Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 40
Parole, Desistance from Crime, and Community Integration 4 Services and Programs for Releasees Recent research on desistance from crime has generated a body of knowledge that examines the underlying conditions that lead to less frequent or a declining rate toward zero offending; yet programs and services for parolees are, in general, not based on these findings, but are instead rooted in the research on individual behavioral change. Farrall and Maruna (2004) have noted that only recently have there been attempts to link findings from desistance studies with evaluations of offender management programs and policies. Part of the problem is that both theory development and research on the mechanisms underlying desistance are limited. We do not know, for example, how individual change and social circumstances such as marriage or work interact to produce desistance. Because the committee believes these unexplored issues are important, we try to provide something of a baseline of information about these linkages by organizing parts of this chapter on programs and services for releases in a desistance framework. For example, in a longitudinal study on probation and desistance (Farrall, 2002), the researchers found employment and family relationship experiences were more critical to successful desistance than differences in probation practice. The desistance literature on parolees has found similar results, though the reasons for them are poorly understood. Using a desistance framework makes it possible to focus on the purpose of the intervention rather than on offending and allows consideration of the broader context required to support behavioral change (Farrall and Maruna, 2004). Although the disconnect between these literatures may make this framework seem forced at times, we believe it
OCR for page 41
Parole, Desistance from Crime, and Community Integration fosters the development of a more coherent structure for future research and policy development. There is an overlap between conditions that promote desistance and the individual effects of change but it is not clear where that intersection lies. Does individual change provide a basis for employment and a stable marriage, or do a stable marriage and a job provide the context for individual change? Intervention research has shown that the most successful programs fostering individual change and leading to desistance are those that start in prison and then continue in the community setting once an individual is released. The chapter begins with a review of research findings on interventions that are offered either before or after release from prison, organized in a desistance framework that includes education and employment, marriage, drug treatment, and individual change. This is followed by a section on current innovations in reentry programming, including prerelease planning and the consequences of early failure. The third section considers available services and their effects, including physical and mental health services, mentoring programs, and best practices. RESEARCH ON PROGRAMS Education and Employment In the United States, adult corrections facilities have a long history of providing education and vocational training as part of the rehabilitation process (Piehl, 1998; Gaes et al., 1999), based on the belief that improving education and job skills will promote desistance. However, participation in these programs has been declining since the early 1990s: among soon-to-be-released prisoners in 1991, 42 percent reported participating in education programs and 31 percent in vocational programs; in 1997 the figures were 35 percent and 27 percent, respectively (Lynch and Sabol, 2001). The reasons for these declines include the rapid growth in the prison population, decreased state and federal funding for in-prison programs, the frequent transfers of prisoners from one facility to another, and greater interest in short-term programs, such as substance abuse and cognitive-behavioral programs (Lawrence et al., 2002). Some studies show that recidivism rates are significantly lower for releasees with more education (MacKenzie, 2006; Adams et al., 1994; Boudin, 1993; Harer, 1995; Stillman, 1999, Fabelo, 2000). Moreover, comprehensive reviews of dozens of individual program evaluations generally conclude that adult academic and vocational programs lead to reduced recidivism (MacKenzie, 2006) and increased employment of 5-10 percent (Gerber and Fritsch, 1994; Gaes et al., 1999; Cullen and Gendreau, 2000;
OCR for page 42
Parole, Desistance from Crime, and Community Integration Wilson et al., 2000; Aos et al., 2006). However, the majority of the evaluations are of poor quality, and a close examination of their methodological problems reduces confidence in their results (MacKenzie, 2006; Wilson et al., 2000) For correctional education programming to be successful, it must be part of a systematic approach that includes programs for employability, social skills training, and other specialized programming (Taxman, 1998). Best-practice correctional education programs are both carefully tailored to individual prisoners and related to vocational and job skills training. Education and job training for prisoners who were low earners are most successful when they provide workers with credentials that meet private-sector demands. Programs that provide training, a range of services and supports, incentives, and access to better employers work well, especially when there are strong incentives for releasees to get jobs (Holzer and Martinson, 2005; Visher