How Is This Different?

Most American cities already have some version of non-police crisis response. Denver sends civilian crisis teams to mental health calls.¹ Portland dispatches civilian responders to behavioral health emergencies.² Harris County deploys crisis specialists across unincorporated areas.³ San Diego County fields 44 civilian crisis units across the region.¹⁶ The question a council member considering a Community Safety Department must answer is not whether to offer these services, but whether the institutional structure around them matters.

The distinctions from what already exists are structural.

CSD vs. Standalone Programs in Existing Departments

The distinction starts with how programs are housed. A city can run a mobile crisis team inside the public health department, a violence intervention initiative out of the mayor’s office, a clean team through a nonprofit contract, and a homeless outreach program through the fire department. Many cities do exactly this. The programs may work well individually. The question is what happens between them.

Denver’s crisis response program lives in the Department of Public Health and Environment, operated by a nonprofit in partnership with a public hospital.¹ It responds to behavioral health calls. But it does not house violence intervention, homeless outreach, or needle cleanup under the same chain of command. If a responder encounters a person whose primary need is chronic homelessness rather than acute psychiatric crisis, the referral crosses an agency boundary.

Harris County’s crisis response program lives in county public health.³ It was designed for a county government structure that does not easily accommodate new standalone departments. It works within that constraint but operates as one program, not a department with multiple divisions.

Albuquerque chose differently. ACS is a cabinet-level department with civilian crisis response, violence intervention, street outreach, and needle pickup all under one roof.⁴ When a responder encounters a person with co-occurring needs spanning multiple categories, the response stays within a single institutional structure. The training academy produces responders who understand the full scope.⁵ The director can shift resources between divisions based on demand. The budget goes to one city council committee for oversight.

Durham’s CSD originally operated four functional areas that work together by design: field response for non-violent calls, a unit for calls involving weapons, a dispatch division resolving calls by phone from inside the 911 center, and follow-up case management.⁶ In July 2025, these were reorganized into two direct-service divisions, though the functional areas continue. Director Ryan Smith described them as “pieces of a larger idea.”⁷

The structural difference: standalone programs coordinate through interagency agreements, memoranda of understanding, and informal relationships. A CSD coordinates through a chain of command. The former depends on the quality of relationships between agency leaders who may change with every election cycle. The latter is built into the org chart.

Polling shows voters prefer the unified department model over separate programs.⁸

CSD vs. Co-Response Programs

Co-response programs pair a licensed clinician with a police officer on every call. The officer provides a security presence; the clinician provides mental health expertise.

A CSD deploys teams independently. No armed officer accompanies the responder. The team is the primary and often only response to its designated call types. These are structurally opposite approaches to the same problem.

The operational differences flow directly from this structural split. In a co-response model, a police officer is still tied up on every call. The workload relief that police departments value does not materialize because an officer is consumed by the same call whether alone or paired with a clinician. In a CSD model, every call handled by a community safety responder is a call the police department does not need to staff at all.

Independent research supports the civilian-only response model that CSDs use. A 2025 Wayne State University study across five Michigan communities found that civilian-only teams outperformed co-response models on diversion and arrest reduction.⁹ Co-response programs produced arrest outcomes similar to police-only response and showed higher hospitalization rates. This evidence applies to the response model, not the governance structure.

St. Petersburg’s police chief explicitly rejected the co-response model when designing the city’s crisis program, saying there were enough non-violent, non-criminal calls that could be “completely diverted” from police.¹⁰ His reasoning was direct: if the goal is to free officers for calls requiring law enforcement skills, pairing an officer with every clinician defeats the purpose.

That said, co-response occupies a legitimate place in the landscape. Durham’s CSD includes a co-response segment specifically for calls involving weapons or threats of violence, where a clinical presence alongside an officer adds capability without removing the security the situation requires.⁶ The distinction is that Durham treats co-response as one tool within a larger civilian-first department, not as the department’s default model.

