Persons With Severe Mental Illness in Jails and Prisons: A Review

Mental health professionals have become increasingly concerned about the number of persons with mental illness in jails and prisons. This issue is a relatively recent one. Reports of large numbers of mentally ill persons in American jails and prisons began appearing in the 1970s (1–3). This phenomenon had not been reported since the 19th century (4).

To better understand this problem, a literature review was conducted. Two of the primary questions addressed were whether large numbers of persons with severe mental illness who commit legal transgressions are being taken to jails and sent to prisons instead of to hospitals or other psychiatric treatment facilities, and whether the number has increased since deinstitutionalization.

The review also examined other aspects of this issue, including the characteristics of mentally ill offenders, factors cited as causes of mentally ill persons' being placed in the criminal justice system, the relationship between mental illness and violence, access to treatment for this population, the role of the police, and society's attitudes toward mentally ill offenders. Finally, recommendations are made about how inappropriate placement of this population in the criminal justice system can be prevented and how to treat mentally ill offenders both in the system and after they are released into the community.

Methods

MEDLINE, Psychological Abstracts, and the Index to Legal Periodicals and Books were searched from 1970, and all relevant references were obtained.

Results and discussion

Incarceration versus hospitalization

Many factors come into play in determining why a person with mental illness is arrested rather than taken to a hospital. Generally, persons who are thought to have committed a felony are arrested and brought to jail regardless of their mental condition. The criminal justice system, charged by society with the responsibility for removing from the community persons accused of committing serious crimes, sees no alternative but to first place the person in custody in a secure setting and then arrange for psychiatric treatment if necessary (5). If the person is thought to have committed a serious crime, the police and the criminal justice system generally do not want to leave this person in a psychiatric hospital where security may be lax, the offense may be seen by staff as secondary to the patient's illness, and the person may be released to the community in a relatively short time.

For persons charged with misdemeanors, the situation becomes more complex. In 1972 Abramson (6) was the first to coin the term “criminalization of the mentally ill”; he observed that persons with mental disorders who engaged in minor crimes were increasingly subject to arrest and prosecution in a county jail system. Subsequently, many authors applied the concept of criminalization to persons with mental disorders who were arrested for serious crimes.

The distinction between arrest and incarceration of mentally ill persons who have committed minor offenses and those who have committed serious offenses is an important one. As Steury (7) notes, no consensus exists on the definition of criminalization of persons with mental disorders. Some researchers define criminalization at the point of arrest (8–11) and others require prosecution (12–15), while others use incarceration in jails and prisons (16,17).

In our opinion, the term criminalization should be used primarily in connection with mentally ill persons who are arrested, with or without jail detention, and prosecuted for minor offenses instead of being placed in the mental health system. As noted, it is clear that persons who have committed serious offenses, no matter how mentally ill, would normally be processed in the criminal justice system (13,18,19). However, it should be acknowledged that many mentally ill persons who commit serious crimes and enter the criminal justice system might not have engaged in such behavior if they had been receiving adequate and appropriate mental health treatment (20).

In 1939 Penrose (21) advanced the thesis that a relatively stable number of persons are confined in any industrial society. Using prison and mental hospital census data from 18 European countries, Penrose found an inverse relationship between prison and mental hospital populations. He theorized that if one of these forms of confinement is reduced, the other will increase. According to this theory, where prison populations are extensive, mental hospital populations will be small, and vice versa. Thus if there is room in prisons and a shortage of hospital beds, many mentally ill persons who come to the attention of law enforcement might well be directed to the criminal justice system. Another corollary of this theory is that if civil commitment is reduced, involvement with the criminal courts will increase (22).

Proportion of incarcerated persons with mental illness

The Bolton Study (23) in 1976 was one of the first extensive and methodologically sound attempts to determine the percentage of county jail inmates with mental illness. In a five-county combined sample of 1,084 adults in California county jails, 6.7 percent were psychotic and 9.3 percent were judged to have nonpsychotic mental disorders, not including personality disorders. For Los Angeles County, the figures were 7.8 percent psychotic and 5.7 percent nonpsychotic.

