Despite what you see on television, a verdict of “not guilty by reason of insanity” is exceedingly rare. Most defendants with mental illnesses end up incarcerated—studies reveal that fully half of all prisoners have at least one mental disorder. That is one million people in the U.S. alone, and the prison system does very little to successfully treat them. As a result, the recidivism rate among released convicts is especially high for those with serious disorders.

Forensic hospitals, on the other hand, which hold and treat offenders found not guilty by reason of insanity, have a very high success rate in preventing disordered individuals from returning to crime. In an analysis of data from California, New York and Oregon, Victoria Harris, a forensic psychiatrist at the University of Washington, reported in 2000 that people at these institutions reoffended at a “much lower” rate than untreated mentally ill offenders. Psychiatrist Jeremy Coid and his colleagues at St. Bartholomew's Hospital in London found in 2007 that forensic patients in the U.K. were 60 percent less likely to reoffend than released inmates and 80 percent less likely to turn to violence.

These and other recent studies show that treatment works, and yet we continue to put offenders with mental disorders in prisons for complex reasons, including our society's views toward mental illness—especially addiction—and the high cost of psychiatric care for inmates. Still, solutions are within reach. A Connecticut program, for example, allows some veterans who have committed crimes to seek psychiatric treatment instead of serving time. Public knowledge of and support for such programs are essential to breaking the cycle of crime that the current prison system perpetuates.

Prisons as Mental Institutions
The prison system functions in substantial part as the successor to our shuttered mental institutions. In 2009 epidemiologist Jacques Baillargeon of the University of Texas Medical Branch at Galveston and his colleagues rightly described this situation as a “national public health crisis” and found that it arose from a baleful synergism of developments. First, the invention of antipsychotic medication in the 1960s led to a movement to close the many psychiatric hospitals then extant. These closed institutions were supposed to be replaced by community facilities, but in reality most were not. At the same time, health insurers restricted coverage for mental health treatment, and finally, the “war on drugs,” begun in the 1980s, increased drug-related arrests and brought in mandatory and fixed sentences. More offenders with psychiatric and substance abuse problems, often one and the same, were incarcerated for many years without treatment and then released into a community that had nothing for them: no jobs, no treatment, no housing. This led to drifting, homelessness, further mental decline and the chronic reoffending we see today.

In 2012 Jason Schnittker, a sociologist at the University of Pennsylvania, and his colleagues reported that legislatures have criminalized “many common psychiatric disorders,” especially substance abuse—which psychiatry's diagnostic handbook, the DSM-5, categorizes as a true psychiatric disorder. This criminalization of drug addiction means, as Schnittker puts it, that “some inmates end up in prison at least partially because of their psychiatric disorders.”

Most of these mentally ill inmates are not treated for their conditions in prison. And their numbers are rising. Schnittker reports that for the past 40 years, the rate of incarceration has quintupled, from 149 per 100,000 in 1980 to 749 per 100,000 in 2009. As people go in, other people come out. Nationally, 700,000 inmates are released every year, which means, according to the National Institutes of Health, that more than 350,000 disordered offenders return untreated to society. In most accountings, most of these people will reoffend.

Clearly, this system does not work. Strikingly, though, it runs in parallel with a system that does work, namely the system of forensic hospitals, which is where defendants end up who are found not guilty by reason of insanity. This outcome is difficult to achieve: the legal defense must demonstrate that the offender had no control over his or her actions or did not comprehend the present reality of the deed done (for instance, shot at a police officer because he thought the officer had been threatening him on television). A small fraction of 1 percent of all criminal defendants are acquitted by reason of insanity.

Forensic hospitals confine people as prisons do but achieve radically different results. Both function by way of the criminal justice system, but prisons cause disordered offenders to break the law more—even more than offenders without a disorder—whereas forensic hospitals treat offenders as patients who can and do recover and who return to society as people who can be expected, for the most part, to be law-abiding citizens.

Cured of Criminality
The radical difference in outcomes from these two systems is illustrated by the experience in Connecticut, where I practice law. In its 2010 Annual Recidivism Report, Connecticut's Criminal Justice Policy and Planning Division analyzed 16,241 inmates released during 2005. Of these, 1,514 were classified as severely disordered, including people who had “chronic schizophrenia or bipolar disorders with frequent psychotic exacerbations, who need medication and assistance with activities of daily living, [as well as persons] with borderline personality disorder with frequent suicidal gestures or episodes of self-mutilation.” Others were even worse off, suffering from “acute psychosis, severe depression, suicidal ideation … and overwhelming anxiety.”

Although the rate of recidivism for the overall cohort of 16,241 inmates was high—67.5 percent within three years—the rate for those with severe disorders was even greater. The department did not indicate by how much, only that it was “significantly” higher.

Given the mental state of these former inmates—people who were not found insane during their trials—patients found not guilty by reason of insanity might seem to be especially disordered, and so they can be. Yet after their release, having been confined in mental hospitals, not prisons, not only are they less likely to reoffend than disordered inmates, but they are even less recidivist than offenders without a recognized mental illness. Indeed, in Connecticut they return to crime so seldom that the department of correction does not have a category for them in its annual recidivism reports. The agency with jurisdiction over acquittees, the Psychiatric Security Review Board, also does not publish data on persons discharged. In response to my inquiry, however, it reported that between 1985 and 2013, four acquittees were arrested while on temporary leave, and one acquittee was arrested while on conditional release.