Several cities have started with co-response and evolved toward civilian-only as their safety records built confidence. Denver explored co-response before launching its civilian-only crisis program.¹ Eugene operated its alternative response within the police department for 33 years before shifting to the fire department (the program later collapsed in April 2025).

CSD vs. Programs Housed in Police Departments

Some alternative response programs live inside police departments. St. Petersburg’s crisis program is managed through the police department’s Special Projects office.¹⁰ The program works, and police leadership has been supportive. But the institutional dynamics are different from a standalone CSD.

When a program lives inside a police department, its budget competes with other police priorities. Its staffing decisions go through police HR. Its culture is shaped by the department it sits inside. If the police chief changes and the new chief has different priorities, the program’s future depends on one person’s discretion rather than an institutional mandate.

When a program lives in a standalone CSD, it has its own budget line, its own hiring authority, and its own director who reports to the mayor or city manager. It cannot be quietly defunded by being absorbed into a larger departmental budget. It cannot be restructured by a police chief who prefers a different approach. The institutional independence is the point.

Portland made this structural choice explicit in June 2025 when the City Council passed a resolution “formally establishing Portland’s crisis response department as an equal branch of the city’s public safety system” with a directive to explore granting staff full first-responder designation and associated employment benefits.¹¹ The resolution moved the department from being a program within another department toward co-equal institutional standing.

The tradeoff: programs housed in police departments benefit from immediate dispatch integration, because police already control 911 routing. Programs in standalone departments must build that integration from scratch or negotiate it through interagency agreements. Durham invested in this from day one, embedding clinicians in the 911 center and building an automated call identification system.⁶ The investment was substantial, but the result was dispatch infrastructure the department controls rather than borrows.

CSD vs. Fire Department-Housed Programs

Fire departments have become increasingly common institutional homes for alternative response. Salt Lake City, Tulsa, Oklahoma City, Lexington, and Philadelphia house behavioral health response within their fire departments or fire-department partnerships.¹² Eugene did the same until its program collapsed in 2025. The logic is appealing: fire departments already run 24/7 shifts, already have dispatch integration, already carry a medical (rather than enforcement) identity, and are broadly trusted by the public.

Eugene’s crisis response program, the oldest in the country (launched in 1989), operated under the police department for 33 years before moving to the fire department. The Eugene program collapsed in April 2025 when its nonprofit operator suffered financial failure; CAHOOTS service continues in neighboring Springfield.

The distinction from a CSD is institutional scope and identity. A fire department that houses a behavioral health team is adding a function to an existing department. A CSD is purpose-built around that function as its core mission. The fire chief’s primary responsibility is fire suppression and emergency medical services; behavioral health response is one priority among many. A CSD director’s primary responsibility is the full spectrum of non-police emergency response.

Lexington, Kentucky, illustrates the fire-department model’s strengths and boundaries. The Lexington Fire Department houses three specialized teams under community paramedicine: a Quick Response Team for post-overdose engagement, a Mobile Integrated Health team for chronic conditions, and a Crisis Response Team for mental health calls.¹² Lt. Alexander Jann described their approach: “We’re going to figure out what you need to solve the problem. We have connections to all these community partners, so we’re able to call up somebody and be like, ‘Hey, this person needs that.’ We’re that middleman to get it done.”¹² Mayor Linda Gorton said the program has “improved patient outcomes, and reduced the burden on frontline emergency service providers like police, fire and EMS.”¹²

The Lexington model houses multiple specialized teams under one roof, which is structurally similar to a CSD. But it remains a division within the fire department, not a co-equal branch of city government. The practical implications: if fire department leadership changes priorities, if the fire budget faces cuts, or if the fire union resists scope expansion, the behavioral health programs compete for institutional attention against the department’s primary mission.

CSD vs. “Defund the Police”

This is a political distinction, not a structural one, but it shapes the public conversation around CSDs.

