Abstract
This study reports on restoration outcomes of a sample of pretrial defendants (n = 877, 69% male) who were found incompetent to stand trial and underwent restoration services in a large urban county. Each male defendant was initially assigned to restoration in one of four settings on a continuum of services of varying intensity (ie, outpatient, jail general population, dedicated jail-based restoration unit, or forensic hospital inpatient unit) based on the defendant's assessed clinical need. Of those who received services on the jail-based restoration unit (n = 398), 40 percent were restored to competency, 31 percent were diverted out of the criminal justice system, and 29 percent were referred for more intensive inpatient services, primarily because of refusal of medication (i.e., the jail would not allow involuntary medication, even if court-ordered). Advantages of restoration on the jail unit compared with inpatient hospitalization included more rapid institution of restoration services and higher rates of diversion out of the criminal justice system at one-third of the cost of inpatient restoration services. A continuum of restoration services that allows the type of restoration service to be matched to the needs of the individual incompetent defendant has significant advantages over routine transfer to a forensic hospital for restoration.
Traditionally, pretrial criminal defendants adjudicated incompetent to stand trial (IST) have been admitted to inpatient state hospital forensic units for restoration of competency. As the number of people with mental illness in jails has grown,1 so, too, have the referrals to state hospitals for competency restoration.2 In many jurisdictions, increased referrals for hospitalization and decreased funding for state mental health facilities have led to long waitlists and a search for alternative methods to restore competency. These waitlists have led some jurisdictions to implement less intensive methods of competency restoration, including outpatient restoration, restoration services provided to those in general population in jails, and jail-based restoration units.1,–,4
Although the delivery of restoration services in the general population of a jail is increasing, there are relatively few jail-based dedicated restoration units. Using one published survey5 and personal communications from directors of several new programs, we have been able to identify fewer than a dozen such programs nationwide, and there are relatively few data regarding their efficacy. Jennings and Bell4 reported a competency restoration rate of 83 percent in a pilot jail-based program in Virginia, and Rice and Jennings2 reported a competency restoration rate of 55 percent in a jail-based unit program in California with a high (85%) rate of medication compliance. Despite these success stories, jail-based programs have been criticized for providing less effective services and prolonging incarceration of mentally ill defendants by restoring them in a correctional rather than a hospital environment.6,7
This article describes the development, functioning, and seven-year outcomes of a jail-based restoration unit in the context of a system of other restoration services that create a continuum of restoration services serving a large, urban county that includes a substantial portion of Atlanta, Georgia.
Development of the Service Continuum
Creating the Jail-Based Restoration Unit
In Fulton County, Georgia, in 2011, the waitlist for IST defendants to be admitted to a state hospital forensic unit had grown to over 60 days. In that jurisdiction, most defendants who were ordered by the court to undergo a competency evaluation were referred to a university-based forensic service that performed the evaluations under a contract with the state. Clinically, many defendants whom evaluators opined were IST did not appear to need the intensity of treatment of an inpatient hospital, but likely could be restored in other settings with a combination of medication (e.g., if they were psychotic) and education (e.g., if they had deficits in understanding legal processes). For these reasons, the state contracted with the university forensic program to develop a 16-bed pilot unit in the county jail to restore male defendants. The unit was initially funded at a daily rate that was less than a quarter of the cost of inpatient hospitalization, not including the jail's daily cost for lodging and food or the cost of medications. The development of this competency restoration unit (CRU) was supported by the jail administration, who saw it as a means of relieving some of the overcrowding that was due to IST defendants waiting for an inpatient hospital bed. The jail's private mental health provider was also supportive because the CRU program offered to provide most of the mental health care for 16 of the jail's most problematic patients. In addition, the restoration program agreed to utilize the private provider's formulary and medical record system, have its clinicians credentialed by the mental health provider, and abide by the provider's policies. The jail did not allow involuntary medication except for short-term emergencies. In short, the program promised to run a specialized day-treatment program on a jail unit with after-hours coverage provided by the jail's mental health provider. Some opposition came from the public defender's office because the attorneys were concerned that their clients would receive second-rate care compared with being sent to the hospital; this concern was later alleviated, as described below.