In a more recent systematic study, Teplin and her coworkers (24) interviewed 728 randomly selected male admissions to the Cook County jail in Chicago. Using a structured psychiatric interview, they found that 6.4 percent met diagnostic criteria for schizophrenia, mania, or major depression. In a second study of women entering a county jail in Chicago, Teplin and her colleagues (25) found that 15 percent had severe psychiatric disorders within the previous six months, 1.8 percent had schizophrenia or a schizophreniform disorder, 2.2 percent were manic, and 13.7 percent had major depression. Guy and associates (26) interviewed 96 randomly selected admissions to the Philadelphia city jail and found that 14.6 percent had schizophrenia or manic-depressive illness.

With regard to state prisons, in a 1987 Michigan study of 1,070 state prison inmates carefully selected through a stratified random sampling procedure, 2.8 percent were found to have schizophrenia, 5.1 percent to have major depression, and 3.8 percent to have bipolar disorder or mania (27). Jemelka and associates (28) used the Diagnostic Interview Schedule with 109 inmates in the state of Washington and found prevalence rates of 4.4 percent for schizophrenia, 10 percent for major depression, and 3.7 percent for mania. Similar rates were found in California and Ohio prisons (29). Steadman and coworkers (30) studied a random sample of 3,332 inmates representing 9.4 percent of New York's general prison population, as well as 352 of the 360 inmates in the prisons' mental health units. They found that 8 percent of the sample had severe psychiatric functional disabilities that clearly warranted some type of mental health intervention, and another 16 percent had significant mental disabilities that required periodic services (specific diagnoses were not given).

Generally, clinical studies suggest that 10 to 15 percent of persons in state prisons have severe mental illness (29). It may be that in recent years, correctional staff have become better able to recognize signs of mental disturbance and, as a result, refer more of these individuals to mental health professionals. Thus better recognition may also contribute to the prevalence rate of inmates identified as mentally ill.

The magnitude of the problem can be seen when we multiply the percentages of mentally ill persons in jails and prisons by the number of inmates. For instance, in 1995 there were more than 483,000 persons in jails and more than 1,587,000 persons in state and federal prisons (31). Thus even a small percentage of such large populations represents a very significant number of mentally ill persons in jails and prisons.

The large number of mentally ill individuals in jails and prisons has presented serious problems for correctional staff. Gibbs (32) noted that second to overcrowding the presence of inmates with psychological problems was the most serious concern for correctional personnel.

Description of the population

In a study of 102 male inmates of a county jail randomly selected from those referred by jail staff for psychiatric evaluation, 99 percent had previous psychiatric hospitalizations, and 92 percent had arrest records (75 percent for felonies) (5). Four-fifths exhibited severe and overt psychopathology, and more than three-fourths met criteria for civil commitment. When arrested, more than a third were transients, and only 12 percent were employed. More than half were currently charged with felonies and 39 percent with crimes of violence. Thus this population is characterized by extensive experience with both the criminal justice system and the mental health system; severe, acute, and chronic mental illness; and poor functioning.

The same study also found that of those charged with misdemeanors, more than half had been living on the streets, on the beach, in missions, or in cheap hotels, compared with less than a fourth of those charged with felonies (5). Persons living in such places obviously have a minimum of community supports. The authors speculated that the less serious misdemeanor offense is frequently a way of asking for help. Still another factor may be that many uncared-for mentally ill persons may be arrested for minor criminal acts that are really manifestations of their illness, their lack of treatment, and the lack of structure in their lives. It was also observed that some inmates, even though overtly psychotic, had underlying antisocial personality problems that appeared to play a major role as causative factors in their alleged criminal behavior. Findings were comparable in a similarly selected sample of 101 inmates of a county jail for women (33).

Other studies have shown that a large proportion of mentally ill persons in a jail population were homeless before arrest and incarceration (34,35). For instance, one study in New York City found that homeless mentally ill persons were grossly overrepresented among defendants with mental disorders entering the criminal justice and forensic mental health systems for both violent and nonviolent offenses (35). Forty-three percent of the defendants with mental disorders were homeless at the time of the crime for which they were arrested. The rate of homelessness was 21 times higher in the overall sample of defendants with mental disorders than in the overall population of mentally ill persons in the city. Moreover, homeless defendants were significantly more likely to have been charged with victimizing strangers.

