/acs/news/albuquerque-community-safety-department-marks-four-years-of-impact-and-innovation

15. Jeff Billman, “A New Model for Public Safety in Durham,” The Assembly NC, June 26, 2024. Ryan Smith “pieces of a larger idea” quote, four-division structure, 5,500 officer hours (pre-expansion figure). https://www.theassemblync.com/politics/criminal-justice/durhams-new-model-for-public-safety/

16. Safer Cities, national survey of 2,400 registered voters. Figures include: 62% prefer unified department, 84% agree on unmet needs from fear of police, 81% agree on reducing injury/death likelihood.

17. Harris County Commissioner Rodney Ellis. Reported in FOX 26 Houston, “Harris County Mental Health Program faces uncertain future amid financial concerns,” May 23, 2024. https://www.fox26houston.com/news/harris-county-mental-health-program-faces-uncertain-future-amid-financial-concerns