Next year, Americans will start hearing a great deal about 988, a new alternative to 911 for “Americans in crisis to connect with suicide prevention and mental health crisis counselors.” Its purpose: to isolate such cases so that they can be responded to in a more specialized fashion—with social workers and, in some cases, specially trained cops, as opposed to just a random officer who happens to be closest to the emergency. The FCC has required phone companies to make 988 operational by July 2022, but the true significance of the move is less about a new number than about a new way of reacting to mental-health emergencies. Advocates believe that the new system may lead to fewer mental-health-related shootings by police and less involvement of the mentally ill in the criminal-justice system.

But after 30 years of experience with enhanced police training in techniques such as de-escalation, cities remain uncertain about what works, and to what degree. We should not expect social-services providers, who don’t always garner much respect from the population they’re trying to help, to perform miracles on the street. Without deeper reform that treats mental illness more thoroughly, the new initiative won’t make much headway. If crisis-response reform diverts attention from deeper reform, it may even prove counterproductive.

Mental-health emergencies are estimated to make up anywhere from 5 percent to 20 percent of police calls for service. Sending out police to respond to such calls is often said to be wasteful and dangerous. In the United States, 200 to 250 fatal police shootings that are in some way mental-health-related happen each year.

Fatal police shootings of mentally ill people are tragic but rare. Spread across a nation of nearly 20,000 law-enforcement agencies, 200 to 250 incidents annually translate to very few per jurisdiction per year, even in big cities—and, in most places, zero. The rarity of fatal shootings of mentally ill people explains much of why we know so little about how to stop them. Any evaluation of a de-escalation program designed to reduce such tragedies will be dealing with a vanishingly small sample size.

Training programs vary in quality and character. Departments don’t track and code mental-health crisis calls in a uniform manner. Even were a researcher to prove, conclusively, that one city’s de-escalation program deserves credit for cutting the number of fatal police shootings of mentally ill people from, say, three in one year to one in the following year, that would prove nothing about programs in other cities.

The essence of alternative-response protocols is persuasion. Advocates want to see cops resort less to force when handling tense situations. In these partisan times, a lack of experience with persuasion seems to have inspired credulousness about the conditions under which it is likely to happen. The situations in which persuasion will be most effective are those least likely to result in tragedy. Conversely, there could scarcely be worse conditions for persuasion than situations involving a man with untreated psychosis armed with a knife and charging at a cop. Antipolice activists heap scorn on the quality of the average officer, while also promoting a superhero conception of policing that implies near-magical powers of persuasion.

Nor should we expect that social-services personnel, in dangerous situations, will excel at persuasion. Police-defunding advocates exaggerate the degree to which mentally ill people like social-services personnel. Social workers are often seen, by the population they’re employed to help, as paternalistic and unreliable, constantly misleading them about being able to do something to help when they know they can’t. Street homeless individuals, in particular, develop notoriously deep “service resistant” instincts as a consequence of having been, as they see it, betrayed by the system for years.

Advocates also expect too much from “peers”—people who have overcome their mental illness and are now employed to help others do the same. It’s presumptuous to believe that someone with one kind of experience of mental illness will possess special influence over someone with an entirely different experience and whom he has only just met at a crisis scene. The appeal of social workers and peers, as response team members, is mainly negative: they are not armed and not authorized to arrest.

Supporters of 988 and reformed crisis protocols invoke the persistence of police shootings to make their case for the urgency of reform. But they lack evidence that proves that their reforms will reduce shootings. They thus retreat to safer ground. One good example is tracking arrests. Claiming that reforming crisis protocols is really about reducing arrests reframes the endeavor as one of “diversion,” which addresses the “criminalization” of mental illness—meaning the high rate of serious mental illness among jail and prison inmates. The goal is to divert as many mentally ill individuals away from more and deeper involvement in the criminal-justice system, and that must start at that system’s front end, with patrol officers.