Current trends

It is often asserted that the number of mentally ill persons currently in our criminal justice system is larger than before deinstitutionalization (4,36, 37). This assertion is consistent with Penrose's theory described above. It can be argued that society's tolerance in the community of the deviant behavior of people with mental disorders appears to be limited. This limited tolerance is especially true for those who have direct contact with mentally ill persons, namely, the courts, families, and other citizens. Many believe that if social control through the mental health system is impeded because of constraints such as fewer long-term state hospital beds, community pressure will result in placement of some of these persons in the criminal justice system.

In the 1970s, studies began to appear showing that the arrest rate for former psychiatric hospital patients was higher than that for the general population (38,39). Various attempts were made to account for the higher rate. Steadman and associates (9) concluded from their data that the increase was due almost entirely to the increased number of persons with arrest records being admitted to mental hospitals. They speculated that “persons who formerly would have been caught in the ‘revolving cell door’ are now bouncing back and forth between state hospitals and jails as solutions are sought in mental health treatment for what are usually nuisance behaviors or property offenses.”

A related explanation in the late 1970s was the theory of the “psychiatricization” of criminals (8,37). This theory hypothesized that the increased rate of violent crime after hospital discharge was due to jail and prison overcrowding and that mental hospitals were increasingly admitting individuals formerly dealt with by the criminal justice system. On the other hand, a 1978 study in a California county showed that former hospital patients with no history of arrests when they entered the hospital were arrested roughly three times more often after discharge than the general county population and five times more often for serious violent crimes (40).

Another explanation for the increased arrest rate of former hospital patients is that a more criminal group of mentally ill individuals is now hospitalized as a result of the stricter criteria for civil commitment, which rely heavily on dangerousness (41). Finally, the relationship between mental illness and violence, as discussed below, may be another factor. Despite the arguments offered, sufficient evidence does not exist to settle these issues definitively.

An important question is whether the number of mentally ill persons in jails and prisons has increased since deinstitutionalization. A number of studies over the past several decades have purported to demonstrate an increase, but Teplin (17) perhaps said it best when she wrote,“It is concluded that the research literature, albeit methodologically flawed, offers at least modest support for the contention that the mentally ill are being [increasingly] processed through the criminal justice system.” This evidence is largely clinical and inferential, and it is certainly highly suggestive. However, because of the lack of good studies of mentally ill persons in jails and prisons before deinstitutionalization, findings of research conducted since that time cannot be considered conclusive evidence that the number of mentally ill persons has increased.

Nevertheless, it appears that a greater proportion of mentally ill persons are arrested compared with the general population. One of the better studies suggesting this disproportionate rate was conducted by Teplin (11). Chicago policemen were observed over a 2,200-hour, 14-month period, and 1,382 police-citizen encounters were documented. The presence of psychiatric illness in a suspect was determined at the scene by a system that took into account behavioral symptoms and the environmental context. It was found that 27.9 percent of the suspects without mental disorders and 46.7 percent of the psychiatrically ill suspects were arrested.

Perhaps two of the more persuasive arguments that a higher proportion of persons with severe mental illness can be found in the criminal justice system since deinstitutionalization are the presence of large numbers of such persons now residing in our jails and prisons and the clinical observations of clinicians and researchers. It is the impression of clinicians and researchers that a large proportion of the severely mentally ill persons they see in jails and prisons are similar in almost every way to long-term patients in state hospitals before deinstitutionalization (42). Obviously, lifetime residents of state hospitals had little opportunity to commit crimes and to be arrested.

In a similar vein, it was observed even in the 1970s that more liberty for the traditional psychiatric hospital patient placed in the community, including the ability to refuse treatment, is likely an important factor in explaining the observed increased arrest rate and violence (39,43). As discussed below, it is generally the untreated mentally ill person who is more violent, particularly if substance abuse is involved.