Staffing and Programming
The restoration program staffs the day shift Monday through Friday. Current staffing includes a part-time (80%) on-site psychologist director, one full-time social worker, one masters-level mental health clinician, and a part-time diversion specialist. Psychiatric care is provided by forensic psychiatry fellows under the supervision of faculty forensic psychiatrists, totaling about 1.1 full-time equivalent (FTE) psychiatrist. Forensic psychology trainees and supervisors, totaling about 1.5 FTE, assist in running groups, working individually with defendants, and conducting psychological testing. Security is provided by two jail deputies for all times when mental health staff are on the unit. When CRU program staff are not on site, the unit is managed by jail personnel. Nursing care, medical care for physical problems, and back-up psychiatric care is provided by the jail's contracted private health provider.
The program at the CRU functions much like the daytime operation of an inpatient forensic unit in a state hospital. The participants in the program all have a daily schedule of groups, which may include legal education, conflict resolution, values clarification, basic reading skills, and medication adherence, as well as individual sessions and community meetings. There is also an emphasis on cognitive remediation activities, which are interwoven into the daily schedule for the purpose of enhancing problem solving, attention, concentration, and memory, all of which are seen as critical to the process of competency restoration. These group activities contribute to a positive therapeutic milieu, which was achieved despite the challenges of creating such an atmosphere in a correctional setting. Most of the defendants work reasonably well together, at times helping the more impaired participants in the program and generally fostering a more interactive environment with a very low rate of aggression and violence. In the seven years of the study period, there have been seven minor inmate-on-inmate assaults, none of which required medical attention, and only one inappropriate touching incident involving a staff person, a rate much lower than that reported by the state forensic hospital. This low rate of violence may be due to the presence of jail deputies who have different approaches to violent incidents than hospital staff, likely serving as a deterrent to violence, and the fact that highly aggressive, severely behaviorally disordered defendants are referred to the hospital. The milieu spirit likely contributes to encouraging some inmates who refused psychotropic medication while in the jail general population to consent to medication after some time on the CRU.
Admission Criteria
All defendants whom an evaluator opined were IST were screened for admission to the unit. The only defendants who were initially excluded were those housed on a maximum security jail unit whom the jail would not transfer to a lesser security unit and those who were so behaviorally impaired as to require intensive mental health or security management that was not available on the CRU (e.g., inmates needing frequent observation for suicide precautions or who were so aggressive as to be considered unmanageable given the available staff). Such inmates were referred directly to the hospital for restoration services. Medication refusal in the general population was not a reason for exclusion. As the continuum of services was developed, defendants who could likely be restored in the jail general population with some individual tutoring about legal processes and those who appeared likely to have their charges dismissed or otherwise diverted out of the jail remained there for treatment with the possibility of later transfer to the CRU if restoration in the general population was unsuccessful.
Initial competency evaluations of defendants not on bond are conducted when the evaluee is in the jail general population. If the defendant was opined IST and moved to the CRU, the evaluator was excluded from the treatment team for that inmate, but that evaluator did conduct the re-evaluation when the treatment team thought it appropriate. In most cases, one or two interview sessions provide sufficient information to reach a conclusion about competency. There are cases, however, when such limited interaction does not provide sufficient information, such as when there is a suspicion of malingered impairments or when the competence appears to be borderline. In such cases, the evaluee is moved to the CRU where he can be observed for an extended evaluation. Such observation for one to two weeks generally provides sufficient clarifying data to reach a judgment about competency and avoids the necessity of moving the defendant to a hospital unit for an inpatient evaluation, which is the method that has traditionally been utilized to address these complex evaluations.
