Beyond Jails: Community-Based Strategies for Public Safety

Watch these ideas in action:

Chicago, IL | Tucson, AZ | Denver, CO | Baltimore, MD

For decades, the United States has responded to social issues like mental health and substance use crises, chronic homelessness, and ongoing cycles of interpersonal violence with jail. This has disrupted the lives of millions of people—disproportionately harming Black and Indigenous people—without improving public safety. There’s a better way. Communities can instead invest in agencies and organizations that address these issues outside the criminal legal system. The proven solutions highlighted in this multimedia report look beyond jails to promote safe and thriving communities.

Who’s in Jail and Why

More than 3,000 jail facilities operate in the United States.[]Zhen Zeng and Todd Minton, Jail Inmates in 2019 (Washington, DC, Bureau of Justice Statistics [BJS], 2021), 11, Before the COVID-19 pandemic, those jails processed about 10 million bookings annually.[]Zeng and Minton, Jail Inmates in 2019, 2021, 1, Some people stayed for hours and others for months.[]Mark A. Cunniff, Jail Crowding: Understanding Jail Population Dynamics (Washington, DC: U.S. Department of Justice [DOJ], 2002), 7, Overall, the number of people in jail has grown exponentially over the past 40 years—from about 220,000 in 1983 to more than 750,000 in 2019.[]BJS, The 1983 Jail Census (Washington, DC: BJS, 1984), 1,; and Jacob Kang-Brown, Chase Montagnet, and Jasmine Heiss, People in Jail and Prison in 2020 (New York: Vera Institute of Justice, 2021), 1,

In response to the COVID-19 pandemic, some jurisdictions took emergency actions to prevent the virus’s spread among incarcerated people and jail staff, which cut jail populations by an estimated 24 percent during the first half of 2020. However, these changes proved temporary; by June 2020, national jail populations were already rising. By the end of 2020, the population had rebounded by more than 50,000 people.[]Kang-Brown, Montagnet, and Heiss, People in Jail and Prison in 2020, 2021, 1-2.

On any given day in 2019, jails in the United States held more than 750,000 people.

Close to 70 percent of all people held in local jails have been charged with violations of drug, property, or public order laws; less than one-third have been charged with offenses that are considered violent.[]Wendy Sawyer and Peter Wagner, Mass Incarceration: The Whole Pie 2020 (Northampton, MA: Prison Policy Initiative, 2020), Unlike in prisons—where incarcerated people have been convicted of a crime—two-thirds of the people in local jails have not been found guilty of their current charges but remain incarcerated pretrial, often because they’re unable to pay even small bail amounts.[]Of the 734,400 people in jail on a given day, 480,700 are unconvicted. Zeng and Minton, Jail Inmates in 2019, 2021, 5. For a discussion of bail, see Léon Digard and Elizabeth Swavola, Justice Denied: The Harmful and Lasting Effects of Pretrial Detention (New York: Vera Institute of Justice, 2019), 2,

Many incarcerated people also experience added challenges like homelessness or behavioral health issues. Forty-four percent of people in jail report having at least one mental health condition.[]Jennifer Bronson and Marcus Berzofsky, Indicators of Mental Health Problems Reported by Prisoners and Jail Inmates, 2011-2012 (Washington, DC: BJS, 2017), 3, And the rate of people with substance use disorders is six times as high in jail as in the community.[]Ram Subramanian, Ruth Delaney, Stephen Roberts et al., Incarceration’s Front Door: The Misuse of Jails in America (New York: Vera Institute of Justice, 2015), 13, People in jails have also experienced homelessness at a rate from 7.5 to 11.3 times that of the broader population.[]For these figures, homelessness is defined as being unhoused at any point during the previous year or, for people in jail, in the one year prior to the beginning of their current incarceration. See generally Greg A. Greenberg and Robert A. Rosenheck, “Jail Incarceration, Homelessness, and Mental Health: A National Study,” Psychiatric Services 59, no. 2 (2008), 170-177; and Applied Survey Research, 2013 San Francisco Homeless Point-in-Time Count and Survey (Sacramento, CA: Applied Survey Research, 2014), These facts are not accidental. They’re the result of policy decisions to use enforcement and incarceration instead of treatment and services.[]Tony Robinson, “No Right to Rest: Police Enforcement Patterns and Quality of Life Consequences of the Criminalization of Homelessness,” Urban Affairs Review 55, no. 1 (2019), 41-73, 43-44,

The U.S. jail population also includes disproportionately high numbers of Black and Indigenous people—they are incarcerated at rates triple and double that of white people, respectively.[]BJS data shows that “Hispanic” people are incarcerated at a rate of 176 per 100,000 compared to 184 per 100,000 residents for white people. Zeng and Minton, Jail Inmates in 2019, 2021, 4. However, the criminal legal system’s categorization of Latinx people in jail data is fraught, and an unknown number of Latinx people are recorded as “white,” if ethnicity is recorded at all. Ram Subramanian, Kristine Riley, and Chris Mai, Divided Justice: Trends in Black and White Jail Incarceration, 1990-2003 (New York: Vera Institute of Justice, 2018), 26, These disparities result from policymaking that has consistently, and at times intentionally, targeted Black and Indigenous people for punishment.[]For more on racial disparities in the criminal legal system, see Elizabeth Hinton, LeShae Henderson, and Cindy Reed, An Unjust Burden: The Disparate Treatment of Black Americans in the Criminal Justice System (New York: Vera Institute of Justice, 2018),; and Ruth Delaney, Ram Subramanian, Alison Shames, and Nicholas Turner, Reimagining Prison (New York: Vera Institute of Justice, 2018), And women are being incarcerated at accelerating rates. Since 2008, the number of women in jails has increased by 11.4 percent, despite an overall jail population decrease of 6 percent during the same time period.[]This compares to a 9 percent decrease for men. Zeng and Minton, Jail Inmates in 2019, 2021, 5.

