One of the most tragic developments in policing in the last forty years has been the massive expansion of their role in managing people with mental illness and other psychiatric disabilities.1 The police have always had to deal with mentally ill individuals whose behaviors are criminal or create a substantial public nuisance. With the massive deterioration in mental health services, the scope and number of these interactions have changed. The police are often the main agency engaged in both emergency and ongoing management of segments of this population. While most such interactions are handled reasonably well, too many result in arrest, incarceration, injury, and even death. The police are particularly ill-suited for this role, given their other functions; relying on police, jails, and emergency rooms to “manage” people suffering from mental health problems is expensive and inefficient, and does little to improve their quality of life.
The United States suffers from particularly inadequate mental health care services. While psychoactive drugs have brought increased independence for many in recent decades, many are unable or unwilling to maintain pharmacological treatment, many do not have access to basic mental health services, and ongoing community-based services are few and far between. As a result, in a crisis, patients and families have little choice but to call 911—and it’s typically the police who respond.
Egon Bittner, in his classic 1967 study, identified the difficult choices officers face when they arrive at a scene.2 Ideally, an officer assesses the situation and decides whether the person should be taken to a psychiatric emergency room for temporary voluntary or possibly involuntary committal, arrests the individual, or attempts to resolve the issue informally. Police typically prefer the latter option, but often feel compelled to take one of the others because the behavior is serious or seems likely to continue unabated if not addressed. In these former cases, the officer must take the person into custody, sometimes against their will. This means using verbal coaxing if possible but, if necessary, force.
US police officers kill hundreds of people with mental illness (PMI) every year, according to a count by the Guardian.3 The Treatment Advocacy Center reviewed the literature on fatal police encounters and estimates that one in every four police killings is of a person with a mental illness, meaning they are sixteen times more likely to be killed by police than other people.4 The killings of PMI take a few general forms. In some cases, police arrive on the scene and encounter someone with something they perceive to be a weapon, such as a screwdriver or kitchen implement. That person refuses to drop the object and sometimes threatens the officer or others, prompting police to open fire. This can be seen in three recent videotaped incidents:
•In August 2014, Kajieme Powell was clearly mentally distraught and had a knife. Officers arrived on the scene and yelled commands at him from dozens of feet away. When Powell took a few steps toward them, they shot him to death.5
•In May 2015, the mother of Jason Harrison called 911 requesting help for her son, who was refusing to take his medication. When police arrived, she casually walked outside, followed by her son, who was carrying a screwdriver. When the officer saw him, he began yelling commands to drop it and within seconds opened fire, killing Harrison.6
•In December 2014, New York police killed a man with a knife who had stabbed someone in a Jewish religious school and was shouting about killing Jews. The video shows local congregants trying to calm him and pleading with police not to shoot, but police destabilized the situation by yelling commands and pointing weapons.7
In each of these cases, officers relied on standard procedure for an armed suspect, which is to yell commands and prepare to use deadly force—even though most of them had received training in how to deescalate confrontations with PMI.
In the United Kingdom and other places where police are less likely to be armed, this dynamic is less common. Police use less lethal means to manage them. Three recent cases reflect this.
•In September 2014, Nicholas Salvador, who had paranoid schizophrenia, beheaded a neighbor and went on a rampage in his London neighborhood. Local unarmed police encountered the suspect and rescued nearby children while engaging him verbally. Eventually, armed police arrived and used Taser shocks to subdue him.8
•In August 2014, a knife-wielding man outside Buckingham Palace was Tasered by police rather than shot.9
•In 2011, a man with a machete was captured after a seven-minute confrontation with up to thirty police officers in South London. Officers used trash cans, batons, and eventually riot shields to contain him and finally overwhelm and tackle him.10
In each of these cases, police put themselves at risk to try to resolve the situation without deadly force, even though they might have been legally justified in using it. In the United States, it seems likely that any if not all of these incidents would have resulted in the person’s death.
