Jill A. Gordon, Ph.D**
Christina M. Barnes, M.S.***
Virginia Commonwealth University
Scott W. VanBenschoten, M.SW., M.PA.
Administrative Office of the United States Courts,
Office of Probation and Pretrial Services
Introduction
The Fundamentals of Jail Diversion for Mentally Ill Offenders
Jail Diversion for Mentally Ill Offenders: Evaluations and Outcomes
The Program: Background, Climate, and Development of the Jail Diversion Program
Research Methods
Client Characteristics and Needs
Client Progress in the Program
Client Outcomes
Discussion
Future Research
Introduction
According to the criminalization hypothesis, the deinstitutionalization era resulted in a shift of mentally ill persons from psychiatric hospitalization to the criminal justice system (Abramson, 1972; Hiday, 1992; Teplin, 1991). As a result of deinstitutionalization, in conjunction with other punitive policies, the institutional and community populations grew in the United States. Today there are over 2 million incarcerated offenders in U.S. jails and prisons, not including those under correctional supervision in the community (Harrison & Beck, 2005). Increases in the correctional population translate into higher numbers of mentally ill individuals coming into contact with the criminal justice system. It is estimated that at least seven percent of all offenders booked annually into U.S. jails suffer from a serious mental illness (e.g., Teplin, 1990b; Teplin, 1994, Teplin, Abram, & McClelland, 1996). This proportion is substantially higher than that found in the general population and excludes those with less severe diagnoses. While some might emphasize that it is only seven percent of all bookings, it is essential to understand that this encompasses approximately 800,000 to 1 million booked individuals, of which 72 percent meet criteria for a co-occurring substance abuse or dependency issue (Abram & Teplin, 1991; Abram, Teplin, & McClelland, 2003). Basically the deinstitutionalization trend caused the criminal justice system to become a de facto mental-health care provider, a role which many argue it is ill-equipped to perform (Steadman, Barbera, Dennis, 1994; Steadman, Deane, et al., 1999; Teplin, 1990a; 1990b; Torrey, Steiber, Ezekiel, Wolfe, Sharfstein, Noble, & Flynn, 1992).
Frustrated by offender outcomes and the misuse of jails as mental hospitals, many of those in the criminal justice system are increasingly opposed to the status quo. Examples of this shift include the creation of problem-solving specialty courts (i.e., drug courts, mental health courts) and diversion programs (both pre- and post-booking) (Petrila, 2003; Steadman & Redlich, 2006). The purpose of this article is to describe the creation of a jail diversion program for offenders diagnosed with a co-occurring disorder and detail the characteristics, progress, and outcomes of the clients who entered a program in which all services are delivered “in-house” rather than through a traditional brokerage system.
The Fundamentals of Jail Diversion for Mentally Ill Offenders
Jail diversion programs for mentally ill offenders provide an alternative to arrest, prosecution or conviction (e.g., Draine & Solomon, 1999; Draine, Blank, Kottsleper, & Solomon, 2005; Shafer, Arthur, & Franczak, 2004; Steadman et al., 1994). Numerous jurisdictions have adopted this approach to help reduce the number of individuals with mental illnesses who enter the criminal justice system (Cowell, Broner, & DuPont, 2004; Petrila, 2005; Shafer et al., 2004; Torrey et al., 1992). Additionally, such programs address a number of other concerns, such as misappropriated and/or underdeveloped treatment protocols for the mentally ill within prisons or jails (Grudzinskas, Clayfield, Roy-Bujnowski, Fisher, & Richardson, 2005). Finally, they have the potential to assist the already overburdened courts and jails (Cowell et al., 2004).
“Jail diversion” simply means redirecting mentally ill offenders away from the criminal justice system into community-based mental health services for treatment (Steadman et al., 1994; Draine & Solomon, 1999). When diversion takes place varies across jurisdictions, ranging from police diversion to judicial diversion. In general, these programs redirect individuals who have committed misdemeanors or non-violent felony offenses. It is argued that jail diversion is an alternative to incarceration or further penetration of the system and that treatment is more imperative than punishment (Draine et al., 2005).