and Courtney, 2006). To be most effective, inmate screening, needs assessment, and the provision of services need to be integrated, but this approach may run counter to other institutional priorities. In many systems, security classification takes precedence over other activities, which may affect the organization of and availability of services. Many prisoners would like to enroll in education and training programs but slots are not available or they are not eligible because of their security status or short sentence length. If the highest need prisoners are also the highest risk offenders, it might make sense to shift some programming resources to higher security institutions where such prisoners are concentrated (Logan, 1993). As noted above, work is a primary feature of successful reintegration and desistance (Sampson and Laub, 1990, 1993; Nagin and Waldfogel, 1998). The time spent and connections made at work probably serve as informal social controls that prevent criminal behavior. Having a job, especially a good job, reduces the economic incentive for criminal behavior. For example, using data from the 1980 National Longitudinal Survey of Youth, economist Jeffrey Grogger (1998) has estimated that the elasticity of crime participation with respect to wages is −1.0, two and a half times higher than the elasticity provided by incarceration.1 Specifically he found that a 10 percent increase in wages would reduce crime participation by 6−9 percent. His estimates suggest that young men’s behavior is very responsive to price incentives and that falling real wages for youth may have been partially responsible for the rise in youth crime during the 1980s and early 1990s. Finding employment is one of the most pressing problems that releasees 1 Elasticity is the ratio of the proportional change in one variable with respect to proportional change in another variable.
OCR for page 43
Parole, Desistance from Crime, and Community Integration face. Although two-thirds of former prisoners were working prior to their imprisonment (Maguire, 1994), their educational level, work experience, and skills are well below national averages for the general population (Andrews and Bonta, 1994; Petersilia, 2005). Moreover, the stigma and legal restrictions associated with incarceration often make it difficult for ex-prisoners to secure employment (Holzer et al., 2002; Bushway and Reuter, 2002). When releasees do find jobs, they tend to earn less than employees with similar background characteristics who have not been in prison (Bushway and Reuter, 2002). Thus, research supports a strong programmatic emphasis on increasing prisoners’ and releasees’ employability, through skills training, job readiness, and, possibly, work-release programs during incarceration and after release. The U.S. Department of Labor (DOL) has a long history of evaluating community-based work programs for former prisoners and people with criminal records, beginning with several large evaluations of job training and financial support for former prisoners in the 1970s. The LIFE and TARP studies conducted in the 1970s are well known (see Rossi, Berk, and Lenihan, 1980). These programs offered ex-offenders varying levels of unemployment compensation and job placement assistance. Random assignment studies in Texas, Georgia, and Baltimore found that income supports reduced property crimes although they also created a disincentive for ex-offenders to find employment (Berk et al., 1980). Other DOL-funded job training initiatives that included ex-offenders were the National Supported Work Demonstration, Job Corps, JobStart, and the Job Training Partnership Act. More recently, DOL has funded faith-or other community-based organizations in 30 sites under the Prisoner Reentry Initiative for employment services and job placements, specifically for clients with nonviolent histories (with varying definitions by site) who are under the supervision of the criminal justice system. The most recent random assignment study, initiated in 1994, evaluated the Opportunity to Succeed Program, which delivered employment services within a set of comprehensive services for drug-using former prisoners. The study found that participants were more likely to be employed full time in the year after release, and they reported less drug use; however, self-reports of arrests and official record measures of recidivism showed no differences between participants and controls (Rossman and Roman, 2003). A meta-analysis that examined the effects of employment training and job assistance in the community for ex-offenders concluded that such programs are responsible for a modest, but significant, 5 percent reduction in recidivism (Aos et al., 2006). However, another meta-analysis, using a very similar, but not identical set of studies and methods, concluded that community-based employment programs do not significantly reduce recidi-
OCR for page 44