But “diversion” raises the question: Diverted to where? What happened to someone instead of being arrested? The deinstitutionalization of the mentally ill was wildly successful in diverting, over the decades, hundreds of thousands of Americans away from mental hospitals—but it diverted them to homelessness and jails and prisons.

The success of diversion hinges on the strength of the local mental-health system and police departments’ relationship with it. The most admired crisis-response systems, such as those found in Arizona, utilize crisis-stabilization centers, facilities to which officers can quickly deposit people in a state of psychiatric emergency instead of booking them on a charge. They’re staffed and designed to feel less like a hospital ER, while also offering many of the same benefits of one. But not every jurisdiction has a crisis-stabilization center, and even the best-run of such facilities raise their own “diversion to where?” questions. Crisis-stabilization centers are meant only to address the 24 hours after a psychiatric emergency hits.

In the most immediate and concrete sense, what 988 will do in most places is direct calls, from anywhere in the country, to the national suicide hotline, which has long been in use but whose number is harder to remember than “988.” Modest federal funding will be provided to help localities develop new response systems, and enabling legislation authorizes states to impose fees on cell-phone bills for that purpose. But in general, it will be up to states and cities to decide how expansively they want to invest in peers, crisis-stabilization centers, and so on. Even where those options exist, dispatchers will have to decide which calls are safest to send social workers to handle.

It also remains to be seen how much 988 catches on with the public, and in what way. Will it function as more than just a suicide hotline—as a “911 system for the brain,” as some hope? In jurisdictions that already have mental-health crisis lines, one person sometimes calls that line about an incident while another calls 911 about the same incident.

Another metric of success to monitor, as 988 gets rolled out: How much does it promote authentic mental-health reform? Such reform targets the mental-health system proper, not cops and dispatch personnel. Crisis-response reform is better understood as a variety of criminal-justice reform, with a mental-health orientation. As the late DJ Jaffe relentlessly emphasized, real mental-health reform equates to a reduced rate of serious mental illness among the incarcerated and homeless populations. If fewer mentally ill are winding up in jail, on the street, and in shelters, we’re probably getting somewhere.

De-escalation is a reasonable idea that I’ve praised in the past. I continue to believe that training cops makes more sense than trying to replace them. Over time, though, I’ve grown increasingly concerned that, for some mental-health advocates, the crisis-response debate represents an elaborate attempt to change the subject. De-escalation could function as part of a substantive mental-health reform agenda. But too often, it functions as an alternative to such an agenda. Pushes to reform crisis response have done nothing to advance civil-commitment reform, the most necessary mental-health reform of all. During the last 30 years, we’ve witnessed the spread of de-escalation training programs and the loss of tens of thousands of inpatient psychiatric beds. Policymakers avoid civil-commitment reform because it’s controversial, though more so among disability-rights groups than with the broader public. Aghast at the spectacle of subway pushings and mass shootings, the public grasps the appeal of more court-ordered supervision for the mentally ill. But we never get around to debating it. We remain absorbed in numerous, far less consequential, matters. When should dispatchers send out social workers only, cops only, or “co-response” teams comprising both? How many crisis-stabilization centers does a community need, and where should they be sited?

Mental-health reform designed to help “everyone” rarely helps the hardest cases. Suicide hotlines represent a public-health response to mental-health emergencies. Public-health programs are, by definition, untargeted, set up to benefit the broader public. Mainstream mental-health organizations will tell you that one in four adults experiences a mental disorder in any given year. But one in four adults is not at risk of being fatally shot by police in any given year; one in four adults does not suffer from incapacitating mania or psychosis. A true “serious” mental illness can qualify people for disability benefits and lead to involuntary hospitalization. Only a small fraction of adults with a diagnosable mental disorder qualify for the former, and an even smaller segment will ever be institutionalized. Call lines will do the most to help people who know that they have a problem and want help. What about those who’ve sworn never to trust the system again, having been betrayed by it too many times? What about people so mentally ill that they don’t even believe that they’re mentally ill?