Forensic hospitals provide one effective alternative to incarceration for disordered offenders. Mental health courts offer another. Dale E. McNeil, a clinical psychologist at the Langley Porter Psychiatric Institute, part of the University of California, San Francisco, reported in 2007 that 34 states have such courts and that they are effective in reducing recidivism and violent reoffending. Typically these courts provide a separate docket for defendants with disorders, with designated judges and counsel, and they offer defendants the option of entering a nonadversarial process in which they follow a treatment plan in return for reduced sanctions. McNeil's study followed 170 people chosen out of 8,325 defendants with mental disorders for a median of 8.3 months, some of whom went through the mental health courts and some of whom went through traditional legal proceedings. His results showed that mental health court participants went longer without reoffending than those who did not participate. Further, the risk of violent offense was halved. Although the study involved subjects not chosen at random and the defendant pool was limited to San Francisco, McNeil's results align with a consistent trend: for criminals with disorders, treatment works, and it works especially well in reducing the rate of violent offenses.

Should Society Bear the Costs?
We know treatment works. Yet the barriers to treating more mentally ill offenders are huge. For one, treatment is not cheap. In Connecticut, the average annual cost per prison inmate is about $33,000. The average annual cost per acquittee in a forensic hospital approaches $500,000. Granted, psychiatrists cost more than prison guards, but I doubt the difference can explain the additional $467,000; there seems to be more than a little redundancy in the system. Outpatient treatment for insanity acquittees is cheaper, although I could find no specific data on how much less. One advantage for states in classifying an offender as a patient is that he or she becomes eligible for Medicaid reimbursement, which means that the federal government covers half the cost of confinement and treatment. Still, that is a cost ultimately borne by society.

The cost to society of treating mentally ill criminals is hard for some people to swallow. If offenders, disordered or not, are morally responsible for their offenses, why not just keep them in prison? It is so much cheaper.

I believe the answer to this dilemma becomes apparent if one considers the etiology of mental disorder, which suggests that circumstances and experiences, rather than innate character flaws, give rise to symptoms. The first DSM was published in 1952, in large part to organize a universe of disorders created by war. More than a million soldiers in World War II suffered enough mental symptoms to be deemed unfit for combat. At the time, doctors had no system of describing most of those symptoms.

These symptoms and disorders were caused by conditions and circumstances, not moral defects; the more combat a soldier saw, the more likely he was to suffer symptoms. We can conclude that those who develop disorders, then, whether by circumstances or genetics, are not responsible for their dysfunction. This view is consistent with Schnittker's finding that “childhood adversities have been linked to adult psychiatric disorders” and that such hardships have been linked to subsequent criminal behavior: “childhood disadvantage is associated with both incarceration and adult psychiatric disorders.” Treating mentally ill offenders, then, is not just the right thing to do to reduce recidivism, but it is the right thing to do for people whose lot is not of their making.

Early Solutions
Given the difficulty of successfully pleading not guilty by reason of insanity—as well as the inherent problems in mixing disordered inmates, whose disorders tend to make them vulnerable, with nondisordered inmates, who are often predators—early attempts at solving this problem have been workarounds. Mental health courts, which are designed to funnel mentally ill offenders who cannot make a legal case for insanity away from prisons and toward treatment, fall into this category. One successful program in Connecticut intervenes on behalf of veterans when they find themselves in criminal court, offering reduced or eliminated sentences in return for closely monitored psychiatric treatment. This policy of intervention speaks to a growing recognition that society is better served by treating mentally ill offenders than by incarcerating them.

We are learning not to hold veterans responsible for how they react to the cauldrons they are thrust into. War, however, is not the only aspect of modern life that can produce disorder. Schnittker notes that in 2006, 7.5 percent of the adult population, or 16 million people, were inmates or ex-inmates, a number that approximates the number of unemployed during the recession of 2008–2009. Indeed, the evisceration of the nation's manufacturing base has coincided closely with the rise in the number of people incarcerated, especially men. Work used to be how many of us redeemed ourselves from early disadvantages. We might hope the reigning powers will, as they did in another era, come to recognize the need people have for productive enterprise and the vital part work has in our individual and national psyches.

Some may debate the connection between disorder and usefulness, but it is beyond contention that the rise in the number of persons incarcerated has been yoked to the rise of disordered persons incarcerated. People who commit crimes and who are treated, especially in institutional settings, reoffend at rates lower than untreated convicts and offenders without disorders—revealing that for most mentally ill offenders, the issue is not moral but psychological. We should recognize that unfortunate conditions produce unfortunate effects. We do not criminalize people who get sick from polluted waters nor those wounded and maimed in wars. We do not call them defective and make of them a pariah class. We treat them, and so should we do with the mentally disordered, however they exhibit that affliction.