It is important to recognize that jail diversion is not a new idea and was advocated by groups such as the National Coalition for Jail Reform in the 1970s and 1980s (Steadman, Cocozza, & Veysey, 1999). However, the number of programs has proliferated over the past 15 years, growing from an estimated 52 programs in 1992 (Steadman et al., 1994) to just over 400 programs in 2006 (Kirkman, personal communication). There are basically two types of jail diversion programs: pre-booking and post-booking. Pre-booking programs divert the individual before he or she is arrested, whereas post-booking programs divert the individual after arrest but before prosecution or sentencing for the pending charge (e.g., Shafer et al., 2004). Under post-booking programs, pre-trial detainment can occur prior to diversion.
While there is no model jail diversion program, key components have been identified. Foremost is interagency collaboration, consisting of cooperation between mental health and substance abuse providers, law enforcement officials, judges and prosecution/public defender offices (Shafer et al., 2004, Steadman et al., 1999; Steadman, Morris, & Dennis, 1995). All of these agencies have a vested interest, although their reasons and philosophies often differ, in reducing the number of mental health clients within the system. Therefore, cooperation is imperative for a well-functioning program. The stakeholders must define the target group for diversion, identify individuals as soon as possible in order to minimize system penetration, negotiate alternatives to incarceration that are community-based, provide appropriate linkages between community-based care, cross-trained case managers and community supervision and promote consistency in dispositions of cases (Steadman et al., 1995; Steadman, Davidson, & Brown, 2001). However, stakeholders must recognize that flexibility and sensitivity to jurisdictional needs is crucial. To illustrate, variations in programs exist due to the size and structure of local criminal justice systems, the perceived need for such services, and the availability of resources and services within the community and local politics (Morris and Steadman, 1994). All of these factors are not constant; they are continuously shifting and a program must recognize and accommodate not only offender variation but also shifts within the broader community.
Jail Diversion for Mentally Ill Offenders: Evaluations and Outcomes
Despite the significant proliferation in jail diversion programs, researchers warn that such implementation has been based upon a sparse empirical foundation (Draine & Solomon, 1999; Petrila, 2005; Steadman & Redlich, 2006; Steadman et al., 1999). Previous studies have suggested that jail diversion programs can result in “positive outcomes for individuals, systems and communities” (e.g., Steadman & Naples, 2005, p. 168). Unfortunately, since many of the programs are new, few studies go beyond the level of description (Steadman et al., 1999). However, a growing body of literature attempts to evaluate the effectiveness of both pre- and post-booking jail diversion programs (e.g., Borum, Dean, Steadman & Morrissey, 1998; Broner, Lattimore, Cowell & Schlenger, 2004; Cosden, Ellens, Schnell, Yasmeen, & Wolfe, 2003; Cowell et al., 2004; DuPont & Cochran, 2000; Hoff, Baranowsky, Buchanan, Zonana, & Rosenheck, 1999; Lamb, Weinberger, & Reston-Parham, 1996; Lamb, Shaner, Elliot, DeCuir, & Folz, 1995; Lattimore, Broner, Sherman, Frisman, & Shafer, 2003; Shafer et al., 2004; Steadman & Naples, 2005; Steadman, Deane, Borum, & Morrissey, 2000; Steadman et al., 1999).
The presented body of literature regarding jail diversion program outcomes provides mixed results. Due to the variations in design and implementation of each program, it is difficult to make generalizable statements regarding these programs. However, there are a few findings that are worth mentioning. First, there is mounting evidence that jail diversion programs can offer positive, safe and viable alternatives to incarceration for individuals with mental illness or co-occurring disorders who have committed misdemeanors or non-violent felonies (Cosden et al., 2003; Lamb et al., 1995; Lamb et al., 1996; Shafer et al., 2004; Steadman & Naples, 2005; Steadman et al., 1999). Second, jail diversion programs can reduce the amount of incarceration time, especially for those arrested for more serious crimes that carry longer sentences (Hoff et al., 1999; Steadman & Naples, 2005; Steadman et al., 1999; Steadman et al., 2000). Third, individuals with substance abuse issues alone have been found to be less likely to be diverted than those with a co-occurring disorder (Hoff et al., 1999; Steadman et al., 1999). Fourth, examination of cost effectiveness typically finds that jail diversion generally results in lower criminal costs and larger treatment costs (Cowell et al., 2004; Steadman & Naples, 2005). Finally, it appears that there is some consistency in characteristics of those diverted. Specifically, older Caucasian females are overrepresented as compared to their numbers within jails (Steadman & Redlich 2006). This pattern actually reflects overall trends seen in criminal justice system decision-making, where lower punitive sanctions are seen among females compared with males and older offenders compared with younger offenders (Steadman et al., 1999).