Parole, Desistance from Crime, and Community Integration vism for ex-offenders (Visher, Winterfield, and Coggeshall, 2005).2 Thus, although work programs can have a significant effect on the employment and recidivism rates of male releasees (Bushway and Reuter, 2002), the effect sizes may be small. Current job assistance and training programs for former prisoners—such as the Center for Employment Opportunities (New York), Safer Foundation (Chicago), and Project Rio (Texas)—are more comprehensive than earlier employment programs for ex-offenders, incorporating other transitional services and reentry support though maintaining a primary focus on job placement (Buck, 2000). The effects of these comprehensive, employment-focused programs on ex-offenders’ employment and recidivism rates are not yet known; several well-designed evaluations are under way. Marriage and Family Support Programs For the general population, research has documented the benefits of marriage for adults and their children (see Chapter 2; Waite and Gallagher, 2000; Lerman, 2002). A recent study concluded that marriage is a “potentially transformative institution that may assist in promoting desistance from criminal behavior” (Sampson et al., 2006, p. 500). Thus, it makes sense to ask whether former prisoners would benefit from marriage and family support programs. Most people in prisons are men, and most of them are fathers. A majority of state and federal male prisoners have at least one child under 18, on average 8 years old (Mumola, 2000). About one-third of the fathers lived with their children just before arrest, and most also lived with the child’s mother. Some who did not live with their children saw them regularly and contributed to their upbringing (Hairston, 2002). Over one-half of incarcerated parents have been married, and about 23 percent are married when they are in prison (Mumola, 2000). In addition to the role of marriage in contributing to desistance, a significant body of research shows positive effects of family support on a variety of reentry outcomes. Greater contact with family during incarceration (by mail, phone, or in-person visits) is associated with lower recidivism rates (Adams and Fischer, 1976; Glaser, 1969; Hairston, 2002). Prisoners with close family ties have lower recidivism rates than those without such attachments (La Vigne et al., 2004; Sullivan et al., 2002). Strong family attachments may keep ex-offenders away from peers who encourage criminal behavior (Warr, 1998). Emotional and financial family support is associ- 2 The reason for these differences is not clear, but is probably due to a slightly different mix of studies. In addition, Visher and her colleagues included only randomized clinical trials in their analysis, while Amos and his colleagues included matched designs.
OCR for page 45
Parole, Desistance from Crime, and Community Integration ated with better employment and the avoidance of illegal substance use (La Vigne et al., 2004; Visher et al. 2003; Nelson et al., 1999; Sullivan et al., 2002). Fathers who report strong attachments to their children have higher employment and lower depression rates after release (Visher et al., 2007). Family support is associated with better results concerning depression and prosocial identity (Ekland-Olson et al., 1983; Uggen, Manza, and Behrens, 2004; Laub and Sampson, 2003). Prison-based programs generally offer parenting education, counseling and support groups, and services to facilitate visitation. Community-based programs usually include counseling, mentoring, assistance with family reunification and rebuilding, continuing parenting education, and family support groups. However, support services specifically focused on marriage and family for prisoners and former prisoners are limited and generally have not directly addressed nonmarital couple relationships. Some programs provide family strengthening services that may affect couples, but indirectly, by attempting to help fathers to be less of a burden to their families, to strengthen the father-child bond, and to avoid behaviors that stress relationships. Other programs focus on family relationships more generally. Programs funded as part of the federal Serious and Violent Offender Reentry Initiative (SVORI) offer general family strengthening services, but only 8 of the 50 adult programs ranked it as one of their top three priorities (see http://www.svori-evaluation.org [accessed June 2007]). Unfortunately, few intervention studies have examined the role of marriage and family support programs in desistance. An exception is La Bodega de Familia, a well-known family-focused reentry program for former prisoners developed in New York City. The program uses comprehensive family case management as a mechanism for working with releasees and people on probation, families, and the community to create a web of support. Comprehensive family case management uses a strengths-based, client-driven approach to help clients and family members navigate service delivery systems and agencies to access treatment and services, maintain employment, tap existing networks for support, and create long-term family well-being and community safety (Shapiro, 2003). An evaluation of the program demonstrated that focusing on returning former prisoners together with their families, and emphasizing the strengths of each unit in addition to addressing the deficits, successfully decreased illegal drug use without additional treatment, reduced new arrests, and increased overall physical and mental health (Sullivan et al., 2002). One negative outcome was that the clients (former prisoners and their families) reported increased conflict in their family relationships. In 2006 the Department of Health and Human Services initiated a demonstration program designed to promote two-parent families and marriage, with a special focus on incarcerated fathers. Ten sites were chosen
OCR for page 46