Another form of this dynamic is “suicide by cop,” in which someone who is suicidal counts on the willingness of an armed police officer to respond to a threat with deadly force. In these tragic cases, the suicidal individuals arm themselves with toy guns or other harmless devices in hopes that they will be sufficient to provoke a deadly response by police, who too often quickly oblige them. In some ways this seems like an unavoidable problem. There are, however, some important caveats. This whole scenario rests on the suicidal person’s assumption that they will be confronted by an armed police officer. The dynamic might be very different if the responder instead was an experienced civilian mental health worker, or even an unarmed police officer. Suicide by cop is extremely rare in the United Kingdom, where police are unlikely to be armed.11
This is not to say that mental health policing in Britain is without problems. The National Health Service offers substantial options for people in crisis or with chronic mental health needs. Police are instructed to take someone with a mental health crisis to a “place of safety,” which could be a hospital, community-based care provider, or, as a last resort, a police station. The UK police rely on a Mental Health Liaison Officer (MHLO) system, in which a few officers receive extensive training and are supposed to respond to difficult calls and smooth bureaucratic processes between service providers and police. In addition, mental health nurse practitioners are stationed in police dispatch rooms to give responding officers patient histories and real-time advice. They are also expanding the number of street triage teams in which a nurse rides along with the responding officer. The overall attitude is one of care rather than threat neutralization. In practice, however, problems remain. After several high-profile deaths and other mishandled incidents, a national commission found in 2013 that training was inadequate, MHLOs were not well supported by the police services, health services in police stations were inadequate, too much force was used to restrain PMI, and there was not always a good working relationship between police, hospitals, and community mental health workers.12
Studies suggest that anywhere from 5 to 20 percent of all US police incidents involve a PMI, and that these incidents take longer to resolve and are more likely to result in arrest.13 In addition, the number of incarcerated PMI has grown dramatically. The National Alliance on Mental Illness (NAMI) found that 2 million people a year are admitted to US jails; of them, 15 percent of men and 30 percent of women have a serious mental illness.14 The largest inpatient psychiatric facilities in the United States are the LA County Jail, New York’s Rikers Island Jail, and Chicago’s Cook County Jail; the PMI in jails and prisons outnumber those in state hospitals ten to one.15 The number-two cause of death in jails and prisons is suicide; jails, which generally receive people straight from police custody, provide only limited screening and inconsistent mental health care.16 NAMI estimates that 83 percent of PMI in jail don’t have access to the treatments they need.17 People are often given medication while in jail and, at best, a bottle of pills and a referral when they are released, leading to a revolving door of arrests and short-term incarceration with no real improvement in the person’s underlying mental health, which is often at the root of the behaviors that get them arrested in the first place.
What we are witnessing is, in essence, the criminalization of mental illness, with police on the front lines of this process. This is especially true for those who are homeless and/or lack access to quality mental health services. Both groups of people have grown significantly in recent decades. While the Affordable Care Act holds the promise of some improvement, as recently as 2011, over 60 percent of people experiencing a mental health problem reported that they had no access to mental health services.18 Even when mental health services are available, they are often inadequate. A lack of stable housing and income exacerbates mental health problems, makes treatment more difficult, and contributes to the public display of disability-related behaviors, all of which make it more likely that the police will be called.
Reducing social services and replacing them with punitive social control mechanisms works less well and is more expensive. The cost of housing people and providing then with mental health services is actually lower than cycling them through emergency rooms, homeless shelters, and jails, as numerous studies have shown.19 The drive to criminalize has more to do with ideology than effectiveness: the mentally ill are seen not as victims of the neoliberal restructuring of public health services but as a dangerous source of disorder to be controlled through intensive and aggressive policing. Any attempt to reduce the negative effects of policing on this population must directly challenge this ideological approach to policing.
Reforms
Training
Efforts to increase and improve officer training attempt several things. First, training details the signs of serious suicidal thinking and actions and offers strategies for stabilizing people so that they can be taken into custody. Second, it provides information about available services such as community-based or outpatient clinics and ways of accessing emergency acute care, including temporary commitment at an emergency room. Officers are also taught about the nature of different mental illnesses and strategies for dealing with a crisis without traditional use of force.
There are severe drawbacks to this approach. First, it is not reasonable to expect a patrol officer to make a meaningful clinical assessment of patients in the field. While experience may help some officers identify certain more common behaviors, a nuanced assessment just isn’t likely, and this could have significant consequences for how the officer approaches the interaction. While some people might respond well to limit-setting language, others might find this threatening and become aggressive, especially when it is attempted by an inexperienced practitioner.
Second, there are few services available in most places, especially for people who are not in severe crisis. A huge amount of police interactions are with PMI they encounter somewhat regularly, often in public places, who are more a nuisance than an actual threat to public safety. Emergency rooms are not appropriate and will generally not accept people in this condition. Telling them about available services—or lack thereof—often just communicates that officers are on their own and must instead rely on either informal resolutions or arrests.
Finally, as mentioned in chapter 1, standard police training instills a warrior mentality. Police are trained to see the potential threat in any encounter and to use their presence, body language, and verbal commands to take charge and to react quickly and aggressively to any threat of violence or the presence of a weapon. This goes directly against best practices for dealing with most PMI. Studies show that standard police approaches actually tend to escalate and destabilize encounters. Yelling commands and displaying weapons may cause a mentally ill person to flee or become more aggressive. Just as problematic, someone having delusions or a psychotic episode may be unable to hear, understand, or comply with police orders. This can have tragic consequences.