“Defund the police” implies reducing police budgets and reallocating that money to social services. CSDs are additive: they create a new department alongside police, not instead of police. In Albuquerque, the police budget is more than twenty times larger than the CSD budget.¹³ No established CSD has been funded by cutting a police budget.

Polling shows that framing affects support levels. When asked about “community safety departments” as described with operational specifics, 82% of voters express support.⁸ When the same concept is framed through the “defund” lens, support drops sharply.⁸

Albuquerque police leadership has offered the most direct operational rebuttal to the “defund” framing: commanders publicly credit the CSD with enabling crime reduction by freeing officers to focus on felony cases, and Commander Jeff Barnard has called for the department to receive additional funding.¹⁴ Durham Police Chief Patrice Andrews (retiring May 2026), despite initial skepticism from officers who “chafed over the political fights that accompanied the program’s creation,” now reports officers requesting CSD teams on the radio.¹⁵

89% of voters agree CSDs “allow police departments to focus on solving serious crimes.”⁸

CSD vs. Traditional Social Services

Social service agencies provide ongoing case management, treatment referrals, housing assistance, and benefits navigation from offices during business hours. They serve an essential function. They are not first responders.

CSDs deploy field teams through 911 dispatch to active situations. The responders go to the person, wherever the person is, at the time of the call. They operate on emergency response timelines, not appointment schedules.

The distinction matters because the people who generate repeated 911 calls are often the same people who have been referred to social services multiple times and have not engaged. CSD responders go to them in the moment and work to bridge the gap between crisis and sustained care.

Durham’s model makes this explicit. The care navigation segment follows up within 48 hours of a crisis encounter and provides ongoing case management: housing search, food insecurity assistance, eviction support, connections to city services.⁶

What We Don’t Know

The central unanswered question about CSDs as an institutional form is whether the department structure itself produces better outcomes than the same programs housed in separate agencies. No study has compared CSD-housed programs against identical programs in public health departments, fire departments, or police departments, controlling for other variables. The theoretical advantages of unified command, shared training, and institutional identity are plausible but unproven at the institutional level. Cities choosing between a standalone CSD and programs housed in existing departments are making a governance bet, not following settled evidence.

---

The Bottom Line

A Community Safety Department is different from standalone programs, co-response models, police-housed teams, fire-housed teams, and traditional social services in one structural way: it is a purpose-built city department, co-equal with police and fire, whose institutional mission is non-police emergency response. Fire department-housed programs share the multi-team structure but sit within a department whose primary mission is fire suppression and EMS. Standalone programs in public health may work well but lack unified command and institutional permanence. Independent research supports the civilian-only response model that CSDs use.⁹ No study has tested whether the CSD governance structure produces better outcomes than the same programs housed elsewhere.

---

Source Appendix

1. Thomas S. Dee and Jaymes Pyne, “A community response approach to mental health and substance abuse crises reduced crime,” Science Advances, Vol. 8, Issue 23 (June 8, 2022). Stanford evaluation of Denver STAR program; program housed in Department of Public Health and Environment. https://pmc.ncbi.nlm.nih.gov/articles/PMC9176742/. See also Urban Institute interim evaluation: https://www.urban.org/research/publication/evaluating-alternative-crisis-response-denvers-support-team-assisted-response

2. Portland State University Homelessness Research & Action Collaborative (Greg Townley, lead evaluator), Portland Street Response Year Two Evaluation, July 2023. https://www.pdx.edu/homelessness/sites/homelessness.web.wdt.pdx.edu/files/2023-07/HRAC%20Portland%20Street%20Response%20Year%20Two%20Evaluation%20Report_FINAL%20FOR%20WEBSITE.pdf

3. FOX 26 Houston, “Harris County Mental Health Program faces uncertain future amid financial concerns,” May 23, 2024. HART under Harris County Public Health, 11,153 calls. https://www.fox26houston.com/news/harris-county-mental-health-program-faces-uncertain-future-amid-financial-concerns