Mental illness and violence

Until recently, it was generally believed that persons with major mental illness such as schizophrenia and bipolar illness were not more likely to commit violent crimes than the general population (44). However, a growing body of evidence has shown a relationship between mental illness and violence, especially among persons who are psychotic and do not take their medications (45-53). This relationship is most striking in relatively nonviolent societies, such as in Scandinavia. For instance, Mednick and his colleagues (49) found that males in Denmark with a severe mental disorder who were admitted to a psychiatric hospital by age 44 represented only 5 percent of the total population of males but were responsible for about 30 percent of all the violent offenses committed by males. Likewise, female mental patients in Denmark constituted about 5 percent of the female population but were responsible for 50 percent of all the violent offenses committed by females. Similar findings were noted in Sweden (54).

Substance abuse also increases the risk of violent behavior, particularly in combination with severe mental illness (44,47,51,53,55,56). While it would appear that the vast majority of persons with serious mental illness are not more dangerous than the general population, the recent literature cited above suggests the existence of a subgroup that is more dangerous. It has been asserted that violent behavior by this subgroup stigmatizes mentally ill persons generally and that it will be difficult to reduce the stigma until the violence of this subgroup is addressed (4).

Causative factors

The factors most commonly cited as causes of mentally ill persons' being placed in the criminal justice system are deinstitutionalization and the unavailability of long-term hospitalization in state hospitals for persons with chronic and severe mental illness, more formal and rigid criteria for civil commitment, the lack of adequate support systems for mentally ill persons in the community, the difficulty mentally ill persons coming from the criminal justice system have gaining access to mental health treatment in the community, and a belief by law enforcement personnel that they can deal with deviant behavior more quickly and efficiently within the criminal justice system than in the mental health system (57,58). A factor less commonly discussed is the public's attitudes toward persons with mental disorders who commit crimes.

In an article about the homeless mentally ill population, Belcher (59) wrote that“a combination of severe mental illness, a tendency to decompensate in a nonstructured environment, and an inability or unwillingness to follow through with voluntary aftercare arrangements and take prescribed medication contributed to involvement with the criminal justice system. Wandering aimlessly in the community, psychotic much of the time, and unable to manage their internal control systems, these people found the criminal justice system was an asylum of last resort.”

Deinstitutionalization.

As noted, the belief that deinstitutionalization is a cause of mentally ill persons' being placed in the criminal justice system is a widely held theory for which some evidence exists (5,17). It can certainly be demonstrated that less room currently exists in state mental hospitals for chronically and severely mentally ill persons. In 1955 when the number of patients in state hospitals in the U.S. reached its highest point, 559,000 persons were institutionalized in state mental hospitals out of a total national population of 165 million. Now the figure is 72,000 for a population of more than 250 million. In about 40 years, the U.S. has reduced its number of occupied state hospital beds from 339 per 100,000 population to 29 per 100,000 on any given day (60). However, these figures may not accurately reflect the numbers of persons who receive highly structured 24-hour care because of the development and growth of a variety of community psychiatric facilities (many of them locked) in the various states that attempt to provide this kind of care (61).

In our opinion, deinstitutionalization set the stage for increasing numbers of mentally ill persons to enter the criminal justice system. Moreover, serious problems in implementing deinstitutionalization have often been encountered, such as inadequate or inappropriate outpatient treatment, insufficient community resources, and insufficient 24-hour highly structured psychiatric care facilities for those who need them. To the extent that deinstitutionalization has resulted in these problems, we believe that it is a significant factor accounting for the placement in jails and prisons of many mentally ill persons who would otherwise be treated in the community or in a hospital.

More restrictive civil commitment criteria.

Many people believe that more stringent civil commitment criteria have contributed not only to deinstitutionalization but to an increased number of mentally ill persons in jails and prisons (57,59,62,63). In 1969 California's then-novel civil commitment law, the Lanterman-Petris-Short Act, went into effect. Within a decade every state and Puerto Rico made similar modifications in their commitment codes. Such a rapid and complete consensus among legislatures is virtually unprecedented. More important, it reflected a nearly universal view that past inattention to the rights of mentally ill persons needed to be corrected.