Public Safety and the Costs of Jail

Put simply, jails don’t make us safer. Research shows increased incarceration has historically contributed less to falling crime rates than broader social and economic factors have.[]Oliver Roeder, Lauren-Brooke Eisen, and Julia Bowling, What Caused the Crime Decline? (New York: Brennan Center for Justice, 2015), 4,; and Pew Center on the States, State of Recidivism: The Revolving Door of America’s Prisons (Washington, DC: Pew Charitable Trusts, 2011), 5-6, 12-13, And during the three-decade drop in crime since the 1990s, several large states decreased their incarcerated populations while experiencing declines in crime.[]These states are California, Michigan, New Jersey, New York, and Texas. Roeder, Eisen, and Bowling, What Caused the Crime Decline?, 2015, 3, 4, 8, 15, and 28. Even as shootings and homicides increased in 2020, evidence suggests that decreased jail populations were not to blame; instead, some experts have pointed to destabilized access to public resources during COVID-19 lockdowns, changes in law enforcement behavior, and an unsustainable dependence on police and punishment as key contributors to the spike in violence.[]Experts are still working to understand the complex factors driving the increase in homicides during 2020. See for example Derek Thompson, “Why America’s Great Crime Decline Is Over,” Atlantic, March 24, 2021, Early evidence appears to rule out one factor: reduced jail populations. For more, see CUNY Institute for State and Local Governance, Jail Decarceration and Public Safety: Preliminary Findings from the Safety and Justice Challenge (New York: CUNY Institute for State and Local Governance, 2021),; and the JFA Institute, The Impact of COVID-19 on Crime, Arrests, and Jail Populations: An Expansion on the Preliminary Assessment (Denver, CO: JFA Institute, 2021), 15, And jail won’t fix it either: other research shows increased use of incarceration can actually increase crime, especially in communities with already high incarceration rates.[]Todd R. Clear, “The Effects of High Imprisonment Rates on Communities,” Crime and Justice 37 (2008), 97-132; Francis T. Cullen, Cheryl Lero Jonson, and Daniel S. Nagin, “Prisons Do Not Reduce Recidivism: The High Cost of Ignoring Science,” Prison Journal 91, no. 3 supplement (2011), 52-58,; and Danielle Sered, Until We Reckon: Violence, Mass Incarceration, and the Road to Repair (New York: The New Press, 2021), 67.

What’s more, incarceration fails to address the issues marginalized communities have identified as important for their safety.[]Katie Buitrago, Amy Rynell, and Sandra Tuttle, Cycle of Risk: The Intersection of Poverty, Violence, and Trauma in Illinois (Chicago: Heartland Alliance, 2017), Some jurisdictions have implemented “reforms” to reduce jail use, but many approaches ultimately increased the role of criminal legal system agencies and used the threat of incarceration to coerce compliance with mandated programming, adding to the approaches' harms. Better paths exist: agencies can contribute space and money to build and sustain community-based services people can access without arrest or incarceration.

This approach is beneficial even when interpersonal violence occurs.

A 2015 Brennan Center for Justice analysis showed that increased incarceration had little to no effect on violent crime rates from 1990 to 2013.[]Roeder, Eisen, and Bowling, What Caused the Crime Decline?, 2015, 3, 8, 15, and 28. And other studies show that spending time behind bars can have “criminogenic” effects, increasing the likelihood that someone will be reincarcerated.[]See for example Cullen, Jonson, and Nagin, “Prisons Do Not Reduce Recidivism,” 2011, 52-58; and Sered, Until We Reckon, 2021, 67. Responding to individual occurrences of interpersonal violence with jail instead of addressing the underlying causes of cyclical violence not only fails to produce safety, but also perpetuates harm to people and communities.

The Harms of Jail Incarceration

Group Created with Sketch. Read More

Nationally, there’s been an amplified call for an ecosystem of services to help people manage conflict, address health issues, and promote socioeconomic stability and public safety without relying on the criminal legal system. This vision prioritizes prevention, accountability, and treatment rather than incarceration. Blueprints for public safety approaches that do not center incarceration—and a variety of strategies to fund them—already exist.[]See for example Leah Sakala, Samantha Harvell, and Chelsea Thompson, Public Investment in Community-Driven Safety Initiatives: Landscape Study and Key Considerations (Washington, DC: Urban Institute, 2018), See also Buitrago, Rynell, and Tuttle, Cycle of Risk, 2017.

Governments can use their authority to dedicate resources to these strategies, including community-based behavioral health crisis services, permanent supportive housing programs, and violence prevention and de-escalation services. Genuine partnership with nonprofit organizations and advocacy groups must be at the center of efforts to create a network of supports that function effectively, equitably, and without funneling people into the criminal legal system.

Responding to Behavioral Health Crises without Incarceration

Because law enforcement is always available, and behavioral health services have fewer resources and limited capacity, police are often the first responders when people experience behavioral health crises—even if they are not the best prepared to render aid.[]Jennifer D. Wood and Amy C. Watson, “Improving Police Interventions During Mental Health-Related Encounters: Past, Present and Future,” Policing & Society 27 (2017), 289-299, Some jurisdictions have tried to address the disconnect between residents’ needs and the services available to them by investing in police-led diversion, additional law enforcement training, and programs that pair officers with behavioral health specialists in the field. But these approaches still center police intervention. Jurisdictions should move beyond these limited options to focus attention and resources on strengthening the broader mental health ecosystem. These investments are essential for successful solutions that operate outside of the criminal legal system because they build the capacity of behavioral health service providers and local health-focused organizations.[]National Alliance on Mental Illness, Divert to What? Community Services That Enhance Diversion (Washington, DC: National Alliance on Mental Illness, 2020),; and SAMHSA, Crisis Services: Meeting Needs, Saving Lives (Rockville, MD: SAMHSA, 2020), 8-13,

The investments detailed in this section can help jurisdictions reduce the use of jail incarceration by meeting people’s prevention and treatment needs without participation in any criminal legal process.

