In the mid-1990s, Florida’s Broward County had a problem. With a paucity of available mental health programs, police often had no alternative to arrest for people who were behaving bizarrely on the streets. Growing numbers of defendants with serious mental illnesses were appearing in criminal court, charged with crimes such as trespassing that often seemed driven by their disorders. After a short jail stay, usually without treatment, their charges would be dismissed or sentences reduced to time served by judges who knew that their reappearance in court was almost inevitable.

Judges around the country, usually with no background in mental health, faced dockets clogged with similar cases. Epidemiologic surveys of jails and prisons found that 14 to 24 percent of inmates suffered from serious mental illnesses, conditions more often neglected than treated during confinement. With little public interest in funding adequate services, the problem of how to deal with persons with mental disorders was being left to the already overburdened criminal justice system. Broward County, though, decided to do something to try to tackle the problem: it established the first mental health court (MHC) in the United States.

Drawing on the earlier model of drug courts, aimed at treating rather than punishing addicts, MHCs were designed to divert defendants with serious mental illnesses from the criminal justice system. Models differ across jurisdictions: some MHCs intervene before trial, others only after a finding of guilt; misdemeanants are the exclusive focus of the Broward court, but defendants accused of felonies are also eligible in some places; most MHCs don’t control their own treatment resources, but a small number—like the Miami-Dade program in South Florida—do. Violent or sexual offenders are often excluded from participation, and some jurisdictions impose other limitations. What they have in common is the ability to offer defendants the opportunity to avoid jail or prison time if they agree to follow a court-approved treatment plan. MHC participation is always voluntary, but participants who complete the program will have charges dropped or sentences suspended. However, failure to abide by the terms of the program can result in a return to the criminal justice system and the prospect of doing time.

MHCs are often paired with other reforms aimed at keeping the mentally ill out of jail. In Allegheny County, Pennsylvania, where Pittsburgh is located, hundreds of police officers have gone through a forty-hour “crisis intervention team” training course on how to manage individuals with mental illnesses. These cops wear small blue “CIT” badges that signal to street people with mental illnesses that they are dealing with officers who understand their conditions and aren’t interested in putting them behind bars. As an alternative to jail, police can escort those who commit minor crimes or disturbances—such as a man who repeatedly tried to direct traffic on a busy main thoroughfare, according to the Pittsburgh Post-Gazette—to special units at local hospitals.

MHCs have become extremely popular, perhaps not surprisingly given the ubiquity of the problem they are meant to address. At last count, there were 357 adult MHCs in the U.S. and another sixty tribunals dealing with juveniles. The MHC model has been adopted in other countries, including Canada and Australia, and has spawned a similar approach for veterans of the armed forces, who often have a combination of mental health and substance abuse problems. Several of the bipartisan legislative proposals stutter-stepping their way through Congress this year would encourage additional jurisdictions to create MHC programs. But as with so many public policy initiatives, enthusiasm for MHCs has often outstripped the evidence for their effectiveness and advocates tend to ignore the downsides of the programs.

What do we know about how well MHCs work? After some early equivocal findings, most recent studies have confirmed that MHC participation is associated with reduced rates of rearrest and reincarceration compared with ordinary handling by the courts and correctional system. A newly published study of the District of Columbia MHC, for example, found that 25 percent of misdemeanor defendants with serious mental illnesses who “graduated” from the program were rearrested within two years, compared with 48 percent of defendants who were eligible for the program but didn’t enter it. (Interestingly, MHC dropouts had the highest recidivism rates of all, at 55 percent.) Similarly positive results come from Ramsey County, Minnesota, in which St. Paul is located: misdemeanants who went through the usual criminal process were two and a half times more likely to be convicted in their first year in the community and served almost five times as many days in jail in that period compared with MHC completers.

Although it is often assumed that reducing rearrest and incarceration will result in a net savings, data on costs associated with MHCs has been inconsistent. A Rand Corporation evaluation of the MHC in Allegheny County, Pennsylvania, suggested that savings begin to accrue after the first eighteen months of participation, largely accounted for by reduced jail time. However, an analysis of four MHCs in different parts of the country funded by the John D. and Catherine T. MacArthur Foundation showed that MHC involvement led to increased annual costs of $4,000 per participant compared with a matched group of non-MHC jail detainees, as the costs of enhanced mental health treatment consistently exceeded the savings from fewer days of incarceration. When it comes to dealing with offenders with serious mental illnesses, MHCs are not offering a free lunch.

What has been harder for researchers to pin down is why MHCs seem to work. Advocates for MHCs, perhaps not unreasonably, assumed that many of the charges faced by defendants eligible for the programs stemmed from behaviors related to the illnesses themselves. Perhaps the euphoria and grandiosity associated with a manic state might lead to the kind of public rowdiness that could result in a charge of disorderly conduct. Or the delusional thinking seen in acute psychosis could cause a homeless person to believe that he owned all the retail stores in the city and hence to walk out of a convenience store without paying for a candy bar. Treat the symptoms of the underlying illnesses, the theory went, and the criminal behavior would go away. Although that commonsense proposition has been remarkably difficult to substantiate, a recent reanalysis of data from the MacArthur Foundation study has confirmed that increased medication compliance and use of mental health services were linked to significantly lower rates of arrest. Nonetheless, it’s still possible that more intensive supervision or other nonspecific factors play a role in reducing recidivism too.

Family members desperate to obtain treatment for a loved one may see no alternative but to have the person arrested, hoping detention will lead to diversion to a mental health court.

Often neglected in the rush to embrace the MHC model are the limitations and potential negative effects of the approach. Since MHCs rarely control their own mental health services, the treatment plans they can impose are necessarily limited by what is available in a given area. Thus, participants who need a service that no local provider offers are simply out of luck. Moreover, given that community mental health services are “maxed out” in many locales, often with waiting lists of weeks or months in duration, creation of an MHC may simply alter who gets access to a scarce resource. Even if a judge’s order jumps an MHC participant to the head of the queue, it will only be at the cost of another person with serious mental illness who has never broken the law but now will have to wait even longer for help. In short, MHCs are not an easy solution to the problem of an underfunded mental health system.

Embedding access to mental health treatment in the criminal justice system can have other perverse effects as well. Family members desperate to obtain treatment for a loved one may see no alternative but to have the person arrested, hoping detention will lead to diversion to an MHC. In addition to the risk associated with an encounter with the police—nearly a quarter of victims of police killings each year have an identifiable mental illness—an arrest record can lead to subsequent difficulty finding housing, getting a job, and accessing community services, all consequences of endowing a person with the dual stigmata of mental illness and a criminal record. Yet the temptation to gain access to treatment by characterizing illness-related behavior as criminal is real. The ironic result may be an increased flow of people with serious mental illnesses into the criminal justice system, exactly the opposite of the original goal.

Finally, we need to acknowledge the paradox of the “success” of MHCs: if adequate community-based services and hospital beds were available for the treatment of serious mental illnesses, it is unlikely that Broward County and the hundreds of jurisdictions that have followed its lead would have developed MHCs in the first place. To be sure, even with an optimal system of mental health care some people with mental illnesses would end up facing charges and could benefit from MHCs and other diversion programs. But as we listen to politicians, bureaucrats, and others extol the MHC model, we would do well to keep in mind that it is only a workaround. Ultimately, our badly funded and poorly organized mental health system itself needs to be addressed in a more thoughtful and systematic way.

Our ideas can save democracy... But we need your help! Donate Now!

Paul S. Appelbaum

Paul S. Appelbaum, MD, is the Elizabeth K. Dollard Professor of Psychiatry, Medicine and Law at Columbia University.