Early Lessons Learned
The jail CRU opened in October 2011. Moving 16 defendants to the unit immediately shortened the wait time for IST defendants awaiting inpatient hospitalization to less than 20 days. When an evaluator writes a report to the court opining a defendant IST, there is frequently a significant delay before the court holds a hearing and makes a formal finding that the defendant is IST, at which point the defendant is put on the waiting list for the hospital. Because almost all of the competency evaluations were performed by the university forensic service, once an evaluator deemed a defendant IST, that defendant could be transferred to the CRU and restoration services initiated before the court made a formal finding of incompetence. In many cases, restoration could be accomplished without the court ever making a formal finding of incompetence.
Diversion Out of Corrections
Early on, it became apparent that, for a significant number of defendants, if a good treatment plan in the community could be developed, the court and prosecutor were often amenable to dismissing or to dead-docketing the charges and releasing the defendant. In some cases, judges were willing to function in a manner similar to a mental health court and release the defendant on a signature bond (i.e., a bond that releases the defendant on his own recognizance with no bail) with continued treatment as a condition of the bond. For release on a signature bond, either the defendant had not yet been formally adjudicated IST or was deemed competent to assent to the bond. Once the possibility of diversion was recognized, the program more actively pursued diversion out of the correctional system for appropriate defendants rather than seeing competency restoration as the goal for all defendants. With this emphasis, about a third of defendants opined IST were diverted out of the correctional system in lieu of being admitted to a restoration program. This result, along with the shortened wait for restoration services, had the effect of markedly reducing the opposition of the public defenders to the program, who now saw a significant proportion of their clients having much shorter stays in an institution (either jail or hospital).
Continuum of Services
With the establishment of the jail CRU, the university forensic program oversaw the initial evaluation, restoration on the jail unit, and admission to the inpatient forensic hospital. From this system-oriented perspective, it became clear that some defendants could be restored in other ways. For example, defendants evaluated while on bond could potentially be restored as outpatients. Some defendants in the jail general population who were psychotic and unmedicated, but willing to take medications, were thought likely restorable with medication treatment in the general population if they did not appear to need the support and educational interventions of the full-day program of the restoration unit. If medication resolved their psychosis but they were still incompetent, they could then be transferred to the CRU. This led to the conceptualization of a continuum of services, essentially matching the restoration needs of a defendant to an appropriate intensity of treatment rather than the uniform approach of transferring all IST defendants to an inpatient forensic hospital. If a defendant could not be restored with lower-intensity services within one or two months, he could then be transferred to a higher-intensity intervention. When the need for a continuum was understood, the program began services for defendants in the jail general population and an outpatient restoration program in the local public hospital. Figure 1 displays the array of services.
A continuum of services.
As noted above, the CRU program described here serves only men. Women are housed in a different jail, and although the forensic service evaluates them for competency to stand trial, it provides no restoration services. After an evaluator opines a female defendant IST, if the defendant is not on the mental health caseload, the evaluator communicates with the mental health provider regarding clinical concerns, medication compliance, symptom severity, behavioral observations, and emergent issues (e.g., self-injury and hygiene). The service does attempt to divert suitable female defendants, but those who are not diverted typically are sent to the hospital for restoration. To provide some services for women, a special program for women was developed (described elsewhere).8 Outcomes of women found IST are also presented below for comparison purposes, although it should be understood that the women as a group differ in significant ways.
Methods
Outcome data were collected in the seven-year period after the CRU started in 2011. At the time of initial evaluation for competency to stand trial, data were collected regarding demographics, court and offense characteristics, evaluator opinion, and the evaluator's clinical assessment as to whether the underlying cause for an IST defendant's impairments involved psychosis, cognitive problems, both, or other. At the restoration sites, data were collected regarding length of intervention and outcome. For those admitted to the CRU (starting nine months after the program was begun), if a defendant was sufficiently organized to participate in psychological testing, he was administered the Wechsler Abbreviated Scale of Intelligence, Second Edition (WASI-II). Outcomes of inpatient hospitalization were obtained from state hospital records. All data were collected as part of the routine operation of the programs. The study protocol was approved by the institutional review boards of Emory University and, for the inpatient hospitalization data, the Georgia Department of Public Health.