To date, the trend in jail diversion is to broker services to a variety of community providers. This is consistent with other correctional options such as probation. However, the current study examines a jurisdictional model that is extremely different, in that all services are provided “in-house.”
Chesterfield County is located in the Richmond, Virginia metropolitan region. The locality is known for its willingness to experiment and develop creative community corrections service delivery. In addition to traditional probation and pretrial supervision, Chesterfield County developed the only locally operated Day Reporting Center (DRC) in the Commonwealth of Virginia. This intensive outpatient model, combined with intensive criminal justice intervention, was a promising strategy to reduce criminal behavior and substance use with the offenders (Walker, 2005).
It was during the development and operation of the DRC that officials in the local criminal justice and mental health systems learned the power of true collaboration between traditionally independent systems. In the Chesterfield County DRC model, the criminal justice system provided funding to the local mental health center so that clinicians worked directly with the DRC. This collaboration included probation officers and clinicians staffing cases and developing appropriate treatment plans and policy as a team.
This model of collaboration permeates the culture of the Chesterfield County criminal justice and treatment systems. As the problem of mental health disorders, and more specifically co-occurring disorders, became increasingly apparent in the criminal justice system, the natural organizational reaction was to address the problem in a collaborative manner.
The Need and Program Development
Within the traditional probation and pretrial programs, Chesterfield Community Corrections was experiencing an ever-increasing number of individuals with co-occurring disorders under supervision. The traditional supervision approaches were failing and access to specialized treatment was limited. The issues of those with co-occurring disorders were so complex that traditional probation and pretrial officers did not possess the expertise to identify symptomology or access the network of required resources to intervene in a timely manner. Additionally, the local jail expressed frustration with the cost of managing mentally-ill offenders and the overall supervision challenges inherent with this population. In order to develop an appropriate plan, the jurisdiction needed to understand the scope of the problem. Inspection of the data revealed that failure occurred most often among offenders who were pre-trial detainees for a period of time prior to release under community supervision.
Following the confirmation of the need, management executives in the criminal justice system approached their peers in the treatment system to seek their input about a possible collaboration to better serve offenders and defendants suffering from co-occurring disorders. Due to the organizational structure and philosophical views within the jurisdiction, it was determined that the ability to develop, “sell,” and implement such a program was more viable with members of the criminal justice system leading the charge. In other words, a program under criminal justice control would be perceived as more “in tune” with the needs of the court, the criminal justice community, and the broader community at large.
A small planning group convened to determine the specifics of the program, including the diversion point and type of program. This group consisted of representatives from the criminal justice system (community corrections, prosecutors, and judges) and treatment systems (both substance use and mental health) within the locality. It was established that diversion would be available to any individual who remained in jail and suffered from both substance use and mental health issues. In other words, only individuals who remained under pre-trial detention were eligible; however, the offenders could not have any prior or pending violent charges. Therefore, the program is viewed as a post-booking jail diversion effort.
Due to the success of the DRC, the planning group decided to use the general DRC model to develop an appropriate program for those who suffer with co-occurring disorders. This jail diversion model program, which became known in Chesterfield County as the Dual Treatment Track Program (DTT), is a highly structured and intensive regimen of supervision and treatment. The services offered to defendants included an immediate evaluation by a psychiatrist, medication management, entry into intensive outpatient services, drug testing, and pretrial supervision. With the exception of the psychiatric visits, all of these services were conducted at the DTT. The psychiatric visits were conducted in an adjacent building within the government complex of the jurisdiction.