In effect, the new civil commitment laws accomplished three things. First, the laws changed the substantive criteria for commitment from more general criteria that simply embodied concepts of mental illness and need for treatment to more specific criteria that embodied either dangerousness resulting from mental illness or the incapacity to care for oneself. Second, the laws changed the duration of commitment from indeterminate and extensive periods to determinate and brief periods. Third, the new laws explicitly provided that persons civilly committed have rapid access to the courts, to attorneys, and, in some cases, to jury trials; this access ensured the kinds of due-process guarantees to civilly committed persons that criminal defendants had obtained over the previous decade (64).

These procedural safeguards and clear commitment standards resulted in fewer as well as shorter commitments. Thus many mentally ill individuals who would otherwise have been civilly committed by family or others were now left to reside in the community. Moreover, the civil commitment standard for dangerousness in some states, such as Alaska (65), California (66), and Washington (67), becomes more restrictive when extended commitments are sought. Therefore, only the most dangerous mentally ill persons remain hospitalized, and the less dangerous are discharged. The result is greatly increased numbers of mentally ill persons in the community who may commit criminal acts and enter the criminal justice system.

On the other hand, it has been observed that changes in civil commitment law have often not had in practice the impact intended by those who wrote them (68). These reforms have been resisted by judges, mental health professionals, families, and even attorneys when they were seen as shifting the focus away from patients' treatment needs. Thus in some instances more restrictive commitment laws may not have been an important cause of an increased number of mentally ill persons in jail.

Access to treatment.

The availability, or lack of availability, of treatment resources in the community has three important aspects. First, it is clear that in most, though by no means all, jurisdictions in this country, mental health treatment, housing, and rehabilitation resources are insufficient to serve the very large numbers of mentally ill persons in the community (69). For instance, case management has come to be viewed as one of the essential components of an adequate mental health program (20,41,70). However, the criminal justice system is ill prepared to provide case management services to mentally ill persons leaving jails and prisons. In many jurisdictions, local mental health agencies have also been slow to provide these services to this population (58).

Second, community mental health resources may be inappropriate for the population to be served (25). For instance, mentally ill persons may be expected to come to outpatient clinics when the real need for a large proportion of this population is outreach services. Some service providers may lack the ability to provide the degree of structure required by many mentally ill offenders.

Third, mentally ill persons who have been in jail may not be able to gain access to community treatment even when it is available. These persons have been described as resistant to treatment, dangerous, seriously substance abusing, and “sociopathic” (58,62,71), characteristics generally not considered desirable by most community mental health agencies. Further, since many of these agencies may not have the capability to provide the needed structure, limit setting, and safety for staff necessary to successfully treat these persons, their reluctance to treat them may be appropriate.

A large proportion of mentally ill persons who commit criminal offenses tends to be highly resistant to psychiatric treatment (57,62,72). They may refuse referral, may not keep appointments, may not be compliant with psychoactive medications, may not abstain from substance abuse, and may refuse appropriate housing placements. As Whitmer (3) has observed, attempts at outpatient treatment with such persons “take on the aspect of a contest that a woefully unprepared therapist must sooner or later forfeit.” Hence, he used the term “forfeited patients” to emphasize that these persons are not just passively lost to treatment, but that mental health professionals have actively struggled to treat them and have had to acknowledge defeat.

Thus the mental health system finds these mentally ill offenders extremely difficult to treat and resists serving them (57,71). This reluctance extends to virtually all areas of community-based care, including therapeutic housing, social and vocational rehabilitation, and general social services (58). Moreover, many mentally ill offenders are intimidating because of previous violent and fear-inspiring behavior. Treating this group is very different from helping passive, formerly institutionalized patients adapt quietly to life in the community (73). Community mental health professionals are not only reluctant but may also be afraid to treat them, especially when measures are not adopted to ensure staff safety. Then these mentally ill persons are left for the criminal justice system to manage (71). On the other hand, we have seen outpatient facilities in which structure is provided, staff are protected, and mental health and criminal justice staff closely collaborate; under such circumstances, many of these persons are successfully treated.

The role of the police.