Parole, Desistance from Crime, and Community Integration for an in-depth process and outcome evaluation; where possible, couples will be randomly assigned to the demonstration program or another set of services (see http://www.acf.hhs.gov/programs/ofa/hmabstracts/summary.htm [accessed March 2006]). When the evaluation is completed, this highly innovative program may offer insights into the relationship between marriage-strengthening programs and successful reentry of releasees. However, given the research on the important role of nonmarital family support after release, demonstration programs are also needed to identify the types of programs and services that would be most effective in promoting desistance. Behavior Management The desistance literature points to individual motivation to change as a key correlate of reduced offending (Maruna, 2001; Laub and Sampson, 2001). Behavior management therapies in correctional programming seek to help individuals understand the basis for their negative behavior and correct their faulty perceptions of themselves, their environment, or both. These therapies provide individuals with skills they can use to monitor their thoughts and correct their behaviors in daily situations, ultimately leading to significant changes in behavior. Behavior management is an umbrella approach that includes contingency management (relapse prevention), social learning, and moral reasoning techniques. In relapse prevention, clients are guided to evaluate situations that may lead to a relapse of illegal behavior and then plan for how to either avoid or cope with them effectively. The most widely recognized behavior management approach to change is cognitive-behavioral therapy or treatment. In the criminal justice field, it is a problem-focused method designed to help people identify the dysfunctional beliefs, thoughts, and patterns of behavior that contribute to their problems and provide them with the skills they need to modify those behaviors, prevent relapse into those behaviors, and maintain successful behavior (Taxman, 2006). This approach also addresses individuals’ readiness and motivation to change by engaging them in self-assessment and the development of treatment goals. The underlying theory of cognitive-behavioral treatment is that behavior is learned, and mechanisms for learning new behaviors must be in place in order for the environment to be part of the change process (Taxman, 2006). It combines two kinds of approaches—cognitive therapy and behavioral therapy. Cognitive therapy concentrates on thoughts, assumptions, and beliefs. Through cognitive therapy, individuals are encouraged to recognize and change faulty or maladaptive thinking patterns that lead to negative behavior. Cognitive therapy enables individuals to gain control over inappropriate repetitive thoughts that often feed or trigger various presenting problems
OCR for page 47
Parole, Desistance from Crime, and Community Integration (Beck, 1995). Behavior therapy concentrates on external factors, addressing the specific actions and environments that either change or maintain behaviors (Skinner, 1974; Bandura, 1977). For instance, people who are trying to stop smoking are often encouraged to change their routine habits: instead of having their daily coffee when waking—which may trigger the urge to have a cigarette—they are encouraged to take a morning walk. Replacing negative behaviors with positive behaviors is a well-known strategy to help change behaviors, particularly when the new behavior is reinforced. Cognitive-behavioral approaches have often been used in correctional programs that target substance use and its associated problems. The theory is that substance use is a learned behavior that is initiated and maintained in the context of environmental factors (Waldron and Kaminer, 2004). Programs built on this premise concentrate on helping people anticipate and avoid high-risk situations as a means to facilitate abstinence. Techniques used to facilitate change include identifying the circumstances surrounding use, learning strategies to manage urges and cravings, and remembering to engage in positive behaviors (Kaminer, 2004). The combination of cognitive therapy and behavioral therapy has been successful, especially among young people, in forestalling the onset, ameliorating the severity, and diverting the long-term consequences of behaviors associated with delinquency, crime, and violence. Research consistently shows that cognitive-behavioral therapy is associated with significant and clinically meaningful positive changes, particularly when therapy is provided by experienced practitioners (Landenberger and Lipsey, 2006; Waldron and Kaminer, 2004). Four problem behaviors have been particularly amenable to change with this approach: (1) violence and criminality, (2) substance use and abuse, (3) teenage pregnancy and risky sexual behaviors, and (4) school failure. Cognitive-behavioral therapy has been successfully applied across settings (e.g., schools, support groups, prisons, treatment agencies, community-based organizations, and churches) and across ages and roles (e.g., students, parents, and teachers). It has also been shown to be relevant to people with differing abilities and from diverse backgrounds. Meta-analyses of programs designed for criminal offenders have shown cognitive-behavioral programs to be very effective in reducing recidivism rates, most notably among higher risk, hard-to-reach offenders (Little, 2005; Lipsey et al., 2001; Landenberger and Lipsey, 2006).3 On average, the therapy reduced the recidivism rates of a general offender population by 27 percent. Interestingly, in a study of prisoners who received cognitive-behavioral treatment either through participation in programs with high-quality implementation (e.g., research and demonstration projects) or in 3 The Landenberger and Lipsky (2006) meta-analysis is particularly notable because it included only studies that used random control designs.