Like most day reporting center models, the Dual Treatment Track Program worked on a level system (see Figure 1 for the original model). New defendants were required to report for services five to six days per week for up to four hours per day. As defendants demonstrated success, the intensity of services lessened. The progression through the level system was based on several behavioral and clinical standards. Clinical services were largely an integration of substance abuse and mental health treatment. The DTT utilized Moral Reconation Therapy (MRT), cognitive behavioral program, process group, medical and symptom management groups and individual counseling.
The program was conceived and implemented as the clinicians and the pretrial officers working together to manage all cases in a team approach. Although there were role distinctions, neither the criminal justice component nor the clinical treatment component took priority. All facets of the case were managed in unison by an integrated team of professionals.
After the planning group identified the point of diversion and the type of program to best reflect the needs of the jurisdiction, the locality applied for Target Capacity Expansion (TCE) funds sponsored by the Substance Abuse and Mental Health Services Administration (SAMHSA). The locality was awarded the funding to establish the program. The planning committee was then expanded to include a larger group of individuals to serve as a steering committee. This committee included representatives from the Commonwealth’s Attorney’s Office, the local Sheriffs Office, the local community treatment agency, the community corrections office, the Department of Social Services, defense counsel, a consumer representative, and a researcher. The steering committee actively convenes to refine program structures and policies over a three-year period, and currently meets on an as-needed basis.
Program Eligibility
The target population for the program consists of offenders who remain in jail and have a dual diagnosis of substance use and mental health issues. Additional requirements are that the offender be at least 18 years old; have a non-violent criminal history and non-violent current charge; and possess a willingness to receive such services. The program is available to both men and women.
The identification of such individuals involves four stages of screening. The pretrial services officer with Community Corrections conducts stage one, the initial screen. Stage two, the subsequent assessment, is a slightly more extensive evaluation conducted by the DTT pretrial services officer. Stage three, the subsequent evaluation, involves a thorough appraisal of the individual and is conducted by a DTT clinician. And stage four, court decision, is the final determination of acceptance by the judiciary.
The sheer volume of clients screened for appropriate selection into the program was enormous. To illustrate, during a two-year time frame a total of 5,344 assessment events occurred. Most of these events were initial screenings of clients by the pretrial service officer (90%, 4,854 events); followed by the subsequent assessment by a DTT pretrial officer (6%, 298 events), the subsequent evaluation by a DTT clinician (2%, 117 events), and the court hearing by the judiciary (2%, 75 events).
To summarize, over 5,000 offenders were screened at some point with just over 152 clients deemed eligible at a minimum of one time in the assessment process, with 75 cases going to court decisions. Ultimately, 68 clients were admitted into the program during March 2003 to January 2005. Eighty-five percent of the clients were white males (76%), with an average age of 34 years old (range 20–50). Seventy-six percent of the target arrests were for a felony offense, the majority of DTT offenders overall were arrested for a property offense (49%).
Research Methods
This descriptive study seeks to educate the practitioner and academic communities on a variety of items to consider when developing a jail diversion program. The goal is to describe the characteristics and needs of the clients admitted into the program; to examine client adherence to the program; and to consider a variety of offender outcomes. The information contributes to the growing body of literature due to the uniqueness of providing “in-house” services, which differs significantly from any existing model. The study has a number of components relying on a number of data points and methods.
Data Collection
There are three primary data sources used to investigate the clients (person tracking, client progress/status information, and official statistics). The person-tracking portion of the study consists of self-report data derived from face-to face interviews, at baseline (within 7 days of entering the program) and after six months for those clients who agreed to participate in the research. The data collected during the interviews examines the general well being of the individual prior to entering and then up to six months following entry into the DTT program. There were 40 clients who agreed to participate in the baseline interview, with 24 (60 percent) agreeing to complete the 6-month interview. Of the 40 percent who were not retained, 20 percent could not be located (although strict protocol was followed to maximize retention efforts), 12 percent refused to participate, and seven percent (3 individuals) were not approached because the individual threatened staff prior to termination from the program.
The client progress/status involves extracting data from the case files for all of those who entered into the DTT program between March 2003 and January 2005. The information retrieved describes the extent of drug screening, the drugs of choice, and the offender termination status.