More recently, some departments have adopted training that emphasizes communication, containment, and coordination with appropriate service providers as an alternative to the command-and-control approach. While this new training has some advantages for de-escalation, it can still lead to tragic results. Officers in New York were using this exact policy when they confronted an Orthodox Jewish man in his apartment after receiving a call. The man had a small decorative hammer used in religious ceremonies. When the man tried to leave his basement apartment with the hammer, officers tried to surround him, in keeping with their training on containment. However, when the man tried to evade containment, they shot and killed him.20 More recently, police in San Francisco used similar tactics in trying to apprehend a man with a knife who had stabbed someone nearby. Officers cornered and surrounded the assailant, demanding that he drop the knife, and fired two beanbag rounds at him, but he continued to hold onto the knife and attempt to leave. Officers then fired fifteen rounds at him, killing him.21 Containment and less lethal weaponry can still lead to deadly encounters.
It is not reasonable to expect that officers who spend the bulk of their time using aggressive methods to establish their authority can just turn that off in a situation where someone might be mentally ill and appears to be a threat to the officer or others. This is why so many encounters with PMI holding weapons end up escalating, even when the officers involved have received mental health training.
Crisis Intervention Teams
The “Memphis Model” relies on a small number of specialized officers who can be routed to calls to deal with a person experiencing a mental health crisis.22 These officers become more knowledgeable and experienced and are better able to assess the situation accurately and take clinically appropriate steps to reduce the chance of escalation. This model has shown signs of success in cities that have embraced it, but only when there are meaningful mental health care services available for police to rely on. The problem is that these services often don’t exist; in addition, it is still a police-centered model with a strong tendency to resolve situations through arrest and other uses of force.
Some places have tried to mitigate this tendency by creating crisis response teams that include trained mental health workers. This approach is common in places like Canada, Britain, Europe, and Australia. Specially trained officers work with mental health professionals to respond to calls involving PMI. In many cases it is the civilian mental health workers who take the lead, with police there only to assist if absolutely necessary. These teams have shown good results in both reducing arrests and the use of force and in reducing hospitalizations as well, since they can make a more complete assessment and take steps to stabilize the person and connect them to appropriate outpatient services.
Outreach Teams
In some places, local officials face chronic problems from mentally ill people in public spaces. Some are homeless; others live in marginal housing or are unemployed and unengaged and spend much of their time on the streets. This population may at times experience acute mental health crises, but they are much more likely to come to the attention of police as a source of disorder, which may take the form of “quality of life” violations like public drinking and urination, disorderly conduct, or sleeping in parks, subways, or sidewalks. In some jurisdictions, officials have attempted to address this problem by developing police outreach teams. Some are designated as focused on homeless people while others deal more specifically with the mentally ill, but the functions overlap.
But why should armed police officers oversee outreach to the chronic and homeless mentally ill? Using armed police is expensive and brings few benefits. Trained mental health and social services outreach workers are perfectly capable of handling this job and, unlike police-based teams, are more likely to be able to build long-term relationships and gain trust, an essential component of outreach to highly isolated individuals with complex mental health and often substance abuse problems. The implied threat of coercive response that police pose drives such people further into isolation, not into proper care. Civilian teams are also cheaper.
Diversion Programs
There have also been efforts to divert PMI from incarceration. Police-based models such as the Law Enforcement Assisted Diversion program (LEAD) in Seattle, allow officers to identify people who are chronically involved in low-level criminality and disorder and place them in programs that try to address their underlying problem, whether it’s a mental health or substance abuse issue or poverty driving them into black-market activities like sex work and drug sales.23 These programs have reduced arrest and incarceration rates; they offer some new services to people in need and some relief for communities. But why do the police need to be the gatekeepers? Framing this as a policing issue bases access to needed services on how much the officer is motivated to resolve a public-order problem. A person muttering to themselves in disheveled and smelly clothing in a high-profile shopping district is more likely to gain the sustained attention of police than a suicidal, homeless teen hiding out under a bridge. Both need services, but police are much less likely to encounter the teen and less likely to treat that encounter as being driven by mental health issues. Mental health outreach workers are likely to see the suicidal teen as more acutely at risk and take steps to stabilize them.
Another important development has been the emergence of a wide array of mental health courts. The purpose of these specialized courts is to divert PMI from jail by connecting them with appropriate services, combined with oversight and the threat of possible incarceration for failure to comply with program goals and court directives.24 Judges tend to take an active role in monitoring and rewarding progress; for some defendants, this represents a rare and important pathway to stability. These courts are not much cheaper to operate than regular misdemeanor criminal courts, but they reduce the number of people being sent to jail, which is tremendously expensive: because of high turnover, jails are much more expensive to operate than prisons, with per-bed costs reaching as high as $200,000 a year or more.25
These courts, however, rely on the constant threat of punitive sanctions. People who fail to follow through with case management plans can always be sent to jail, since a guilty plea is often a condition of receiving treatment. Also, they can only access the services the court provides if they have been arrested, meaning that many people in need of services remain unable to obtain them. As with the LEAD program, the focus is on abating nuisances and saving money rather than developing a rational system for delivering necessary mental health care.