Results
In the seven-year study period, the forensic service issued initial opinions on competency to stand trial in 1,662 cases (1,400 men, 262 women; 9% white, 88% African American). Of the 1,400 evaluations of men, 638 (46%) were opined IST. Of the 262 women, 158 (60%) were opined IST. These IST rates are considerably higher than those reported in most other studies, i.e., around 20 percent.9,10 While this could potentially reflect that the university forensic service evaluators had a lower bar for an IST finding than the national norm, the rate of IST findings has increased over the years despite stable supervisors. One author (P.A.) has conducted teaching sessions of state forensic evaluators with audience surveys and has noted high consistency between the ratings of members of the university service and other state evaluators. Anecdotally, findings in the vast majority of evaluations were quite clear-cut, and practically no findings of incompetence were contested in court. Several factors likely account for the high rate of IST findings. First, the forensic service conducts annual trainings of the public defenders, which has resulted in much better screening of defendants being referred for evaluation. It is rare for the program to receive a referral of a defendant who is obviously competent: practically all have significant mental health concerns or intellectual disabilities. Second, about a quarter of the sample were charged with a misdemeanor, often resulting from behavior attributable to a mental illness (e.g., criminal trespass), and the rate of an IST opinion for misdemeanor defendants (64%) was considerably higher than for felony defendants (39%). Third, the rate of cognitive impairment in our sample was quite high, due in part to histories of head trauma and substance abuse. While IQ measures were not obtained for all defendants, after nine months of operation the CRU began routine testing of defendants who could cooperate utilizing the WASI-II. The median full-scale IQ of tested defendants (n = 177) on the unit was 69 and the mean was 71. Only 36 percent of tested defendants had a full-scale IQ score above 85. It should be noted that, for many psychotic defendants, a low IQ score likely reflects poor effort or disorganized thinking rather than their maximum capacity when psychosis is in remission.
One of the advantages of a jail-based unit is that restoration services can begin before the court issues an order finding the defendant IST and transferring him to the hospital. In our jurisdiction, the delay in obtaining such an order could be significant. All that is required for the internal transfer is the request of the CRU; 72 percent of admissions to the CRU began receiving restoration services before an IST order was obtained. Outcomes by the type of disposition outcomes are shown in Figure 2.
Outcomes by restoration intervention (percentages given are within intervention group).
Jail Restoration Unit Outcomes
As can be seen in Figure 2, the rate of restoration of those treated on the CRU (n = 398) was 40 percent. An additional 31% of defendants were diverted out of the criminal justice system; 76 percent of misdemeanor defendants were diverted, while only 25 percent of felony defendants were diverted. The mean length of stay was 98 days and did not vary markedly by outcome.
Somewhat less than a third of defendants (29%) could not be restored or diverted and so were referred for inpatient hospitalization; 70 percent of those referred to inpatient hospitalization were defendants who refused or were only intermittently compliant with taking antipsychotic medications. The restoration unit milieu had some success in persuading defendants who initially refused medication to become more accepting, but 74 percent of initial refusers continued to refuse. By jail policy, involuntary medications, even those that are court-ordered, cannot be given. For those with felony charges, the program sometimes recommended the court hold a Sell hearing11 to obtain a court order authorizing involuntary medication to restore competency. Then, when a defendant arrived at the hospital and continued to refuse medication, he could immediately be medicated, which reduced the length of hospitalization and eliminated the burden on hospital staff of having to seek and testify at a Sell hearing.
Admissions to the CRU were classified by the nature of the impairment that interfered with competency: 72 percent were psychotic without cognitive impairments, 10 percent had primarily cognitive impairments, 13 percent were both psychotic and had cognitive impairments, and 4 percent were classified as other, generally reflecting that they were admitted for suspected malingering that was later confirmed or had transient conditions such as substance-induced states. Other conditions, such as mood disorder or personality disorder, were very rarely causes of IST. Table 1 shows the outcomes of admissions classified by type of impairment and whether the defendant was compliant with taking recommended medications.
