In Rock Bar, in San Francisco’s Bernal Heights neighborhood, a bartender watched as a man who appeared to be homeless shouted on the street outside. The man waved his hands and yelled at the street, seemingly at no one; then, he abruptly ran in front of an SUV that was driving down the block: “Get out of here, kids!” he yelled, blocking traffic. “I’m not gonna hurt anyone!” The bartender watching was concerned, but he did not dial 911. “I don’t want to call the cops on somebody and have them locked up,” he explained in the Mother Jones article where this incident was reported. “That’s what normally happens.” Eventually, someone else in the neighborhood called 911. Contrary to the bartender’s expectations, however, the police did not show up to make an arrest: instead, a red van staffed by three unarmed crisis responders and stocked with supplies such as blankets, socks, snacks, Narcan, tampons, and toothbrushes, arrived at the scene. The crisis responders included a paramedic from the San Francisco Fire Department, a behavioral health clinician, and a peer specialist. The latter two were wearing informal clothes, and all of them were unarmed. The responders gave him a water bottle and a snack, took his vitals, and talked to him while he gradually calmed down.

The team offered to connect him with a shelter or a drug detox clinic for long-term care, but the man refused. They made a note to follow up with him later through the San Francisco Office of Coordinated Care, whose outreach workers connect with people again after an initial encounter with crisis responders. The man declined a referral or a follow-up, which happens on around 16% of the calls that the team responds to. Still, he was much more relaxed by the time he said goodbye. “God bless you all,” he said quietly as he walked away. The bartender watching was impressed by the team. He commented to Samantha Michaels, who reported the story, that in his sixty years of living in the neighborhood, and throughout his own struggles with substance abuse, he had never seen such a response from the city.

The van of crisis responders that came to the man’s aid is part of the San Francisco Street Crisis Response Team. As mentioned in our introductory post, the Street Crisis Response Team is one of many community responder models that are going into operation in cities across the U.S., in which unarmed mental health and medical professionals respond to low-risk, nonviolent 911 calls instead of the police. They represent an emerging fourth branch of emergency response that offers social resources and psychological support to supplement the traditional police, fire, and emergency medical departments. As these models decrease police contact with the public, they have the potential to disrupt systems that sometimes criminalize the mentally ill, and that disproportionately affect Black, Latinx, and Native American communities. Below, we will take a closer look at the San Francisco Crisis Response Team and contrast it with other Bay Area alternative first responders, including the Mobile Assistance Community Responders of Oakland (MACRO), the Community Mobile Response Teams in Santa Clara County, and the developing Specialized Care Unit in Berkeley.

Evolution of Alternative Emergency Response

Before we discuss these programs in relation to each other, it is worth considering why they may differ from each other. In the wake of the murder of George Floyd at the hands of police in 2020, researchers and politicians held up the CAHOOTS program, which has been operating in Eugene, Oregon, since the 1990s, as a “proven model” for unarmed alternative response. However, as Tim Black, the former CAHOOTS Operations Coordinator has said, “CAHOOTS isn’t a cookie-cutter program. This is not a one-size-fits-all program. We have something that’s been shaped by the communities of Eugene and Springfield.” Indeed, the CAHOOTS program emerged to suit the specific needs of Eugene and Springfield, which, at that time, were the behavioral health and addiction issues that sometimes surfaced at concerts, music festivals, and anti-war protests. Their program was an early, innovative forerunner for alternative response; however, it was designed to suit the particularities of these communities, and can not be blindly copied and pasted into every city.

One major difference between CAHOOTS and many of the models being piloted today is that the cities that CAHOOTS operated in are more than 80% white. As a result, the team behind CAHOOTS did not have to navigate the lack of trust that some racial and ethnic communities may feel towards calling 911. In the same vein, it is worth noting that while we are drawing attention to differences between these programs, we do not mean to hold up any model as an ideal, because each program operates in circumstances that are difficult to compare. The difference in contexts raises local questions for each program about how a model like CAHOOTS can be adapted to build trust and serve racially or ethnically marginalized communities, and how alternative first response may need to differentiate itself from some of the negative connotations that may accompany police responses.

The Model Matters: How the Structure of a First-Response Team Can Impact its Effectiveness

The roles that operate on a first-response team are diverse. The occupations that cities generally chose from to serve on their teams include: a peer specialist who has lived experiences with issues that clients may be struggling with, such as homelessness, substance use, or experience with the criminal justice system; a case manager who can link the individual to other community resources and create follow-up plans; a crisis intervention specialist, who has completed training in de-escalation, care for substance abuse, and psychological support; a behavioral health clinician who can also provide emotional support, as well as assess whether involuntary hospitalization may be necessary; and some kind of emergency medical professional or community paramedic who can provide immediate healthcare. Researchers have commented that the workforce and career pipelines for these kinds of roles do not exist yet, and the question of how to build employment infrastructure to serve this field remains pressing.

These roles may vary by city and can change in response to the needs of the program. In San Francisco, for example, the Street Crisis Response Team originally consisted of a behavioral health clinician, a peer specialist, and a community paramedic. Recently, the city made the decision to replace the behavioral health clinician with an Emergency Medical Technician. Now, the clinicians work on case-by-case follow-ups with people after the Street Crisis Response Team has already met them. This change drew criticism from clinicians who were formerly on the team who contended that the team was already successful, and that the removal of the clinician detracted from the quality of the emergency mental health care that the team could provide. The city explained that the change was made to dedicate more attention towards long-term care and follow-up; a focus on dedicated, repeated interaction has proven successful for another one of San Francisco’s homeless aid programs. Difficulty in hiring and retaining staff to perform these roles, a challenge which we will elucidate more on in a future article, may also have played a role in their decision.

