Strategy overview

  • Responding to behavioral health crises: The foundation of non-police emergency response is to alter or divert 911 call responses in the event of a mental health crisis or behavioral emergency. Rather than deploying police officers, jurisdictions send trained behavioral health professionals (such as a social worker), who may be accompanied by a medical professional (i.e. a nurse or EMT). The responding team is typically trained in a wide range of emergency support services focused on health and de-escalation.
  • Partnering with health clinics: Most alternative emergency response programs are a partnership between a local jurisdiction’s emergency response agency and community health providers. Oftentimes, a health clinic will provide crisis workers (who may be social workers, substance abuse specialists, community health workers, and more), who are trained to respond to mental health emergencies, including conflict, substance abuse, suicide threats, and more. The jurisdiction may also provide a nurse, paramedic, or EMT to provide medical services as needed. Crucially, no member of the response team carries a weapon.
  • Ensuring 911 calls are low-risk: Diversion of 911 calls is strictly reserved for low-risk situations, wherein an individual poses no physical threat to others or themselves. Successful programs often feature specific criteria that help dispatchers determine the extent to which an individual poses a threat to themselves and others.
  • Reducing emergency costs: In many instances, a resident in crisis does not need to be sent to the emergency room via ambulance. Because first responders are trained in behavioral health crisis de-escalation, emergency calls are more likely to be resolved onsite and without engaging costly medical equipment and staff.

What evidence supports this strategy?

Two quasi-experimental studies on alternative emergency response models indicate promising results. More comprehensive scientific research is needed.

  • A 2022 quasi-experimental study on a community response pilot in Denver found that the intervention reduced reports of minor crimes (e.g., disorderly conduct, drug use) by 34 percent and had no effect on serious crime.

  • A 2021 quasi-experimental study found that individuals receiving a co-response intervention (police and medical professionals) in Indianapolis were less likely than those receiving a traditional police response to be arrested following the incident. However, there were no effects on justice involvement, while demand for emergency medical services increased at 6- and 12-month intervals.

Is this strategy right for my community?

Providing non-police emergency response services has been shown to improve outcomes predictive of upward mobility. These outcomes, identified by the Urban Institute, are safety from trauma, safety from crime, and just policing.

City and county leaders can assess local conditions for each of these outcomes using the metrics below, identified by the Urban Institute. This assessment can be used to determine whether this strategy is appropriate for their community. (Note: these metrics are a starting point for self-assessment and are not intended to be comprehensive.)

All cities and counties with populations over 75,000 can receive a customized data sheet here.

  • Measuring safety from trauma in your community: Examine the number of deaths due to injury per 100,000 people. These data are available from the National Center for Health Statistics’ Mortality File and the CDC’s WONDER database.

  • Measuring safety from crime in your community: Examine reported property crimes per 100,000 people and reported violent crimes per 100,000 people. These data are available from the Federal Bureau of Investigations’ Uniform Crime Reporting Program.

  • Measuring just policing in your community: Examine the number of juveniles arrested per 100,000. High rates of juvenile arrests provide a strong indicator of overall system involvement and over-policing. These data are available from the Federal Bureau of Investigation’s Crime Data Explorer.

Best practices in implementation

  • Partner with experienced mental health crisis responders: To deliver the model effectively and with maximum impact, prioritize developing partnerships with health clinics demonstrating mental health crisis response as a core competency. Ideal partner clinics may also be able to train new crisis responders to meet capacity as a program expands within a jurisdiction.
  • Set clear diversion criteria: Effective alternative emergency response begins at the point-of-contact with a 911 caller. Jurisdictions should partner with mental health experts, especially a clinic partner, to set clear criteria for determining that a police presence is unnecessary. Additionally, resident callers must be able to request alternative emergency services.
  • Implement the program gradually: In many jurisdictions, diversion of 911 calls to alternative responders represents a significant departure from decades of practice. Given the significance of the change, effective programs are often piloted in a small geographic boundary to help manage capacity constraints, refine the program, and demonstrate the impact of the model to the public. A gradual, precinct-by-precinct implementation can help a jurisdiction balance the benefits of the model with the challenges of altering emergency response protocols.
  • Invest in strong, public-facing data infrastructure: Alternative emergency responders, like social workers and community health specialists, are trained to collect and analyze various wellbeing measures that do not normally fall within the scope of 911 response (such as mental health measures and housing status). Many implementing jurisdictions partner with universities or other institutions who can conduct a program evaluation to determine its effects on homelessness, public health, and more.