A large proportion of acutely mentally ill persons come first to the attention of the police (74-77). Even if the police consider the problem to be mental illness, the mental health option can involve a number of problems and irritants. There may be long waiting periods in emergency rooms during which police officers cannot attend to other duties. Mental health professionals may question the judgment of police and refuse admission, or they may admit for only a brief hospital stay a person who just a short time before constituted a clear menace to the community (57,78,79).

On the other hand, the police know very well that if they refer a psychiatric case to the criminal justice system, the offender will be dealt with in a more systematic way. He or she will be taken into custody, will probably be seen by a mental health professional attached to the court or in the jail, and will probably receive psychiatric evaluation and treatment. Thus arrest is a response with which police are familiar, one over which they have more control, and one that they believe will lead to an appropriate disposition (57,80). Moreover, when persons who are socially disruptive are excluded from psychiatric facilities, the criminal justice system becomes the system “that can't say no” (62).

With regard to minor offenses, a number of factors have been proposed to explain why a mentally ill person is arrested rather than taken to a hospital. A person who appears mentally ill to a mental health professional may not appear so to police officers, who, despite their practical experience, have not had sufficient training in dealing with this population and are still laypersons in these matters (63,81). Also, mental illness may appear to the police as simply alcohol or drug intoxication, especially if the mentally ill person has been using drugs or alcohol at the time of arrest. Still another factor is that in the heat and confusion of an encounter with the police and other citizens, which may include forcibly subduing the offender, signs of mental illness may go unnoticed (5).

In addition, law enforcement officers may be more inclined to take mentally ill persons to jail if they believe no appropriate community alternatives are available (82), a practice that has been referred to as“mercy booking.” Although this practice may be viewed as unconstitutional, the vast majority of states have not enacted legislation against detaining noncriminal mentally ill people in jail (19).

The demands of citizens also come into play. Many retail stores have a policy that anyone caught shoplifting should go to jail, and store managers are instructed to make a citizen's arrest and call the police without exception. In another kind of situation, people who have just been assaulted by a psychotic person are frequently not inclined to be sympathetic to their assailant even when mental disturbance is evident. Thus an angry citizen may insist on signing a citizen's arrest and having the person taken to jail.

>Society's attitudes.

The public has traditionally believed that any sentence other than prison is too lenient for serious offenders, even if they are mentally ill (83). Moreover, some view mental illness as volitional and perhaps a deliberate attempt to avoid punishment (84,85). Still another important factor is the public's fear of mentally ill persons who commit criminal offenses.

The public's growing intolerance of perpetrators, whether mentally ill or not, is demonstrated by its acceptance of and desire for more restrictive detention laws for offenders. With respect to offenders with mental disorders, some states have repealed sexual psychopathology laws that permitted mental health treatment for sex offenders rather than criminal processing and imprisonment. Diminished capacity, which can be a factor in granting a more lenient sentence, has also been repealed in a number of states. Moreover, legislation has been passed whereby offenders with mental disorders in prison can have their periods of social control extended if they are identified as dangerous before their parole date or the expiration of their sentence. For example, in California mentally ill offenders considered to be dangerous (86) and sexually violent predators (87) are usually transferred on their parole date or on expiration of their sentence to state mental hospitals, where they are confined for treatment for renewable periods of one or two years. In our opinion, these laws reflect the attitudes of society toward mentally ill offenders.

Although psychiatric interventions exist in the criminal justice system, mentally ill persons are more strictly controlled in that system than are patients in psychiatric hospitals (57). Moreover, the criminal justice system, despite protestations to the contrary, appears to have little interest in decriminalizing persons with psychiatric disorders even though they represent a considerable burden and utilize scarce resources. In a thoughtful article, Laberge and Morin (57) observed that a general decriminalization of psychiatric cases would threaten the criminal justice system to its foundations because such an approach might be perceived as undermining the principle of equality of all before the law. This perception would exist even where criminal law recognizes mental disorders as conferring a special status.

Specific treatment of mentally ill persons in the criminal justice system is often seen as special treatment both by the general public and within the criminal justice system. For instance, the insanity defense is perceived by most Americans as a frequently raised defense, as well as a way to evade justice. However, studies have shown that this defense is seldom used and rarely successful (88,89). In addition, it has been demonstrated that persons who successfully use this defense may be detained for considerably longer periods than others convicted for the same offenses (90,91).