Official statistics are examined to identify the outcome measures of recidivism and the number of days spent in jail one-year prior and one-year post entry into the DTT program. Specifically, Virginia Criminal Information Network (VCIN) was used to gather arrest information and local jails provided client-specific data regarding the number of days housed.
Measurement
Client characteristics and needs presents a rich examination of the client’s overall status at program entry. The specific information captures educational status, employment status, additional sources of income (i.e., food stamps, Veteran’s benefits, spouse/partner), ability to manage daily activities, and level of trauma. Specifically, clients were asked during the baseline interview to respond to four statements regarding their current aptitude in the areas of managing day-to-day life, household responsibility, work, and leisure time/recreational activities with a corresponding scale of “no difficulty,” “little difficulty,” “moderate difficulty,” “quite a bit of difficulty,” and “extreme difficulty.” The interview also asked about overall trauma exposure in order to gain a better understanding of the offenders’ experience. This information is typically limited to female offender populations but the research on mental health clients suggests it is necessary to examine the levels of trauma across genders. Specifically, the trauma section was measured with the DC Trauma Collaboration Study Violence and Trauma Screening. The nine statements tap the extent to which a person witnessed a violent event, experienced sexual violence, and experienced physical violence of self and others over both the course of a lifetime and during the past 12 months.
A variety of outcomes is examined. First, substance use during the prior 30 days was measured at baseline and six-months. The clients were asked to report use of alcohol, use of alcohol to intoxication, and use of illegal drugs, with the assumption that clients reported use. Second, the Mental Health Statistics Improvement Program (2000) was administered during the six-month interview to ascertain any self-reported change in mental health symptoms since program entry. The tool asked seven mental status statements and the clients responded using a four-point Likert scale. Third, the type and number of arrests both one-year prior and one-year post DTT entry is captured. And, finally, the number of days an individual spent detained in a local jail (or prison) is examined for one-year prior and one-year post DTT entry.
Limitations
It must be noted that there are several limitations to the study. First, the Targeted Capacity Expansion (TCE) funding of the jail diversion programs determines this venture’s primary focus, which was on designing and implementing services for a specific offender population rather than for a rigorous evaluation. Hence, the only comparison that can be made is between status of the admitted DTT clients one-year before and one-year following entry. There is no comparison group to see if the absence of treatment made a difference. However, examining the larger body of jail diversion literature demonstrates a widespread difficulty in finding an adequate comparison group (e.g., Draine & Solomon, 1999; Cowell et al., 2004; Draine et al., 2005; Hoff et al., 1999; Lamb et al., 1996; Lattimore et al., 2003; Shafer et al., 2004).
Second, the sample size is small, although the primary purpose of this report is descriptive rather than causal. And finally, the sample consists of the first set of clients to enter a new program. As with any new program, there is a period of adjustment until the full implementation and cadre of services are adequately provided.
Client Characteristics and Needs
Although not all individuals agreed to participate in the interview process or were approached1, the information ascertained from the 40 clients who agreed provides valuable insight into their characteristics. Among those who agreed to participate in the face-to-face interview process, a reported 65 percent of the clients attended high school, with 25 percent having graduated from high school. In addition 12 percent went on to college, with 5 percent receiving a college degree.
Furthermore, more than half (58 percent) were employed at the time of arrest, with 33 percent working full-time and 25 percent part-time. Additional reported income sources include SSI (10 percent), food stamps (8 percent), Veteran’s benefits (2 percent), from spouse or partner (12 percent), family or friends (25 percent), or non-legal sources (20 percent).
Examination of the clients’ level of difficulty with several pro-social expectations is presented in Table 1. Most of the clients report “quite a bit” to “extreme difficulty” with managing day-to-day tasks (65 percent) and work (46 percent), while “no difficulty” is reported with regard to household responsibilities (35 percent) and leisure time or recreational activities (33 percent).
Table 2 reveals the level of trauma reported. In general, a high level of trauma has been experienced either directly or indirectly by most individuals over their lifetime. To illustrate, regardless of gender the clients experienced a high level of physical violence and witnessed violent events in general (98 percent and 73 percent respectively). The level of sexual trauma over the lifetime did vary by gender, with 82 percent of females compared to 8 percent of males reporting this type of victimization.