OCR for page 48
Parole, Desistance from Crime, and Community Integration routine correctional practice, those in the former group experienced higher reductions in recidivism rates after release (49% on average) than the latter group (11% on average). The largest effects on recidivism were seen for higher risk offenders who received treatment from providers with at least moderate training in cognitive-behavioral therapy and as part of research and demonstration projects. For this group, recidivism rates were reduced by nearly 60 percent. Overall, however, Landenberger and Lipsey (2006) found no difference in the effectiveness of various “brand name” programs of cognitive-behavioral therapy in comparison with generic forms of the therapy. In their latest review of the evidence from 291 rigorous evaluations of adult corrections programs throughout the United States and other English-speaking countries, Aos et al. (2006) found that programs for the general offender population that use cognitive-behavioral treatment significantly reduced recidivism by an average of 8.2 percent compared with treatment as usual.4 That is, without the cognitive behavioral approach 49 percent of offenders will recidivate, and with cognitive behavioral treatment, 45 percent of offenders will recidivate, a reduction of 8.2 percent in the recidivism rate. Prison-based drug treatment programs that use cognitive-behavioral approaches have been found to reduce recidivism by nearly 7 percent (Aos et al., 2006, Exhibit 1). Finally, cognitive-behavioral treatment programs have also been shown to be cost-effective, yielding $2.54 to $11.48 for every program dollar invested in comparison with punishment-oriented interventions, which have yielded returns of only 50 to 75 cents for every program dollar (Aos et al., 2001). Treatment for Drug-Involved Offenders Illegal drugs are related to crime in multiple ways, and the connection between drug use and crime has been well established in the research literature. According to the Bureau of Justice Statistics (1998), 31 percent of crime victims reported that their assailants were under the influence of drugs or alcohol. In 2005 approximately 20 percent of state prison inmates and 55 percent of federal prison inmates were incarcerated for a drug related offense, and among state prisoners, three-fourths had some type of involvement with drug or alcohol abuse prior to their incarceration. Substance use among former prisoners presents significant challenges to their reentry. Only 15 percent of offenders involved with drugs and alcohol abuse received treatment in prison (Karburg and James, 2005). Serious drug use problems—involving daily or weekly use—likely affects about 4 As with the other meta-analysis, there were no significant differences in outcome between the “brand name” and generic forms of treatment.