Alternatives
We can never fully eliminate interactions between the police and PMI. There is indeed a need for more training of all officers, and even the participation of officers in some crisis-response scenarios. The situation we have today, however, represents a gross criminalization of mental illness. This system does not require that individual police officers be biased against PMI or regularly misuse their discretion—which studies show they usually do not.26 It only requires that we have a fundamentally flawed mental health system that fails to provide adequate care to people—which we do. This means responsibility for dealing with people in crisis invariably falls on the police, whether they like it or not. Yes, crisis response teams, specialized courts, and improved training can reduce the impact of the criminal justice system on the mentally ill and the impact of the mentally ill on the criminal justice system, but these are not replacements for a rational, functioning mental health system.
Thoughtful police officers and leaders are well aware of this. Many view interactions with PMI as one of the least desirable and most fraught aspects of the job. Many are deeply frustrated by the revolving door of emergency room visits, jails, and police lockups, which never seem to solve the problem. Too often police are forced to arrest someone because a hospital, clinic, or other program is either unavailable or won’t or can’t accept them. Police officials are starting to speak up as well, like former Chief Michael Biasotti from New Windsor, New York. As chair of the New York State Association of Chiefs of Police, he backed measures to increase funding for mental health services, pointing out the irrationality of housing 350,000 PMI in prisons and jails. He notes that a real diversion program
would be expanding services to the seriously mentally ill, and getting treatment before the police are at your door, before you are standing before a judge, and before you find yourself in jail … Increased services mean less involvement with the criminal justice system and improved quality of life for those with mental illness and their families.27
Mike Koval, chief of the Madison, Wisconsin, police force, has spent years advocating for community-based mental health services in the wake of police killings of PMI. He realizes that even with enhanced training and specialized response, there are still limits to what the police can do: “The unique challenges presented in these calls are going to result in more tragic outcomes unless or until there is a commitment to provide more proactive, pre-emptive, and collaborative interventions BEFORE an individual’s mental health issues have declined to critical levels.”28 He even got permission from the city of Madison to undertake litigation against the state for closing down a mental health clinic, arguing that the loss of its services diverts considerable police resources and money away from patrolling, as officers must now transport people longer distances.
According to the Florida Mental Health Institute, chronically mentally ill people are a major source of spending for the criminal justice system. Its study identified ninety-seven “chronic offenders” who, over five years, accounted for 2,200 arrests, 27,000 days in jail, and 13,000 days in crisis units, state hospitals, and emergency rooms. The costs to taxpayers for these people alone was nearly $13 million, or $275,000 per year per mentally ill person. In Miami-Dade jails, some 1,400 inmates take psychiatric drugs, making the corrections system the largest warehouse for PMI in Florida. Mental health care there costs taxpayers $80 million per year.29 The Vera Institute of Justice found that incarcerating PMI costs two to three times what community-based treatment does.30
Instead of just funneling ever-increasing amounts of money into specialized police units and enhanced mental health services in jails and prisons, we need a major overhaul of our mental health systems. Billions of dollars have been cut from public mental health services in recent decades, as states have closed down expensive and poorly run hospitals but failed to fund community-based care. Instead of relying on forced treatment, we should be providing easy access to varied, culturally appropriate community-based services as needed. Even people with severe disabilities can live independently and with a limited impact on the community with long-term supportive care in a stable living situation. Some places are trying to move in this direction. Miami officials are working to turn a shuttered hospital into a rehabilitation hub for people with serious mental illnesses. The facility would provide safe drop-in spaces, treatment facilities, and access to short-term housing.31 While this is a step in the right direction, it still doesn’t provide long-term stable housing with medically appropriate support services. Part of the facility will also be used to house a mental health court—resources that could be better spent on housing and medical services.
Special attention is needed for services for those with severe problems such as schizophrenia, which, when untreated, can result in significant antisocial and even potentially dangerous behavior. Giving people medication and sending them to a homeless shelter or welfare hotel is not adequate. Without stability and support, patients are more likely to stop taking their medication. A safe, supportive housing environment is more likely to produce stability than incarceration or forced pharmacological treatment. For those who are currently homeless and off their medication, we need civilian outreach teams and access to safe drop-in spaces.
Finally, when people do experience a major mental health crisis, we should always attempt to approach that situation in the least confrontational way possible. Trained civilian responders should be the default preference. They pose the least threat to the PMI and are the least likely to escalate the interaction. Yes, these interactions can be dangerous, but people trained and experienced in dealing with PMI know these risks and have techniques for dealing with them. Even in state mental health hospitals that contain people with a history of violence, staff are generally able to manage patients with a minimum of violence. Force is used and is even sometimes excessive, but a well-trained team is much less likely to cause a death than an armed police officer.