Crisis Call Centers

A crisis call center is a 24-hour clinically staffed, central location designed to provide immediate phone support to people who may be experiencing a behavioral health emergency—similar to the 911 system.[]SAMHSA, National Guidelines for Behavioral Health Crisis Care: Best Practice Toolkit (Washington, DC: U.S. Department of Health and Human Services [HHS], 2020), 14-15, These centers conduct behavioral health assessments and help callers problem-solve, develop coping strategies, and connect to other support services. Some crisis call services collaborate with local police departments to divert 911 calls that a behavioral health specialist can address. With the proper technological support, regional call hubs can also enhance coordination by using real-time information to track the availability of mobile responders, monitor the capacity of treatment providers, and verify when a person has been connected to services. These centers can help reduce the use of police response to behavioral health crises, which decreases the likelihood of arrest; physical harm to the person in crisis, other residents, or officers; and inadequate connection to care.[]Wood & Watson, “Improving Police Interventions,” 2017; and Eric Westervelt, “Mental Health And Police Violence: How Crisis Intervention Teams Are Failing,” National Public Radio, September 18, 2020, In July 2022, the Federal Communications Commission plans to roll out 988 as a three-digit dialing code to reach the National Suicide Prevention Lifeline, which provides free emotional support 24/7 to callers experiencing suicidal crisis or emotional distress.

Warmlines are another telephone-based service offering behavioral health crisis assistance without traditional emergency responders. They provide people a confidential space to speak with a trained responder about their needs and symptoms. Warmlines differ from 24/7 crisis hotlines in that they are not typically used for emergencies and are generally staffed by “peers,” or people who have direct experience with behavioral health issues.[]Truven Health Analytics, Crisis Services: Effectiveness, Cost-Effectiveness, and Funding Strategies (Rockville, MD: SAMHSA, 2014), 12, Warmlines can help de-escalate situations that may have otherwise resulted in an emergency department visit or 911 call.

Mobile Crisis Response Teams

Mobile crisis response teams are staffed by nurses and behavioral health specialists trained in crisis response, including at least one clinician who can provide assessments, de-escalation, and connections to other services as needed (including transportation). These teams may also include trained peers. Mobile crisis response teams may request police backup when they deem it necessary but are designed to respond without law enforcement.[]SAMHSA, National Guidelines for Behavioral Health Crisis Care, 2020, 18. These teams also coordinate with local emergency medical services (EMS) and can operate as either an alternative to, or an extension of, EMS.


One of the best-known mobile crisis response programs is Crisis Assistance Helping Out On The Streets (CAHOOTS) in Eugene, Oregon. The program uses two-person teams pairing a medic and a behavioral health crisis worker to provide immediate stabilization, referrals, and/or transportation to further treatment resources.

Learn more about the CAHOOTS mobile crisis response model.


In June 2020, Denver, Colorado, launched Support Team Assisted Response (STAR), a program that dispatches a mental health clinician and a paramedic instead of armed officers to respond to behavioral health crises or low-level incidents related to poverty or homelessness, such as trespassing. STAR responders can connect community members to resources like food assistance, shelter, and ongoing mental health care. Dispatchers send STAR through 911 when appropriate calls for service come in, or STAR teams can be requested through Denver Police Department’s non-emergency line. In the program’s first six months, STAR teams responded to 748 calls and none resulted in police involvement or arrest.


Portland, Oregon, has also instituted specialized mobile crisis response to reduce police interaction. Portland Street Response (PSR) started in 2021 as part of a $500,000 pilot program to reduce police contact with people who are experiencing homelessness and/or behavioral health issues. When a 911 call about street homelessness or public disorder comes in, PSR dispatches specially trained medics alongside peer support specialists who have direct experience with being unhoused. In addition to providing care for non-life-threatening medical issues and connecting people to services, the team may provide transportation to shelters, clinics, or another destination the person being helped selects.

Crisis Stabilization and Receiving Services

Jails have become some of the largest institutions providing psychiatric care in the United States.[]Ailsa Chang, “Insane': America's 3 Largest Psychiatric Facilities Are Jails,” National Public Radio, April 30, 2018,; and Matt Ford, “America’s Largest Mental Hospital is a Jail,” Atlantic, June 8, 2015, Corrections officials, behavioral health professionals, advocates, and others have called for more resources to enhance behavioral health treatment within jails—and for ways to displace jail as a behavioral health provider for people in crisis.

One approach is to provide residents, mobile crisis teams, and other first responders with a rapidly accessible location in the community to use when a person is experiencing a behavioral health crisis that cannot be handled onsite. These crisis response centers also provide treatment space for residents and their loved ones to proactively access without relying on first responders.

Crisis receiving and stabilization centers offer a therapeutic, non-hospital environment for temporary observation and rapid service delivery to handle acute behavioral health crises.[]SAMHSA, National Guidelines for Behavioral Health Crisis Care, 2020, 22. They’re designed to accept everyone who accesses the center 24/7, whether they walk in, are referred by someone in the community, or are brought by a first responder. Once a person arrives at a receiving and stabilization facility—sometimes called a drop-off center—they are assessed, stabilized, and connected to the appropriate levels of care, all within 24 hours. When police respond to someone experiencing a behavioral health crisis, a drop-off center can serve as a quick and suitable destination that is neither jail nor the emergency department.[]SAMHSA, Executive Order Safe Policing for Safe Communities: Addressing Mental Health, Homelessness, and Addiction Report (Washington, DC: HHS, 2020), 9-10,

In addition to short-term stabilization and receiving, drop-off centers can facilitate connections to employment/vocational assistance, legal help, food and nutrition assistance, emergency housing, substance use treatment, and other services to foster people’s success in the community.[]Rafik Nader Wahbi, Sterling Johnson, and Leo Beletsky, From Crisis Response to Harm Prevention: The Role of Integrated Service Facilities (San Francisco: The Justice Collaborative, 2020), Although these services alone are not sufficient for managing behavioral health needs long-term, they can help build an infrastructure capable of reducing jail use by providing immediate diversion from criminal legal system contact and connections to long-term support. Crisis stabilization centers and other similar facilities have increased the use of less restrictive treatment options, reduced unnecessary hospitalizations, and shortened inpatient stays when psychiatric hospitalizations did occur.[]Elie Francis, William Marchand, Marlene Hart et al., “Utilization and Outcome in an Overnight Psychiatric Observation Program at a Veterans Affairs Medical Center,” Psychiatric Services 51, no. 1 (2000), 92-95,

Non-Jail Crisis Responses in Chicago and Pima County

In Chicago, the Westside Community Triage and Wellness Center provides urgent behavioral health care and serves as a hub to connect the neighborhood’s largely Black and Latinx residents to ongoing behavioral health services.