OCR for page 49
Parole, Desistance from Crime, and Community Integration one-third to one-half of all drug-abusing offenders. Forty-one percent of men released from Maryland prisons and returning to Baltimore reported using heroin daily in the 6 months before their incarceration; about 33 percent of men released from Illinois prisons reported weekly illicit drug use before incarceration (Visher et al., 2004; Visher et al., 2003). Because chronic drug abuse alters the brain chemistry in people addicted to heroin, cocaine, and even nicotine, they are at higher risk of relapse to use even after long periods of abstinence (National Institute on Drug Abuse, 2006). In addition, releasees who are substance users also have high rates of other mental health problems, so they may need integrated drug and psychiatric treatment (Compton et al., 2003). Such treatment may need to be relatively prolonged, because research has shown that multiple episodes of treatment may be required to help substance users maintain abstinence over time (National Institute on Drug Abuse, 2006). A comprehensive assessment is the first step in developing a treatment regimen, and tailoring individualized services is an important component of drug abuse treatment for criminal justice populations (National Institute on Drug Abuse, 2006). The three primary treatment approaches for drug-abusing offenders are therapeutic communities, outpatient treatment, and “12-step” programs. Therapeutic communities are intensive programs that typically have stand-alone custodial units and use a hierarchical model with treatment stages that reflect increased levels of personal and social responsibility. Outpatient treatment involves counseling by certified drug treatment specialists and often includes pharmacotherapy. And 12-step programs are organized by peers, such as Alcoholics Anonymous (AA) and Narcotics Anonymous (NA). It is widely believed that in-prison drug treatment for offenders leads to reductions in drug use and subsequent criminal behavior and to better outcomes in other areas, such as employment. However, there are few rigorous evaluations of in-prison drug treatment programs. Evaluations of the three most well-known model programs—Key/Crest (Delaware), the Amity therapeutic prison (California), and Kyle New Vision (Texas)—suffer from several methodological shortcomings, including noncomparable treatment and control groups, inadequate controls for selection bias, and poor outcome measures (Pearson and Lipton, 1999; Gaes et al., 1999). This lack of high-quality evaluations is especially troubling in light of congressional appropriations of more than $450 million to states in the last decade to establish residential substance abuse treatment programs in correctional institutions. Unfortunately, even for the small group of individuals who have access to and take advantage of treatment programs in prison, available evidence suggests that fewer continue to receive such community-based treatment after release (Winterfield and Castro, 2005). In an analysis that examined
OCR for page 50
Parole, Desistance from Crime, and Community Integration the extent to which correctional treatment was matched to individual needs, a recent study of prisoners with drug problems found that only 58 percent of those who either had objective indicators of serious drug use or indicated a need for drug treatment received in-prison drug treatment (including Alcoholics Anonymous or Narcotics Anonymous) (Winterfield and Castro, 2005). A recent meta-analysis of correctional programs concluded that drug-treatment programs reduce recidivism: the reduction is about 5 percent for releasees who received only in-prison treatment and 12 percent for releasees who received both in-prison and community treatment (Aos et al., 2006). An earlier study showed that some offenders may benefit from diversion into treatment, but others may require intensive monitoring with the threat of criminal justice sanctions (Marlowe, 2003). Offenders who complete prison-based treatment and continue with treatment in the community have the best outcomes (National Institute on Drug Abuse, 2006; Gaes et al., 1999; Harrison and Beck, 2006). Continuing drug abuse treatment in the community is believed to be necessary to help new releasees deal with problems that only become salient at reentry, such as learning to handle situations that could lead to relapse, learning how to live drug free, and developing a drug-free peer support network. Moreover, better outcomes are also associated with treatment that lasts longer than 90 days, and studies have shown that legal pressure can improve retention rates (National Institute on Drug Abuse, 2006). Increasingly, medications are an important part of treatment for serious drug abusers with long histories; those medications include methadone, buprenorphine, topiramate, and naltrexone (National Institute on Drug Abuse, 2006; Witten, 2006). Although these medications may not be appropriate for all drug-using offenders, the criminal justice system has been slow to embrace these approaches; and most parolees and other ex-offenders who are under criminal justice supervision in the community are not offered this type of treatment.5 Yet postrelease monitoring of drug use through urinalysis or other objective methods, as part of criminal justice supervision, has been found to reduce both relapses of drug use and criminal behavior (Taxman, 2006). Ongoing coordination between treatment providers and courts or supervision officers is required to address the needs of the drug-abusing releasees (Marlowe, 2003), but collaboration and communication between the treatment and community criminal justice supervision systems have been limited to date. In summary, although sustained abstinence is associated with substantial reductions in crime (perhaps 50 percent or more), only a small percentage of drug-abusing offenders receive appropriate treatment for the length 5 A few adult drug courts do include medication-based treatment (see below and Chapter 5).