Outcomes of CRU Admissions by Type of Impairment and Medication Compliancea
As can be seen in Table 1, the majority of defendants who were not restored and referred for inpatient treatment were medication-refusers who had psychosis as a contributing impairment. Although low IQ is associated in the literature with a low probability of restoration,12 in the CRU sample, medication refusal was a much stronger predictor of failure to restore than low IQ (Table 1). Taking medication markedly increased the likelihood of psychotic defendants being restored.
Very few defendants (n = 3) with cognitive impairments alone were referred to inpatient care, and those were defendants who were thought non-restorable but potentially civilly committable. If, after an attempt at restoration, the CRU evaluator concluded a defendant was not competent, not restorable, and not committable (n = 53), that opinion was reported to the court which then released the defendant, generally with an aftercare plan, without referral to the hospital.
Outcomes of Other Services
About a third (n = 7) of those seen in outpatient restoration were restored. Overall, few defendants were referred for outpatient restoration. Poor attendance was a significant factor in the outpatient restoration outcomes.
Of the 165 defendants in the general population who were not admitted to the CRU but received restoration services, about half (n = 83) received restoration services in general population from CRU staff, and the remaining group had their cases managed in other ways (most commonly medication provided by jail medical services or CRU staff consulting with attorneys to facilitate diversion). Some (n = 26) were restored to competency. More than half (n = 107) had their charges dismissed or bonded out, whereas 32 were referred for inpatient hospitalization, and seven were resolved in other ways (2 died, and five were transferred to other jurisdictions on other charges).
For defendants who were admitted for inpatient hospitalization and for whom a final determination was reached (n = 282), the overall restoration rate was 83 percent. Table 2 gives the restoration rate by category.
Restoration Outcomes of Those Admitted to Forensic Inpatient Hospital Unit
For those defendants sent to an inpatient unit after a restoration attempt on the CRU, the rate of restoration in the hospital for medication refusers was 80 percent. This likely reflects that the hospital unit could administer medication involuntarily if required for safety or with a court order. It was reported some defendants explicitly stated that they were refusing medications while they were in jail but would be willing to take them voluntarily in a hospital setting. Of the 23 defendants who were medication-compliant on the CRU but who were not restored within three months and were admitted to the hospital, 59 percent were ultimately restored after a mean length of hospitalization of more than seven months. Those who were not restored tended to be those with delusional disorder that did not respond to medication. Overall, the hospital had high rates of restoration of those who were admitted.
Discussion
In comparison to a system in which the only option for restoration of competence is an inpatient unit, having a continuum of services such that the intensity of service is matched to the competency deficit of each defendant offers several advantages. Such a continuum allows for increased efficiency in decreasing both time to beginning restoration and cost. Outpatient restoration programs and restoration in the general population have been tested in numerous jurisdictions and afford considerable cost savings, but using restoration in the jail general population as a routine measure has been criticized as often providing defendants with inadequate services.6,7 A jail-based restoration unit can be a useful addition to the continuum of services. In our jurisdiction, if the jail CRU did not exist, all the defendants admitted to the CRU would have been put on a list for the inpatient hospital, adding approximately 5,800 patient-days each year to the forensic inpatient load. Forensic inpatient costs vary across jurisdictions but are estimated at over $750 per day, while the cost of the CRU, including the costs to the county in running the jail, is less than a third of that, so the net savings are on the order of $3 million per year. The main disadvantage of jail restoration is that when it fails to restore competency, the attempted lower-intensity services lead to a delay in beginning inpatient hospital restoration.
Another benefit of services provided in a jail is the more rapid institution of restoration services, particularly in jurisdictions where there are significant waitlists for inpatient services. In the case of Trueblood v. Washington State Department of Social and Health Services,13 the court found that IST defendants in need of inpatient restoration services must be sent to a hospital within a week of the IST order, and similar litigation has been brought in other states. In our sample, 72 percent of defendants who were not referred to the hospital began receiving restoration services prior to being found IST by a court. It should be noted that participation in treatment was voluntary, and defendants much preferred the CRU to the jail general population because they had single-occupancy cells, rather than double-occupancy cells, and they could spend more time out of their cells with the option of participating in groups and other programming.