In Pima County, Arizona (Tucson), the Crisis Response Center offers 24/7 access to care resources for people who are experiencing behavioral health crises to avoid jail or emergency room settings.


What Practitioners Should Consider

  • Regularly review policies, practices, and eligibility criteria to ensure they do not systematically exclude people who may benefit from the services. Eligibility criteria limit the number of people reached by crisis call centers, mobile crisis response teams, and crisis receiving and stabilization services. It’s important to ensure widespread, equitable access across categories of race, class, gender, and ability.
  • Make first responders aware of community-based crisis response options. Non-jail solutions will not be used to their full potential if people in need, emergency personnel, and police are unaware of them. For example, Chicago’s Westside Community Triage and Wellness Center substantially increased its clientele following an in-depth training for the local police after an evaluation showed law enforcement personnel did not understand the benefits of referrals.
  • Implement programs in partnership with a diverse set of stakeholders, accounting for histories of racialized harm and prioritizing the perspectives of communities that have been most impacted by incarceration. Some prominent mobile behavioral health crisis responses (like CAHOOTS and PSR) originated in overwhelmingly white jurisdictions, and these models may not have the same outcomes in other cultural contexts.[]CAHOOTS was first implemented in Eugene, Oregon, which in 2019 was estimated to be 83.3 percent white. Similarly, the population of Portland, Oregon, the home of Portland Street Response, was 77.4 percent white. For more, see U.S. Census Bureau, “QuickFacts: Eugene, Oregon,”; and U.S. Census Bureau, “QuickFacts: Portland, Oregon,” For example, linking behavioral health service providers to 911 and law enforcement does not guarantee universal access to crisis response services because many people, especially in communities of color, are hesitant to call 911.[]Molly Harbarger, “Police Cuts Give Portland Alternative First Responder Program a Boost—But Can It Respond to the Moment?” The Oregonian, July 4, 2020, Centering racial equity and accounting for cultural differences are vital to success.
  • Explore and pursue multiple funding sources and sustainability models. Crisis stabilization and receiving centers need physical space, so funding can be a key challenge. The National League of Cities has highlighted that jurisdictions can support these programs through capital funds from municipal bonds, Community Development Block Grants, and in-kind support like using city-owned property.[]National League of Cities (NLC), Triage Centers as Alternatives to Jail for People in Behavioral Health Crises (Washington, DC: NLC, 2019), 6, Still, more partnerships may be necessary to ensure sustained funding. Crisis call centers, mobile crisis response teams, and crisis receiving and stabilization centers all rely on a broader treatment infrastructure and strong partnerships with other service providers.[]Sara Fleming, "Civilian Team May Respond to 911 Crisis Calls Instead of Denver Police," WestWord, November 8, 2019,; and Street Roots, Portland Street Response (Portland, OR: Street Roots, 2019),

Addressing Chronic Homelessness without Incarceration

Jail incarceration frequently worsens the health problems, employment barriers, strained familial relationships, and other issues chronically unhoused people face. And conventional housing options often exclude people with criminal legal system involvement.[]Prisoner Reentry Institute at John Jay College of Criminal Justice, Fortune Society, Corporation for Supportive Housing, and Supportive Housing Network of New York, A Place to Call Home: A Vision for Safe, Supportive and Affordable Housing for People with Justice System Involvement (New York: John Jay College Prisoner Reentry Institute, 2017), 5-7, Some approaches intended to help unhoused people, who often have multiple unmet needs, may require them to participate in programs to demonstrate their independence and address other underlying issues, like substance use, as a prerequisite for housing access.[]Josh Leopold and Mary K. Cunningham, “To End Homelessness Carson Should Continue Housing First Approach,” Urban Institute, January 18, 2017, However, these requirements often present barriers instead of supports. Research shows that prioritizing direct access to housing can make it easier for people to address economic and health-related needs that may drive their chronic homelessness.[]The Housing First movement treats everyone as “housing ready” and provides for permanent housing without preconditions. In many cases, people experience improvements in other areas of life once their housing needs are met. See United States Interagency Council on Homelessness, Housing First In Permanent Supportive Housing (Washington, DC: U.S. Department of Housing and Urban Development, 2014),

Permanent Supportive Housing

Permanent supportive housing (PSH) programs provide permanent affordable rental housing and access to tailored, voluntary services—without prerequisites or stringent conditions.[]Leopold and Cunningham, "To End Homelessness," 2017. Participants receive rental assistance and other supports that enable them to sign a standard lease with a local supportive housing provider.[]National Alliance to End Homelessness, Housing First (Washington, DC: National Alliance to End Homelessness, 2016), Once housed, people have access to ongoing support from a case manager, who can connect them to public benefits, treatment, and other wellness services. Because PSH provides long-term housing for people with extremely low incomes and high service needs, a combination that disproportionately affects communities of color, it’s a promising homelessness response strategy to advance racial equity.[]Oakland-Berkeley-Alameda County Continuum of Care, Centering Racial Equity in Homeless System Design (Oakland, CA: EveryOne Home, the City of Oakland, the City of Berkley, and Alameda County Health Care Services Agency, 2021), 26-27,

Additionally, PSH programs substantially reduce the number of days participants spend in jail compared to nonparticipants. They can also improve outcomes for people returning home from incarceration, who may have few options other than the streets, shelters, or unsuitable housing, which makes reincarceration more likely.[]Emily Peiffer, “Five Charts that Explain the Homelessness-Jail Cycle—and How to Break It,” Urban Institute, September 16, 2020, Some PSH programs are specifically tailored to reduce jail incarceration, using eligibility criteria to prioritize people who are frequently involved with criminal legal, shelter, and hospital systems.