OCR for page 52
Parole, Desistance from Crime, and Community Integration (Jenkins, Griswold, and Gillespie, 1993; see http://nicic.org/Library/011018 [accessed September 2007]). The importance of developing these reentry plans to enable people to succeed once they have been released has been described by Petersilia (2003) and Healy (1999). Unfortunately, such individualized reentry plans are not yet standard operating practice prior to release, largely because of lack of resources to fund staff to prepare them. Early Failure and Its Consequences When they leave prison, releasees most immediately need transitional services. Transitional services include photo identifications, appropriate clothes, housing, access to transportation, and, if they are eligible, getting signed up for public assistance. Unfortunately, these kinds of immediate needs are often not addressed before release, and it falls to family and friends to help arrange them for new releasees. In fact, most postrelease programs are not available to releasees in the first few days after release. This lack of immediately available services has been shown in recent research to have serious consequences: releasees are at high risk of dying or being rearrested within the first few days and weeks after release. In a recent special article for the New England Journal of Medicine, Binswanger and colleagues (2007) found that former prison inmates were at relatively high risk of death after release, particularly during the first 2 weeks. The article reports on a retrospective cohort study of all inmates released from the Washington State Department of Corrections from July 1999 through December 2003. The risk of death during the first 2 weeks after release, adjusted for age, sex, and race, was 12.7 times that of other state residents. The leading causes of death for former inmates were drug overdose, cardiovascular disease, homicide, and suicide, which are different from the leading causes of death in the state’s general population and in the prison population.6 It has been well established that a large proportion of parolees who return to prison fail in the first weeks and months after their release (Maltz, 1984; Schmidt and Witte, 1988; Ezell, 2007; Haapanen et al., 2007). In a recent analysis, Rosenfeld and colleagues, using data from the Bureau of Justice Statistics, calculated arrest probabilities by month for each of the 36 months postrelease for a sample of 243,334 released prisoners in 13 states. Roughly two-thirds of prison releasees are arrested at least once during this 6 The study found that of 30,237 released inmates, 443 died during an average follow-up period of 1.9 years. The overall mortality rate was 777 deaths per 100,000 person-years. The adjusted risk of death among former inmates was 3.5 times that for other state residents. During the first 2 weeks after release, the risk of death among former inmates was 12.7 times that for other state residents, with a markedly elevated relative risk of death from drug overdose, a shocking 129 times that of the general population.
OCR for page 53
Parole, Desistance from Crime, and Community Integration 3-year postrelease period (Langan and Levin, 2002). In one analysis, the researchers assumed that the entire sample was eligible for arrest in any given month. A second analysis adjusts that probability by subtracting out persons who were in jail, in prison, or dead during the month and therefore not eligible for arrest. In both cases, the probability of arrest declines with months out of prison: that probability during the first month out of prison is roughly double that during the 15th month, and it then stabilizes through the end of the 3-year period (see Figure 4-1). The probability of arrest after release from prison differs by type of crime. Prison releasees arrested for property or drug offenses are more likely to be arrested early in the postrelease period than those arrested for violent offenses. This pattern is illustrated in Figure 4-2, which shows the probability of arrest for releases arrested for property, drug, and violent crimes. The arrest probabilities have been adjusted for time off the street. Although risk for arrest declines over time for all three crime types, a much steeper decline occurs for property and drug offenders, whose arrest risk drops by nearly 50 percent between the 1st and 15th month after release; for violent offenders, the decline is only about 20 percent from the 1st to the 15th month out of prison. Given these data, it is difficult to overstate the importance for parolees and their communities of access to FIGURE 4-1 Probability of arrest for a new crime after release from prison. NOTE: Probabilities adjusted for time off the street. SOURCE: Richard Rosenfeld, personal communication, January 10, 2007.