Additionally, for those who reported a specific type of trauma during their lifetime, a follow-up question centering on the prior year was utilized to gauge the extent of current trauma. As one would expect, the overall levels of trauma are reduced; however, the trends seen with lifetime trauma remain. For example, the highest level of trauma experienced overall is physical trauma (36 percent). This information is extremely valuable in understanding the needs of the clients who enter the DTT program. Traditionally, trauma treatment is more commonly offered to the female population, but the information ascertained here speaks to the need of all clients who have entered the DTT program.
Client Progress in the Program
Several items related to a client’s program progress are considered. This captures all 68 clients to enter the program during the time of the research. As discussed earlier the clients are subject to routine substance use testing. Table 3 reports, on average, that 76 drug screens were given to each client, with 31 percent of all clients receiving over 100 screens. The average number of positive screens for each offender is 2.59, with a range between 0 and 10. Cocaine appears to be a drug of choice, with 51 percent reporting positive for this substance.
Table 4 focuses on the outcome of the positive drug screens. On average, clients missed 5.2 screens, with a range of 0–20. Those who tested positive received an average of 2.28 technical violations (range 0–10). Additionally, as a result of the combination of violations, the clients spent an average of 10 days in jail (range 0–46).
Table 5 indicates the termination status of the clients. As shown 69 percent of the clients were terminated. Reasons for termination can range from sanction violations to threatening staff. The program’s absconder rate was 43 percent, which attributed to this termination rate as well. The program quickly learned that acquiring appropriate housing for the offenders was necessary in order to retain them in the program.
Client Outcomes
This section provides information related to changes in substance use, mental health status, arrest, and jail days. It provides a summary picture of any improvements among the clients.
Substance Use
Table 6 examines the use of a substance during the prior 30 days during both the baseline and 6-month interviews. Specifically, during the month prior to entering the DTT program 68 percent of clients report using alcohol compared to 25 percent at the 6-month comparison. Likewise, 53 percent of clients consumed alcohol to the point of intoxication immediately prior to entering DTT, compared to 16 percent at the 6-month interview point. And finally, 88 percent of the clients used an illegal drug during the 30 days prior to entering DTT compared with 16 percent at the 6-month follow-up period. Additionally, the average number of days (mean) decreased at all time points.
Mental Health Status
Table 7 reports the percentage of offenders who “agree” or “strongly agree” with the listed statement. Overall, the program results are favorable in that each of the categories shows improvement. Specifically, during the 6-month follow-up interview at least half of all clients agreed to each statement, with 60 percent or greater agreeing that they “deal more effectively with daily problems,” “better able to control my life,” “getting along better with my family,” and “do better in social situations.” These are all marked improvements in one’s quality of life that can have direct benefits on a pro-social lifestyle. In general, the self-report data reflect overall improvement among the clients regarding overall mental health status and lower use and intensity of use of alcohol and drugs.
Arrest and Jail Days
A look at the entire sample
As shown in Table 8, the likelihood of arrest is relatively unchanged one year prior and one year post entry into DTT. However, a closer look at the types and seriousness of the offenses committed has shown enhancement after entry into the DTT program (Table 9). Specifically, the table reflects a reduction in the percent of clients arrested at the felony level, with 76 percent of the arrests for the target offense (the offense that initiated DTT) and 49 percent for a felony arrest one year prior, compared to 41 percent at the felony level during the one-year after entry into the DTT program. Likewise, the percent of minor arrests increases during the follow-up period; from 32 percent for target offense and 44 percent for one year prior to 61 percent during the follow-up period. This is interpreted to signify that although the clients are still involved in the criminal justice system, the types of offenses committed appear to be less serious. The category of “minor offense” most commonly is associated with failure to appear and technical violations (e.g., drug screen failure, not reporting). Overall, when we examined the types of offenses arrested for during the follow-up period, every category decreased except for the minor offense category.