For example, the Supportive Housing Social Impact Bond (SIB) initiative in Denver, Colorado, serves unhoused people who have been to jail at least eight times in the past three years. SIB participants spent an average of 19 days in jail per year compared to 77 days for similarly situated nonparticipants.[]Sarah Gillespie, Devlin Hanson, Mary Cunningham et al., Engaging the Most Vulnerable in Supportive Housing: Early Lessons from the Denver Supportive Housing Social Impact Bond Initiative (Washington, DC: Urban Institute, 2017), 20,

How Permanent Supportive Housing Keeps Denverites Safe

In Denver, the Social Impact Bond Initiative connects people experiencing homelessness to stable housing and then provides wraparound services in collaboration with community-based providers.


PSH enables jurisdictions to deliver a targeted, comprehensive response to chronic homelessness that’s more effective than incarceration. For example, the first statewide PSH study in Illinois found supportive housing was associated with a 39 percent decrease ($2,414 per participant per year) in total costs related to the medical care, behavioral health, county jail, and state prison systems.[]The Heartland Alliance Mid-America Institute on Poverty, Supportive Housing in Illinois: A Wise Investment (Illinois: Heartland Alliance Mid-America Institute on Poverty, 2009), 4-10, The average per-person cost related specifically to the use of county jails decreased 68 percent.[]Ibid., 22. Given how far the costs associated with chronic homelessness extend beyond the limited scope and timeframe of the study, the full range of cost savings is likely higher than these initial estimates. Plus, none of these numbers captures the broader social benefit derived from providing people with suitable, sustainable housing. PSH residents in Illinois described supportive housing as vital to improving their perceptions of self, familial relationships, life skills, and overall health.[]Ibid., 4-10.

What Practitioners Should Consider

  • Establish metrics of success that presume substance use crises, cycles of incarceration, or other challenges may not cease immediately once a person is housed. PSH programs serve people with complex needs, for whom other interventions have failed over time. Although accessing PSH can substantially improve outcomes for them, including reducing the likelihood of incarceration, it doesn’t eliminate the risk of being arrested or experiencing a behavioral health crisis while housed. Challenges associated with the transition itself may include adjusting to new responsibilities, coping with distance from the social networks and supports built while unhoused, and addressing long-untreated health issues.[]Gillespie, Hanson, Cunningham et al., Engaging the Most Vulnerable, 2017, 18-23. Like other investments discussed in this report, a robust ecosystem of care providers and social supports is necessary to ensure long-term success.[]Heartland Alliance, Supportive Housing in Illinois, 2009, 26; and National Alliance on Mental Illness, Divert to What?, 2020.
  • Coordinate with a wide array of service providers to connect with PSH participants. Because many chronically unhoused people’s lives are transient, reaching potential participants can be challenging. Denver’s PSH program, for example, found that coordinating outreach and funding, sharing information, and educating community members improves referrals for the program.[]Gillespie, Hanson, Cunningham et al., Engaging the Most Vulnerable, 2017, 17-25.

Interrupting Cycles of Violence without Incarceration

Conversations around reforming the criminal legal system and reducing jail incarceration often exclude crimes considered violent, categorizing them as one uniform type of offense.[]Alexi Jones, Reforms without Results: Why States Should Stop Excluding Violent Offenses from Criminal Justice Reforms (New York: Prison Policy Initiative, 2020), However, interpersonal violence encompasses a diverse range of behaviors. Incarceration frequently exacerbates the root causes of interpersonal violence (such as exposure to violence and unmet economic needs), fails to promote accountability once violence has occurred, and doesn’t empower communities to peacefully resolve conflicts on their own.[]Danielle Sered, Accounting for Violence: How to Increase Safety and Break Our Failed Reliance on Mass Incarceration (New York: Vera Institute of Justice, 2017), 4-8,

Incarceration often worsens the root causes of interpersonal violence, fails to promote accountability once violence has occurred, and doesn’t allow communities to lead their own peaceful conflict resolution.

Additionally, incarceration frequently fails to meet the needs of people harmed by crime.[]Sered, Until We Reckon, 2021, 22-49. In 2016, the first-ever national survey of survivors’ views on safety and justice found that by a margin of 3 to 1, crime survivors believed incarceration was more likely to lead someone to commit crimes in the future than it was to interrupt cycles of harm. The same survey found that most respondents preferred a focus on prevention and treatment to incarceration.

Various stakeholders have developed strategies to prevent, de-escalate, and respond to interpersonal violence by accounting for the factors that shape it and centering the people who experience it. Such strategies include community mediation services and public health–based violence intervention programs. The leaders of these programs may collaborate with criminal legal system agencies that are making referrals to services or undertaking community engagement efforts but rely primarily on support from other sources.

These strategies have been shown to improve conflict resolution skills, minimize criminal legal system involvement, and reduce violent crime.[]Toran Hansen and Mark Umbreit, "State of Knowledge: Four Decades of Victim-Offender Mediation Research and Practice: The Evidence," Conflict Resolution Quarterly 36, no. 2 (2018), 99-113.