Moreover, it appears the criminal justice system is more inclined to interpret and deal with criminal behavior in terms of illness when the deviant person acknowledges the illness and is willing to undergo treatment for it (92). Clearly, the appropriateness of treating mentally ill offenders safely in the community should be assessed. However, undertaking successful treatment for this population can be daunting. For instance, Brelje (93) wrote that effective psychotherapy for mentally ill offenders involves the patient's insight, an awareness of vulnerability to or presence of a mental disorder, a realistic understanding of the nature of the mental illness, a motivation to change or prevent recurrence of symptoms, an acceptance of treatment goals and strategies, realistic personal goals, and the patient's awareness of his or her legal status and its meaning.

However, Laberge and Morin (57) have observed that many mentally ill offenders do not take responsibility for their illness or their offenses and do not acknowledge their need for treatment. They refuse a therapeutic relationship and refuse to take medication and keep appointments. Therefore, they are often not seen by society as persons who should be “excused” for their legal transgressions. It appears that despite the concern of mental health professionals and many family members about mentally ill persons in jail, the general public would show little support for not placing social controls on individuals who commit offenses and refuse to submit to treatment that sets limits on their behavior.

Thus criminalization of mentally ill persons who have committed minor offenses cannot be seen as resulting simply from the usual explanations of lack of long-term hospitalization, lack of adequate support systems in the community, difficulty in gaining entry into the mental health system, and more restrictive criteria for civil commitment. Another crucial factor is society's concern that criminal offenses be dealt with and that persons committing them be controlled and punished, especially if they are not clearly willing to accept the patient role.

Conclusions and recommendations

Much has been learned about what needs to be done to prevent mentally ill persons from being inappropriately placed in the criminal justice system and about how to treat them once they are there and after they are released into the community. What has been lacking is widespread and comprehensive implementation of interventions shown to be effective (29). Several of these strategies are summarized below.

Steps should be taken to prevent inappropriate arrest of mentally ill persons (94). The police are often the first to respond to emergencies involving people with severe psychiatric disturbances (75). However, the police may not always recognize a need for, or have access to, emergency psychiatric resources. Clearly, mental health expertise is needed at this point to prevent criminalization. There is a pressing need for formal training of police officers to help them better understand mental illness and to improve their attitudes toward individuals with mental disorders (81,95).

Mental health consultation provided to the police in the field can result in a response that combines the specialized knowledge and expertise of law enforcement and mental health professionals. Such an approach can greatly increase the number of mentally ill persons given appropriate access to the mental health system rather than inappropriately diverted to the criminal justice system. For example, an evaluation of psychiatric emergency teams consisting of police officers and mental health professionals found that the teams were able to deal with psychiatric emergencies in the field, even with a population characterized by acute and chronic severe mental illness, a high potential for violence, a high prevalence of serious substance abuse, and long histories with both the criminal justice and the mental health systems (77). These teams took or sent almost all of the persons in crisis to the mental health system and not to jail.

For individuals who are arrested and placed in jail, it is generally recommended that the facility routinely screen all incoming detainees for severe mental disorder and that jail administrators negotiate programmatic relationships with mental health agencies to provide multidisciplinary psychiatric teams (24,78). These teams should be established inside jails to provide short-term crisis evaluation, treatment, and referral to a psychiatric hospital if necessary. The teams should include psychiatrists so that psychoactive medications can be prescribed.

Mentally ill detainees who have committed minor crimes, such as trespassing and disorderly conduct, should be diverted to the mental health system entirely, or at minimum for treatment. For instance, mental health teams should be readily available for consultation to the arraignment courts and especially to the municipal courts, where many acutely psychotic patients appear with very minimal criminal charges. Steadman and associates (96) found that only a small number of U.S. jails have diversion programs for mentally ill detainees. They also observed that objective data on the effectiveness of these programs are lacking. On the other hand, it has been found that court-mandated and -monitored treatment in lieu of jail was effective in obtaining a good outcome for chronically and severely mentally ill persons who committed misdemeanors (97).