4. City of Albuquerque, “Albuquerque Community Safety Department Marks Four Years of Impact and Innovation,” press release, September 2025. Multi-division structure, 120,000+ calls, 85% diverted. https://www.cabq.gov/acs/news/albuquerque-community-safety-department-marks-four-years-of-impact-and-innovation

5. Central New Mexico Community College, “CNM and City of Albuquerque Launch Annual ACS Academy to Train Alternative First Responders,” October 2025. 140-hour curriculum, accreditation pathway. https://www.cnm.edu/news/cnm-and-city-of-albuquerque-launch-annual-acs-academy-to-train-alternative-first-responders

6. CSG Justice Center, “Durham, NC — Expanding First Response Program Highlights,” updated December 2024. Four-division structure, 911 integration, embedded clinicians, automated CAD, care navigation. https://csgjusticecenter.org/publications/expanding-first-response/program-highlights/durham-nc/

7. Jeff Billman, “A New Model for Public Safety in Durham,” The Assembly NC, June 26, 2024. Ryan Smith “pieces of a larger idea” quote. https://www.theassemblync.com/politics/criminal-justice/durhams-new-model-for-public-safety/

8. Safer Cities, national survey of 2,400 registered voters. Figures include: 62% prefer unified department model, 82% support with operational description, 89% agree on police focus benefit.

9. Leonard Swanson, Catherine Zettner, Amy Watson, Melanie Hinojosa, Juliette Roddy, and Sheryl Kubiak, “Eleven-Month Arrest Outcomes Among Three Crisis Response Models in Michigan,” Psychiatric Research and Clinical Practice, Vol. 7, Issue 3, pp. 182-190 (May 2, 2025). Mobile crisis IRR 0.548 (45.2% less likely arrested); co-response not statistically significant vs. law enforcement only. https://pmc.ncbi.nlm.nih.gov/articles/PMC12418740/

10. Results for America, “Non-police emergency response: St. Petersburg, FL” case study. Chief Tony Holloway and Megan McGee quotes, civilian-only model rationale, 11,000+ contacts, 97% diversion rate. https://catalog.results4america.org/case-studies/non-police-emergency-response-st-petersburg-fl. See also WFLA: https://www.wfla.com/news/pinellas-county/st-pete-police-creating-new-group-to-respond-to-non-violent-911-calls-instead-of-officers/

11. Portland City Council Resolution No. 37709, “Support and expand Portland Street Response as a co-equal branch of the first responder system,” adopted June 25, 2025. Passed 10-2. https://www.portland.gov/council/documents/resolution/adopted/37709. KGW8 coverage: https://www.kgw.com/article/news/politics/city-council-passes-resolution-portland-street-response/283-ec59262b-7171-4d84-8f23-6c06e1903bb7

12. City of Lexington, “Mayor cuts ribbon to open Community Paramedicine office,” August 5, 2025. Lt. Alex Jann and Mayor Linda Gorton quotes, three-team structure. https://www.lexingtonky.gov/news/mayor-cuts-ribbon-open-community-paramedicine-office. See also Spectrum News 1, Sabriel Metcalf, August 13, 2025: https://spectrumnews1.com/ky/louisville/news/2025/08/12/lexington-s-opens-community-paramedicine-center

13. Murat Oztaskin, “Sending Help Instead of the Police in Albuquerque,” The New Yorker, February 4, 2023. Police budget “more than twenty times larger.” https://www.newyorker.com/news/dispatch/sending-help-instead-of-the-police-in-albuquerque

14. 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/

15. Tradeoffs / The Marshall Project, “How Durham Got Police Onboard with Unarmed Crisis Response,” May 2, 2025. Andrews quotes on initial skepticism, officer adoption, radio requests. https://tradeoffs.org/2025/05/02/how-durham-north-carolina-got-police-onboard-with-unarmed-crisis-response/

16. San Diego County News Center, “Mobile Crisis Response Program Celebrates Four Years,” January 22, 2025. 44 teams. https://www.countynewscenter.com/mobile-crisis-response-program-celebrates-four-years/