Community Mediation Centers

Community mediation empowers people to identify grievances, talk through sources of conflict, and establish their own solutions to violent or otherwise harmful confrontations. Trained mediators who reflect the identities of the people seeking mediation guide participants through this process. Community mediation as a practice varies widely, but the National Association for Community Mediation advises centers to commit to addressing conflict at the earliest possible stages; providing an alternative to criminal legal system involvement; creating a forum to address conflicts; and engaging in public awareness activities, all while being community based, open, accessible, low cost, and inclusive. Mediation centers provide a variety of services to help prevent interpersonal violence; de-escalate existing conflicts; and/or create a mutually acceptable, peaceful resolution when violence has already occurred.

Evidence indicates that the services and resolution processes available through community mediation can be effective without legal system involvement. They produce outcomes that are more satisfying for the people who have been harmed and are more likely to reduce costs, incarceration, and recidivism.[]Hanson and Umbreit, "State of Knowledge,” 2018. Community mediation centers may also act as a hub for additional services designed to address conflict. For example, Neighbors in Action (formerly known as the Crown Heights Mediation Center) in Brooklyn, New York, facilitates several youth development, violence prevention, legal aid, and community-building programs on an annual basis. Neighbors in Action also runs Save Our Streets, a violence interruption program that de-escalated 370 violent or potentially violent conflicts and completed at least 40 high-risk mediations in 2017.

In Baltimore, Maryland, the Baltimore Community Mediation Center (BCMC) provides mediation services for Baltimoreans experiencing any stage of conflict, including mediation within jails and prisons for people approaching reentry. To ensure mediation services are accessible, BCMC partners with other public services and community-based institutions including libraries, churches, and recreation centers to receive referrals and provide space for mediation across the city. In 2018, with help from around 60 volunteers, BCMC held close to 600 mediation sessions at more than 130 different locations across the city.

Helping Baltimore Residents Address Violence

BCMC provides a hub for services that empower Baltimoreans to manage conflicts before, during, and after violence.


What Practitioners Should Consider

  • Account for the racial, ethnic, and gender-specific needs of the people being served. Make use of established and respected community-based institutions that can facilitate residents’ access to mediation. Drawing on these resources can enable mediators to hold sessions in locations that are close and comfortable for participants. For example, the Baltimore Community Mediation Center partners with organizations around the city to provide multiple locations where mediation sessions can take place. Selecting ADA-compliant spaces is important to ensure equitable access for people with disabilities.
  • Avoid connecting mediation center operations too closely with legal system agencies, such as courts. In some instances, programs rely heavily on support from courts that want to shrink their dockets by referring cases to non-court services. Research has linked programs’ reliance on criminal legal system agencies for operational support with reduced autonomy on the part of mediation centers, loss of perceived program legitimacy within the community, compelled participation for people involved in the court system, and loss of focus on community empowerment.[]See generally Timothy Hedeen and Patrick G. Coy, "Community Mediation and the Court System: The Ties That Bind," Mediation Quarterly 17, no. 4 (2000), 351-367, When restorative processes are structured or overseen by criminal legal institutions, the values and conventions surrounding the mediation process can reproduce imbalances that disadvantage participants of color.[]Theo Gavrielides, “Bringing Race Relations Into the Restorative Justice Debate: An Alternative and Personalized Vision of ‘the Other,’” Journal of Black Studies 45, no. 3 (2014), 216-246, 237.
  • Consider the racial and cultural backgrounds of all parties involved, including the mediator. To promote positive outcomes, it’s important to ensure that mediators are trained to navigate cultural differences in a way that facilitates relationship-building with participants and that they reflect the identities of people undergoing mediation.[]Mark Umbreit and Robert Coates, Multicultural Implications of Restorative Justice: Potential Pitfalls and Dangers (St. Paul, MN: Center for Restorative Justice and Peacemaking, 2000), 13-19, For example, one study found decreased hope among participants for a productive resolution when the mediator’s race did not match that of the people seeking mediation.[]Lorig Charkoudian and Ellen Kabcenell Wayne, “Fairness, Understanding, and Satisfaction: Impact of Mediator and Participant Race and Gender on Participants' Perception of Mediation,” Conflict Resolution Quarterly 28 (2010), 23-52.

Public Health–Based Violence Prevention

Public health approaches to violence intervention and prevention prioritize the “contagious” nature of many forms of interpersonal violence, based on research indicating previous exposure to violence is a major predictor that someone will use violence in the future.[]Li-Yu Song, Mark Singer, and Trina Anglim, “Violence Exposure and Emotional Trauma as Contributors to Adolescents’ Violent Behaviors,” Archives of Pediatric and Adolescent Medicine 152 (1998), 531-536, In practice, these approaches engage people involved in violence, health professionals, and the broader community to prevent, intervene in, and reduce instances of interpersonal violence.[]"Who We Are," Cure Violence Global, See also Lashonia Thompson-El, “Great Mentors Create Great Mentors,” Vera Institute of Justice, May 18, 2021, Public health models seek not just to respond to violence when it occurs but also to address the social factors behind violence.

Community violence intervention (CVI) programs work to reduce violence by establishing relationships in communities acutely affected by it. CVIs rely on outreach workers—many of whom have previously engaged in violence, been personally harmed by violence, lost loved ones to violence, or experienced incarceration.

A prominent example of this approach is Cure Violence, which conducts public education campaigns to change attitudes about violence at the neighborhood level, seeks to build relationships with the residents who are most likely to engage in violent behavior, teaches those residents how to avoid violent conflicts, and reduces the likelihood that they turn to violence to satisfy their basic needs by connecting them to economic opportunities.[]Sheyla Delgado, Laila Alsabahi, and Jeffrey Butts, “Young Men in Neighborhoods with Cure Violence Programs Adopt Attitudes Less Supportive of Violence,” JohnJayREC DataBits, March 16, 2017, The Cure Violence model relies on trained “credible messengers,” people who have lived experience with violence in the neighborhood. Cure Violence programs have been successful in multiple cities and are associated with a 30 percent reduction in shootings in Philadelphia.