Jail-based alternatives to hospital restoration have been criticized on the grounds that it is simply wrong or inappropriate to restore people who have mental health needs in a jail. Most IST defendants have chronic mental conditions that, aside from their IST legal status, would not qualify them for admission to a state inpatient unit. State inpatient units have become increasingly restrictive in limiting civil admissions to acute situations, a trend that has been recognized as contributing to the number of persons with mental illness in jails. If a defendant's underlying condition does not warrant hospitalization, it would seem to follow that the defendant's IST status does not confer a right to hospital treatment if timely and effective restoration services can be provided in a less intense setting. Our experience demonstrates that jail restoration can be instituted more quickly than hospitalization because it does not require a court finding of IST or a wait for hospitalization.
The main limitation in the CRU is the problem of treating those defendants who are incompetent on the basis of psychosis and refuse medication. Further research that identifies which defendants will likely continue to refuse medication after several weeks of refusal would assist in the more efficient use of resources. The CRU had a lower rate of restoration than the program described by Rice and Jennings2 or rates reported in studies of hospital restoration,3 or indeed by the hospital to which defendants were referred if they were not restored on the CRU but were thought to be potentially restorable. Three factors distinguish our sample from other reported samples: the high rate of medication refusal without the option of involuntary medication, the high rate of cognitive impairments, and the fact that many defendants who were potentially restorable were diverted out of the system. That said, it is clear that the higher level of intensity of services in an inpatient hospital will lead to higher rates of restoration when compared with a jail restoration unit.
The high rate of diversion out of the jail came as a surprise in the early years of the program and led to increased consideration of that possibility. Diversion is rarely mentioned in the competency-restoration literature. More typically, a defendant sent to an inpatient unit remains there until restored or deemed non-restorable. This may reflect that many mental health courts release defendants on a bond that requires outpatient treatment, and there are questions as to whether a defendant ruled IST can intelligently consent to such a bond. In our sample, diversion was generally achieved by having the charges dismissed if a promising outpatient treatment plan was in place. This diverted the defendant out of the correctional system. In only a few cases did the court find that defendants were competent to agree to outpatient treatment as a condition of bond even if they were not deemed competent to participate in a trial. There is no clear reason why such diversion could not be actively explored as an option for hospitalized IST defendants, and our results suggest that a significant portion of IST defendants are suitable for that approach.
The positive results of this study are not intended to state that the particular mix of restoration services and the procedures described in this study are suitable for all jurisdictions. Jurisdictions vary widely across such dimensions as the demographics and volume of IST populations, laws pertaining to competency restoration, handling of misdemeanor defendants, receptivity of judges to diversion, availability of diversion programs and state hospital beds, and procedures for handling involuntary medications. Differing laws, organizational structures, and local cultures call for different approaches. As just one example, if the jail restoration program operates at a county level, it requires a fairly large jail to provide a sufficient number of IST defendants to fill a 16-bed unit, and the downward scalability of such a program is unclear. Even most large county jails, such as ours, do not house a sufficient number of female IST defendants to fill a 16-bed unit. Jail restoration units that receive admissions from multiple counties or from the whole state would be one potential solution to this problem, an approach that is used in Colorado.14
Conclusion
A jail-based competency restoration unit is best utilized as one option in a continuum of restoration services. Matching a defendant's restoration needs to the type and intensity of restoration services received allows for a more efficient use of resources than is possible with a more limited set of options, especially when only inpatient hospitalization is available. This is particularly important in an increasing number of jurisdictions where more demand for services and limited forensic hospital beds has led to significant waits in jail for inpatient services. A continuum of approaches allows for more rapid commencement of restoration services at markedly decreased costs. Further research will help inform what characteristics of defendants deemed IST predict restorability, and at what intensity of intervention. In addition to attempting to restore competency, diversion of IST defendants out of jails and hospitals holds promise for decreasing the need for services and decreasing the time these defendants remain in an institution.
Footnotes
Disclosures of financial or other potential conflicts of interest: None.
- © 2020 American Academy of Psychiatry and the Law