Table 10 presents the average (mean) number of jail days for the three time points examined: target offense, one-year prior, and one-year after entry into the DTT. Clients spent an average of 44 days in jail for the incident that initiated entry into DTT (target arrest) and an average of 64 days in jail for the year prior to entering the program (this includes the target offense as well). In comparison, the total average number of days in jail during the follow-up period is 71 (an average of 36 days for a new arrest and an average of 35 days for a sanction or sentence due to the target incident). This information is based only on the 55 clients who had a complete one-year follow-up period when the data were extracted. When examining the sheer average number of days there is a slight increase in the total number of jail days before entry (64 days) and after entry (71 days).
A Comparison of Successful Participants to Terminated Participants
This section investigates the clients who received a stronger dosage of treatment (successful release) to those who did not (terminated clients). Although this is not a strong methodological comparison, such an examination assists with locality-specific feedback on the usefulness of the program. Tables 11–13 reflect information pertaining to arrest and jail time served based on release status. Table 11 examines the target offense information and reveals that, regardless of release status, the majority of offenders were arrested on a felony crime that was classified as a property or minor offense.
Table 12 shows the likelihood of arrest, charge level and type of charge one-year prior (excluding the target offense) and one-year after. Some interesting findings appear. First, all of the successful clients were involved with the criminal justice system at least once (22.23 v. 52.72), and fewer number of days in jail overall during the one-year follow-up period (23.11 v. 93.79), with all of the differences being statistically significant. The one item that is not statistically significant indicates that clients successfully released from the program spent more time in jail during the prior year than those terminated (92.60 v. 49.81).
When taken together, on face value, it appears that the program has an impact on those who may be at an immediate higher risk level. The factors indicating this are the slightly higher rate of arrest during one-year prior to entering the DTT program (excluding the target offense) combined with an increased number of days in jail one-year prior to DTT entry compared to those terminated. Again, this needs to be viewed cautiously, because it only covers a one-year time frame and there is no control group; for instance, combined with the fact that we do not know the status of their family support system or housing issues, to name a few, compared to those terminated. In addition, there were a number of individuals in the “terminated” group who could be considered high risk with this limited definition of the term. At any rate, assuming this is true it would be consistent with the larger literature on correctional intervention that suggests that highly structured programs have the best impact on high-risk offenders and a potential negative risk on low-risk offenders (Andrews and Bonta 1994; Andrews, Bonta, and Hoge 1990). Additionally, the finding is consistent with previous research that indicates that reduction in jail time is most apparent in the following year for those individuals charged with more serious offenses (e.g., Hoff et al., 1999; Steadman & Naples, 2005).
Overall Summary of Results
A number of points are revealed by this sample of clients: clients have a significant trauma history, clients show signs of mental health improvement, clients report a reduction in substance use, the program appears to maintain a level of safety in the community as reflected by the less serious charges after entering into the program, and the program shows strong improvements among clients who are successfully released compared with those who are terminated. Discussion
Fifty years ago, Sutherland observed, “For a century or more, two rival policies have been used in criminal justice. One is the punitive policy; the other is the treatment policy.” (Sutherland, 1950 as cited in Grudzinskas et al., 2005, p.278). This statement embraces the response to the deinstitutionalization of mentally ill offenders. Jail diversion programs for the mentally ill attempt to balance the individuals’ unique needs while addressing the system’s desires for punishment in order to restore the disparity caused by the specific offense. This response requires inter-agency collaboration among systems that traditionally have philosophical divergence. For this reason, the literature overwhelmingly stresses the need for active and continual conversations between the criminal justice system, treatment system, and the community at-large (Shafer et al., 2004; Steadman et al., 1995; 1999). Chesterfield County is a locality that has productively learned to meet this challenge.
There are a number of situations that illustrate that the locality’s success is due to inter-agency collaboration. To illustrate, during the infancy of the program implementation it became readily apparent that a high percentage of clients were being terminated. Closer inspection of the reason for termination revealed a high rate of absconding among clients who had unstable housing situations. One response could have been to alter the entry requirements to include stable housing, though that would have eliminated a number of potential clients. However, the criminal justice leaders approached the Community Service Board to investigate options for housing. The conversation resulted in obtaining temporary housing for clients on an emergency basis. This has stabilized the issue and reduced the rate of absconding in the last year of data collection.