Hospital-based violence intervention programs (HVIPs) are another public health approach to address interpersonal violence. HVIPs form partnerships between hospital medical staff and community-based organizations to reach people who have been hospitalized after being injured by interpersonal violence. Violence tends to be concentrated in small geographic areas and often recurs because people who are harmed by violence are likely to sustain new injuries resulting from conflict or to use violence themselves—in fact, 41 percent of people treated for violent injuries are reinjured within five years.[]Frank Kennedy, James R. Brown, Karin A. Elliot Brown, and Arthur W. Fleming, “Geographic and Temporal Patterns of Recurrent Intentional Injury in South-Central Los Angeles,” Journal of the National Medical Association
88, no. 9 (1996), 570-572, See also Andrew V. Papachristos, Christopher Wildeman, and Elizabeth Roberto, “Tragic, But Not Random: The Social Contagion of Nonfatal Gunshot Injuries,” Social Science & Medicine 125 (2015), 139-150,

These cycles of interpersonal violence are driven largely by socioeconomic insecurity, isolation, and shame.[]Sered, Accounting for Violence, 2017, 5. HVIPs rely on credible messengers and hospital staff to interrupt the cycle of violence because people are more receptive to interventions that promote behavioral change in the immediacy of hospitalization.[]“Our Background,” Health Alliance for Violence Intervention, During hospitalization, violence intervention professionals engage the patient and their loved ones, providing crisis intervention while offering links to follow-up assistance and other longer term case management. For example, D-LIVE (Detroit Life is Valuable Everyday) is built on a partnership with Detroit’s Sinai-Grace Hospital and has been successful in using individualized therapeutic plans to both connect young people to employment and educational opportunities and reduce the likelihood that they will be reinjured. Of D-LIVE’s 70 participants to date, none have been seriously reinjured and more than 80 percent have either enrolled in an educational program or obtained employment.[]American Hospital Association, Hospitals Against Violence Case Study: DETROIT LIFE IS VALUABLE EVERYDAY (DLIVE) (Washington, DC: American Hospital Association, 2018), 2,

Advocates in Black, Latinx, Indigenous, and poor white communities that have been most impacted by incarceration have long highlighted its harms and limitations. These stakeholders have indispensable knowledge about the needs and resources of marginalized residents. Genuine partnership between government and community-based organizations, particularly those led by formerly incarcerated people, advances racial equity and facilitates power-sharing by bringing traditionally marginalized people to the center of decision-making.[] See for example National Institute of Corrections, Implementing Evidence-based Principles in Community Corrections: Collaboration for Systemic Change in the Criminal Justice System (Washington, DC: DOJ, 2004), 2, It can also expand the lenses policymakers use to understand problems and solutions. Without collaboration, reforms may fail to meet the needs of residents and cause unintended harm for communities of color and other marginalized communities that are already overburdened by incarceration and a lack of effective public investment.

Local advocates and organizations have identified assets in their communities, diagnosed needs, and outlined residents’ policy priorities. Nationally, stronger partnerships between policymakers who direct resources and the organizations that support residents in need could provide for more informed government decision-making. And in some places, resident-led efforts independently drive these initiatives to inform change. For example, in Milwaukee, Wisconsin; Detroit, Michigan; Portland, Oregon; and New York City, community-based organizations have mapped and amplified the perspectives of marginalized residents through grassroots efforts with little to no involvement from local government.

Grassroots Strategies to Elevate Community Expertise

Group Created with Sketch. Read More

In other places, such as Los Angeles and Washington, DC, local governments and community-based organizations have partnered directly to identify improved approaches to safety and justice. For example, Offices of Violence Prevention (OVPs) work to move public safety supports away from being exclusively police- and criminal legal system-oriented and into the hands of community members through learning exchanges, leadership development, and capacity building for data and reporting. OVPs call for investments in community-based interventions, prevention, and development to increase the expertise and effectiveness of city agencies.

Although collaborations like these can be powerful in gathering input from a wide array of local stakeholders, they can also heighten existing distrust between community-based organizations and local government if the information gathered is not meaningfully reflected in future policymaking. Successful, collaborative efforts are built on the principles of respect, transparency, and partnership with a clear plan for the findings to shape future decision-making.

Collaborating with Community-Based Organizations

Group Created with Sketch. Read More

Why Criminal Legal System Responses Are Not Enough

With growing recognition of the human and financial toll of jail incarceration, local governments across the country have sought new ways to promote safety. However, many current approaches to reducing the use of jails still rely on criminal legal system agencies and the threat of incarceration. They focus on growing staff, budgets, and other resources within criminal legal system agencies to expand their options for addressing problems related to poverty and behavioral health.[]Mike Crowley and Betsy Pearl, “Reimagining Federal Grants for Public Safety and Criminal Justice Reform,” Center for American Progress, October 7, 2020, These efforts fail to address many of the underlying causes of violence and other criminalized behaviors that would be better addressed through other agencies, organizations, and community-led efforts—issues like unstable housing, poverty, limited educational opportunities, poor health, and inadequate access to services.[]National Alliance on Mental Illness, Divert to What?, 2020. They also fail to account for the racialized harm caused by decades of investments prioritizing criminal legal system agencies over community-based services and often ignore existing problematic system practices. These shortcomings limit both their efficacy and their reach.

Homelessness courts and police-led diversion are two prominent examples of limited-impact reforms that make access to services dependent on interactions with legal system agencies.

Homelessness courts respond to chronic homelessness with diversionary proceedings to connect unhoused people to resources. However, these courts have been criticized for legitimizing the criminalization of homelessness and failing to address its root causes because they use the court system and prosecution as gateways to services.[]Sarah Lustbader, “Are Problem-Solving Courts Impeding Progress?” The Appeal, January 7, 2020, Even in jurisdictions where police issue civil citations instead of arresting unhoused people for low-level “quality of life” charges, a person’s inability to pay the associated financial sanctions can trigger other penalties that make continued homelessness or future arrest more likely.[]Jessica Mogk, Valerie Shmigol, Marvin Futrell et al., “Court-Imposed Fines as a Feature of the Homelessness-Incarceration Nexus: A Cross-Sectional Study of the Relationship Between Legal Debt and Duration of Homelessness in Seattle, Washington, USA,” Journal of Public Health (Oxford, England) 42, no. 2, 2019, 1-6,; and Robinson, “No Right to Rest,” 2019, 41-73, 43-44.