Pushes to reform response protocols are heavily premised on the idea that police are prejudiced toward mentally ill people—they think they’re violent, when they’re not. So cities have cops take time off from their patrol duties to listen to mental-health professionals give lectures about anxiety, PTSD, personality disorders, and “stigma.” This kind of instruction is poorly targeted. Teaching patrol officers about many mental disorders beyond just incapacitating mania and psychosis will elevate their general knowledge of mental health, yes—but we could elevate cops’ knowledge about many social-policy topics only loosely connected to their duties: housing, the ins and outs of Medicaid financing, the history of the subway system. Being force-fed lessons about how most mentally ill people aren’t violent serves as poor preparation for how to handle mentally ill people who demonstrably are violent. Between January 2015 and June 2021, the Washington Post documented 1,474 fatal police shootings that were somehow mental-health-related. In 916 of those cases (62 percent), the victim was attacking someone and was usually armed with a gun or knife. In two-thirds of the cases not coded by Washington Post researchers as involving an attack, the victims were armed with a gun or knife. The database codes only 43 out of the 1,474 incidents (3 percent) as involving victims definitively not attacking anyone and definitively unarmed. About 10 percent of police shootings are instances of “suicide by cop,” which mostly involve mentally ill people. The calls for service likeliest to end badly, such as those involving a psychotic person wielding a knife, are the ones that social workers are the least well suited to handle.

Knowledge is power in fast-moving crisis situations, but the most valuable form of knowledge may simply be that of the community. People with untreated schizophrenia or severe bipolar disorder behave erratically. They do and say things that strangers interpret as threats but that their families and neighbors know to be innocuous. Knowledge of a community and its members is not gained through listening to lectures about stigma but through experience: walking a beat or responding to hundreds of calls on patrol, attending barbecues, and so on. Giving cops knowledge of the community requires a serious commitment on the part of governments because experience is an expensive mode of instruction. “Community policing,” rightly understood, is labor-intensive. The cheapest model of policing is to keep a small number of cops in reserve and available to respond to dangerous emergencies, and not much beyond that. Experienced cops are costlier than new recruits. Knowledge of the community is, of course, jeopardized by waves of retirements of veterans.

One benefit of the “criminalization of mental illness” debate is that it incorporates the voices and experiences of criminal-justice-system participants. Judges, prosecutors, jail and prison staff and managers, and police commissioners and patrol officers have built up, over the years, extensive knowledge of serious mental illness. Some research suggests that over 90 percent of patrol officers have had encounters with mentally ill people. What percentage of the general population has substantial experience with people with mental illness, especially the particularly disturbing psychotic variety? What percentage of “mental-health professionals” has substantial experience with people with untreated psychosis and violent tendencies?

We will always need cops to be involved in responses to mental-health emergencies. Transportation alone guarantees it. Helping someone in crisis often entails taking him to another location, such as a jail, hospital, or crisis-stabilization facility. That will often require the assistance of police. We could, instead, hire entirely separate and new teams of “Crisis Transport Security Officers,” but if we did that right, they’d resemble cops in many ways. Social workers sometimes welcome the presence of cops on the scene of crises because not having to worry about their personal security helps them focus on their particular expertise.

Cops, for their part, would just as soon not have to deal with mental-health emergencies. But it’s not true to say that they’re unqualified to do so. Police have a great deal of hard-won knowledge about the nature of untreated mental illness that they can usefully bring to bear to resolve particular crises, or as contributions to the broader debate over mental-health-policy reform. Mental-health emergencies are too important to be left to the experts.

Photo: Contrary to the claims of some advocates, cops will always have a role in dealing with disturbed people. (BRITTANY MURRAY/THE ORANGE COUNTY REGISTER/AP)


City Journal is a publication of the Manhattan Institute for Policy Research (MI), a leading free-market think tank. Are you interested in supporting the magazine? As a 501(c)(3) nonprofit, donations in support of MI and City Journal are fully tax-deductible as provided by law (EIN #13-2912529).

Further Reading

Up Next