A second example of true collaboration arose with the need for aftercare. The larger body of correctional effectiveness research stresses the importance of including an aftercare component in order to meet the changing needs of the offender (Gendreau, 1996). The locality quickly understood that the need for continued services was enormous after the first few clients graduated. Through collaborative efforts between the treatment and criminal justice system, two responses were developed: continued psychiatric and substance use services upon completion of the program and continued DTT sessions to clients on an as-needed basis. Therefore, towards the final stages of the funding, an aftercare or “booster” program was developed.
We suspect that a primary reason for the high degree of success with collaboration is due to the program structure. The locality was innovative in utilizing a Day Reporting model that essentially provides all services on-site and resulted in a more consistent method of service delivery across clients. Such an approach reduces the potential for system fragmentation that is strongly reflected in the literature as a primary indicator for program success or failure. In other words, as indicated by Grudzinskas et al. (2005), jail diversion programs must have appropriate and effective linkages between the courts and service providers. The model presented here saw the linkages working in unison on a daily basis for each client.
A paramount concern of jail diversion programs for mentally ill offenders is public safety (Steadman et al., 1999). The data suggest that public safety was maintained, as illustrated by a change in the types of incidents the clients were involved in and overall improvement in mental health symptoms. Additionally, a reduction in both the type and number of criminal justice system contacts was revealed among clients who remained in treatment for a longer period of time.
While this assessment adds to the body of knowledge on the impact of redirecting offenders suffering from co-occurring disorders, a number of issues brought forward in the literature must be echoed as concerns regarding the state of jail diversion programs. The treatment of mentally ill offenders must be multi-faceted since their needs are great. The criminal justice system and the community at-large hold unrealistic expectations for such programs. Although the diversion programs differ in length, most are a year or less, a limited time frame in which to satisfy the varying needs of the clients that include but are not limited to, stabilization of mental health needs, treatment of general medical needs, poverty, substance use, joblessness, and homelessness (Grudzinskas et al., 2005). Additionally, it must be recognized that the creation of jail diversion programs is proliferating at a rate that far exceeds the knowledge basis concerning policies, procedures, and appropriate clientele (Steadman & Redlich, 2006). Future Research
The explosion of jail diversion programs reflects the ease of diverting mental health clients from the criminal justice system, however; the essential question now turns to the appropriateness of services provided. Future research will be challenged to answer this question. The complexity of the answer is muddied by the varying points at which offenders are diverted, community structure and needs, admission criteria for entry into a diversion program, and a lack of standardization across programs. Furthermore, the research has a number of methodological challenges. The literature points to the difficulty in identifying appropriate comparison groups (e.g., Draine & Solomon, 1999; Draine et al., 2005; Hoff et al., 1999) and sufficient sample sizes (Steadman et al., 1999; Steadman & Naples, 2005 ) due to the highly select admission criteria.
The balance between rewards and punishments within a program and how this may impact the likelihood of engaging in criminal behavior should be examined. In essence, perceived deterrence theory suggests that offenders are less likely to engage in criminal activities when the certainty of detection is high and the recognition for accomplishments is immense (e.g., Akers, 1990; Gibbs, 1975; Tittle, 1980; Zimring & Hawkins, 1973). While this concept has been tested among drug court participants (see Marlowe, Festinger, Lee, & Patapis, 2005), it has not been addressed by the jail diversion literature. Examination should reveal the validity in the types of sanctioning system, rewards systems, and the overall impact of the balance on offender outcomes.
___________________________________________________________________________
* This project was supported by Grant Number 07-A5260OT07 awarded by Substance Abuse & Mental Health Services Administration, Targeted Capacity Expansion, U. S. Department of Health and Human Services. Points of view in this document are those of the authors and do not necessarily represent the official position or policies of the U. S. Department of Health and Human Services.
** Please direct all inquiries to Dr. Jill A. Gordon, 923 W. Franklin St., Richmond, VA 2328.
*** The authors would like to thank the staff (especially Glen Peterson and Kristina Bryant) and clients in the Dual Treatment Track Program located in Chesterfield County, Virginia. Additional thanks to the TAPA Center staff for assisting with the data management.