With police-led diversion, law enforcement officers connect a person in crisis to services instead of booking them into jail.[]S. Rebecca Neusteter, Ram Subramanian, Jennifer Trone et al., Gatekeepers: The Role of Police in Ending Mass Incarceration (New York: Vera Institute of Justice, 2019), 32-35, Although such programs have been associated with fewer jail bookings, their reach is limited.[]For a more thorough review of existing evaluations of pre-arrest diversion programs, see Robin S. Engel, Robert E. Worden, Nicholas Corsaro et al., Deconstructing the Power to Arrest: Lessons from Research
(Cincinnati, OH: International Association of Chiefs of Police/UC Center for Police Research and the John F. Finn Institute for Public Safety, 2018), 41-62,
The process usually begins with an officer’s decision to use it, so diversion depends heavily on both the organizational culture of the law enforcement agency and the discretion of individual officers.[]Ibid., 41-42. In many agencies, the practices, priorities, and organizational culture can be counter to serving people who are experiencing a behavioral health crisis.[]For example, the crime fighter mentality that often permeates police cultures can result in perceiving people needing help as threats to themselves, officers, or other community members. One study found that law enforcement officers are between 1.4 and 4.5 times more likely to use force in interactions with someone who has a mental health condition. Robin Shepard Engel and Eric Silver, “Policing Mentally Disordered Suspects: A Reexamination of the Criminalization Hypothesis,” Criminology 39, no. 2 (2001), 225-252, See also Westervelt, “Mental Health And Police Violence,” 2020. For example, law enforcement departments often generate a “warrior culture,” in which officers identify primarily as fighters of crime. This creates distance, and sometimes tension, between officers and residents through an “us versus them” mentality.[]Sue Rahr and Stephen K. Rice, From Warriors to Guardians: Recommitting American Police Culture to Democratic Ideals (Washington, DC: DOJ; 2015), 5-7, Even with diversion programs and additional training, the high costs of relying on law enforcement for behavioral health crisis responses have persisted—including the risk of serious injury or death.

To be responsive to residents’ needs and account for the harm caused by incarceration, jurisdictions across the country must look beyond jails and the criminal legal system for public safety solutions. Effectively ending the current dependence on jail incarceration requires an ecosystem of services and supports that enhance the mental, physical, and socioeconomic well-being of the people who have been most marginalized. The list of programs presented here is not exhaustive, and despite sharing common challenges, no two communities are exactly alike. It’s important to tailor approaches to fit specific local contexts. Investments should be made with a spirit of innovation and experimentation toward a goal of transformation and repair, acknowledging histories of harm and past failures. Policymakers and practitioners must build relationships founded on partnership and power sharing with community-based organizations, particularly those led by formerly incarcerated people, use direct funding or other in-kind support to help identify local assets and needs, and provide resources to expand their work.

Ultimately, a network of community-based services and supports could go a long way to address criminalized behaviors in ways safer and more effective than jails. Expanding non-jail solutions is a key part of a broader strategy to improve racial and economic justice for all communities, but it is not a panacea for the harms of the current system—both past and present. Commitment from government agencies and community-based organizations alike is crucial to shrink the criminal legal system’s footprint and end reliance on arrest and incarceration to address social concerns like homelessness, behavioral health crises, and interpersonal violence.


Author: Melvin Washington, program associate II, Vera Institute of Justice

Editor: Elle Teshima, editor, Vera Institute of Justice

Production Consultant: Gail Ablow

Video Editor: Judith Wolff

Web Design: Michael Mehler, web engagement director, Vera Institute of Justice, Karen Ball, front end developer, Vera Institute of Justice, Madison Hindo, digital content coordinator, Vera Institute of Justice, and Surbhi Chawla, digital designer, Vera Institute of Justice

Photo Credits:

Baltimore: Elvert Barnes, Community Mediation Center, Blink O’Faneye

Chicago: Wilhelm Joys Anderson, Thee Erin, Sean T. Evans, Patrick Griffin, PMonaghan, Tripp

Denver: Urban Institute Courtesy of Maura Friedman, Robert Kash, Brandon Marshall

Tucson: Vincent Aguilar, Anthony Dolce

Vera thanks representatives from the Westside Community Triage and Wellness Center in Cook County, Illinois; the Colorado Coalition for the Homeless in Denver, Colorado; Connections Health Solutions in Pima County, Arizona; and the Baltimore Community Mediation Center in Baltimore, Maryland. The members of these organizations were gracious enough to volunteer their time during a global pandemic and share their perspectives via video interviews. We are beyond grateful for their willingness to contribute their insights to this project. We thank Patrick Griffin of the MacArthur Foundation for his support throughout the development of this report. We also thank Madz Reeve for their help coordinating and conducting the interviews; Gail Ablow and Judith Wolff for their amazing work producing and editing the interviews; Jose Enriquez and Jonathan Prellwitz for coordinating the photo shoot in Tucson; and John Wehmann for helping with the Urban Institute photos. We also would like to thank our colleagues Elizabeth Swavola, Cindy Reed, Kica Matos, Daniela Gilbert, Jason Tan de Bibiana, Jackson Beck, and Cameryn Okeke for their thoughtful reviews and feedback; Maris Mapolski, EpsteinWords, and Jac Arnade-Colwill for editorial support; Sara Duell for design assistance; and Chris Choi for digital strategy.

This report has been made possible by funding from the John D. and Catherine T. MacArthur Foundation’s Safety and Justice Challenge, which seeks to reduce over-incarceration by changing the way America thinks about and uses jails.

Chicago, IL
Tucson, AZ
Denver, CO
Baltimore, MD