Belcher (59) wrote that a system that relies solely on voluntary compliance may not provide adequate structure for mentally ill offenders. He and others (4,94,98-101) recommended such mechanisms as outpatient commitment, court-monitored treatment, treatment as a condition of probation or parole, and psychiatric conservatorship or supervision by agencies such as Oregon's Psychiatric Security Review Board (102). Freeman and Roesch (103) acknowledged that the court or parole board has a right to set conditions for release to the community that include mandatory treatment. Nevertheless, mental health professionals have an ethical and legal obligation to fully inform patients about the nature of the treatment and obtain their consent for it.

It is important to recognize that persons with mental disorders who are discharged from psychiatric or correctional institutions experience multiple problems that cannot be adequately treated in traditional community-based facilities (104,105). Thus placement in the community often results in rehospitalization or reincarceration (106). To reduce this cycle, assertive case management programs are recommended.

The great majority of mentally ill offenders need the basic elements of case management, which starts with the premise that each person has a designated professional with overall responsibility for his or her care (20,107). The case manager formulates an individualized treatment and rehabilitation plan with the participation of the mentally ill person and often the supervision of the court. As care progresses, the case manager monitors the mentally ill person to determine if he or she is receiving and complying with treatment, has an appropriate living situation, has adequate funds, and has access to vocational rehabilitation.

The case manager not only provides outreach services, but also serves as an advocate for the individual and makes sure that the mentally ill person is not drifting away from the supportive elements of such a network. An assertive case management program deals with clients on a frequent and long-term basis, using a hands-on approach that may necessitate meeting with clients “on their own turf” or even seeing clients daily (106). This form of contact and familiarity with clients helps the case manager and client anticipate and prevent significant decompensation.

Important advances have been made in recent years in the management of the violent behavior of severely mentally ill persons (108,109). Behavior therapy and pharmacotherapy—in particular, the use of the new atypical antipsychotic medications—are but a few examples. It is crucial that these modalities be widely implemented.

Mental health agencies in the community must be able to provide the degree of structure and limit setting needed by mentally ill offenders, as well as ensure the safety of staff. When highly structured 24-hour care is required, it should be provided.

The role of family members is an important aspect in the care of mentally ill offenders. Often overlooked are family members' needs for guidance and support. Families should be instructed in ways to help stabilize their relative (107). They should also be involved in support programs to help them during crises and in self-help programs so they can benefit from the experience of other families in similar situations (110).

We believe that a significant increase in mental health services for severely mentally ill persons, from outpatient treatment and case management to highly structured 24-hour care, would result in far fewer mentally ill persons' committing criminal offenses. Thus one of our most important recommendations is for increased mental health services. The criminal justice system should not be viewed as an appropriate substitute for the mental health system. Moreover, it has been our experience that an enormous stigma is attached to people who have been categorized as both mentally ill and an offender, and it is thus extremely difficult to place them in community treatment and housing. The difficulty is even greater when they have been in a forensic hospital.

Clearly, many mentally ill persons who commit criminal offenses present formidable challenges to treatment because of their treatment resistance, poor compliance with antipsychotic medications, potential dangerousness, high rate of substance abuse, and need for structure. To a large extent, the public mental health system has given up on them and allowed them to become the responsibility of the criminal justice system. We believe these recommendations would contribute to successful treatment of this population.

Implementing these recommendations would mean tailoring mental health services to meet the needs of mentally ill offenders and not treating them as if they were compliant, cooperative, and in need of a minimum of controls. The lives of a large proportion are characterized by chaos, dysphoria, and deprivation as they try to survive in a world for which they are ill prepared. They cry out for treatment and for structure, and we believe it is the obligation of the mental health system to provide it. If effective and appropriate interventions are provided, these individuals may not only improve psychiatrically but may also engage in considerably less criminal behavior.

Dr. Lamb is professor of psychiatry and director of the division of mental health policy and law of the department of psychiatry at the University of Southern California School of Medicine, 1937 Hospital Place, Los Angeles, California 90033. Dr. Weinberger is associate professor of clinical psychiatry and chief psychologist at the Institute of Psychiatry, Law, and Behavioral Sciences of the University of Southern California School of Medicine.