For context, similar to San Francisco, Santa Clara County’s new TRUST mobile crisis response team also does not have a licensed clinician on board: their three-person team consists of a crisis intervention specialist, a first aid responder, and a peer support specialist. The MACRO team in Oakland employs a case manager as well as a crisis intervention specialist and an emergency medical technician. They explain that their lack of mental health professionals is because licensed clinicians are difficult to recruit and train, and because the field as a whole lacks diversity, which can compound uneasiness among residents. They further explain that they reserve clinicians for more serious emergency calls that involve police officers so that the clinicians can assess for involuntary hospitalization. The Specialized Care Unit that Berkeley is planning currently does include a behavioral health specialist, in addition to a peer counselor, and a medical worker.

The question of whether a clinician should work on a mobile crisis response team does not yet have a clear answer. While there is no research-backed consensus, the federal Substance Abuse and Mental Health Service Administration’s minimum expectations for mobile crisis teams include staffing “a licensed and/or credentialed clinician capable of assessing the needs of individuals within the region of operation.” They clarify that this could manifest as a Bachelor’s level clinician on the team and a Master’s level clinician on-call as backup, who is consulted as needed. In practice, however, the composition of crisis response teams varies greatly across the country. CAHOOTS, the widely cited Eugene model that was discussed earlier, does not have a clinician and is staffed by a crisis intervention worker, a paramedic, and an emergency medical technician instead. In contrast, STAR, an alternative first responder in Denver that Stanford researchers have deemed effective for certain kinds of calls, staffs a two-person team composed of a mental health clinician and a paramedic. In general, regarding the choice of first responders, experts in the field urge that the design of the team should be customized to fit the needs of the specific community that it serves. Kari Auclair, director for Western Montana Mental Health Center has suggested that “[i]n some communities it’s going to be the church group that’s going to be part of a crisis response because that’s who people go to and that’s what they’ve got.”

The Substance Abuse and Mental Health Service Administration’s best practice guidelines capture this need for community trust in their recommendation for including a “peer support worker” on a mobile crisis response team. The San Francisco Street Crisis Response Team does staff a peer specialist, seemingly similar to Berkeley and Santa Clara County, and similar to the Community Intervention Specialist in Oakland. The peer specialist can play the important role of bridging the lack of trust that may persist towards institutional first responders, when, in some communities, first responders have been associated with arrests and police violence. While the paramedic on the team wears an official Fire Department uniform, which can be “seen as very institutional,” the peer specialist instead wears civilian clothes and approaches the client from the perspective of having lived through similar experiences. This is one way in which, through design choices, the alternative first response team can differentiate itself from the institutional connotations of the police department.

Future of Alternative First Response in the Bay Area

Still, this lack of trust is not easily overcome. Even while cities pilot these alternative response programs, some nonprofits have taken matters into their own hands to provide a less institutionalized community response to crises. In Oakland, in parallel with the official government MACRO program, activists from the Anti Police-Terror Project established Mental Health First Oakland, a hotline that is completely unaffiliated from 911, and offers phone line support under the principle that “police should not be involved when responding to a crisis unless asked by mental health responders as a last resort.” This kind of project highlights the fact that when a person calls 911, even in a city with an alternative first response team, they are not always guaranteed an unarmed response. There is still a chance that the police may show up, as what happened last August in San Jose, when a non-profit called 911 to request an ambulance for a woman in a mental health crisis, but instead got 13 police officers who prevented the crisis team from approaching the patient.

Non-profit leaders in San Jose and Santa Clara County have cautioned that there is a need for patience while the departments implement and adjust alternative first response measures. Still, incidents like this illustrate why communities may be hesitant to rely on 911-based support, and raise concerns about how alternative response can be implemented and scaled while remaining transparent about their involvement with the police. Additionally, these incidents raise the broader question about whether, even once their growing pains are resolved, a government responder is the most appropriate choice to resolve the crises that happen in a community, especially in a community with a history of systemic racialized police violence.

The way that these alternative response programs are implemented, scaled, and expanded varies by city, drawing attention to the early choices that policymakers make when planning alternative first response pilots. In one year, the San Francisco Crisis Response Team expanded from a single team that supported the Tenderloin neighborhood to six teams that provide San Francisco with 24/7 citywide coverage. In a May 2022 report, they wrote  that their response times have become even faster than the police’s response times for comparable calls. As of February 2022, the Street Crisis Response Team’s now seven response teams have expanded the call types that they are eligible to respond to in order to serve a wider range of calls.

Other alternative first responder programs in the Bay Area remain in more preliminary stages of development. In Oakland, MACRO is in the midst of an 18 month pilot, which was initially launched in East and West Oakland and now serves the entire city. It currently operates 7 days a week, from 6:30am to 1:45pm, with swing shifts from 2pm-10pm that are being offered as staffing permits, while it aims to eventually achieve 24/7 deployment. Meanwhile, three mobile crisis response teams operate in Santa Clara County, one of which recently received a grant to expand into North and South County and the West Valley, and another of which was launched last Autumn. Finally, in Berkeley, the pilot version of their alternative first responder, called the Specialized Care Unit, will launch in June. Even while considering the diversity of contexts that these programs operate in, the variance in program design engenders a growing set of questions researchers and policymakers should consider: What does effectiveness look like for an alternative first responder? What metrics are useful in assessing the success of a certain program? And which programs are the most successful?

In the next article we will consider the difficulties that go into scaling these programs from experimental pilots into established departments, and the lessons that have been learned about alternative first response that other cities and non-profits can consider.