OCR for page 54
Parole, Desistance from Crime, and Community Integration FIGURE 4-2 Probability of arrest for a violent, property, or drug crime 36 months after release from prison. NOTE: Probabilities adjusted for time off the street. SOURCE: The analysis for this figure was conducted by Rosenfeld et al. (2005). both supportive and transitional reentry services in those first days, weeks, and months out of prison. Housing Needs and Barriers Securing housing is perhaps the most immediate challenge facing prisoners after release. Although most new releasees can count on family or friends to provide housing, those who cannot have very limited housing options. The situation is often complicated by a host of factors: the scarcity of affordable and available housing in many cities, legal barriers, preconceptions that restrict tenancy for this population, and local eligibility requirements for federally subsidized housing that may exclude many releasees, such as those who were convicted of drug offenses. Housing eligibility restrictions on ex-offenders are a critical public policy factor in planning reentry. Such restrictions and other barriers may lead to a concentration of releasees in low-rent, distressed neighborhoods, environs that are hardly conducive to successful reintegration. Released prisoners who do not have stable housing arrangements are more likely to return to prison (Metraux and Culhane, 2004). This finding suggests that the obstacles to securing both temporary and permanent hous-
OCR for page 55
Parole, Desistance from Crime, and Community Integration ing warrant the attention of policy makers, practitioners, and researchers. Supportive housing programs—such as the Oxford House model, originally conceived as a drug and alcohol addiction recovery program for substance users, and halfway houses that include on-site services (see Roman and Travis, 2004)—could be an option for former prisoners, but they have not been implemented on a wide scale (Jason et al., 2006).7 Mutual-help models like Oxford House are particularly effective at providing positive peer networks and may be an appropriate model for postrelease housing for former prisoners (Olson et al., 2005). Peer Support and Individualized Services More generally, greater peer support from other formerly incarcerated people is associated with less recidivism (Broome et al., 1996). In fact, the presence of other recovering peers has been shown to be more effective in achieving desistance than the involvement of clinical staff or correctional officers alone (Wexler, 1995). The most rigorous research study on this question was a randomized trial of an Oxford House program that tracked more than 90 percent of the participants, (substance abusers, often with criminal records) for 2 years: it found 50 percent less recidivism, among Oxford House residents than among a control group. Employment at the 2-year follow-up was also significantly higher among Oxford House participants (Jason et al., 2006). Another reentry programming approach is a package of individualized services for new releasees that has been referred to as “wraparound” service delivery. As the name suggests, it involves a comprehensive array of individualized services and support networks that are wrapped around clients, rather than presenting them only with set, inflexible treatment programs (Walker and Bruns, 2003). In the wraparound model, treatment services are usually provided by multiple agencies working together as part of a collaborative interagency agreement, and each new releasee’s service plan is developed and managed by an interdisciplinary team that includes a caseworker, family members and community residents, and several social services and mental health professionals. Wraparound interventions are different from traditional case management programs, which simply provide individuals with one caseworker 7 Researchers at DePaul University have been actively investigating this model for over 10 years and have produced dozens of manuscripts on its advantages; see, for example, Davis et al. (2006). One of the researchers, Dr. Brad Olson, is an adviser to the Safer Foundation’s programs. The mission of the foundation as posted on its website is to reduce recidivism by supporting, through a full spectrum of services: the efforts of offenders to become productive, law-abiding members of the community (see http://www.saferfoundation.org/viewpage.asp?id=4 [accessed August 2007].
OCR for page 56
Parole, Desistance from Crime, and Community Integration whose job is to guide them through the existing social services and try to ensure that they receive appropriate services. Wraparound programs feature several basic elements, including the collaborative team described above; interagency agreements; care-case coordinators to supervise diverse cases and treatment management; a family orientation; and a unified plan of service delivery. Wraparound approaches also emphasize the importance of recruiting committed staff and creating programs that are culturally competent and “strengths based” (see below) (Barton, 2006; Walker and Bruns, 2006). Wraparound programs with basic elements have become increasingly popular since the model was introduced in the 1980s. Strengths-based approaches attempt to rebalance the traditional focus on an individual’s problems, pathologies, and deficits, which can be demeaning and unproductive; instead, they focus on resilience, positive psychology, empowerment theories, and brief therapeutic approaches, such as solution-focused work and assessment and practices to capitalize on individual strengths (DeJong and Berg, 2002; Burnett and Maruna, 2004; Taxman, 2006). This approach has been found to work with families, substance abusers, and people with mental health problems, especially juvenile offenders (Early and GlenMaye, 2000; Rapp-Paglicchi and Roberts, 2004). The focus is on assessing and leveraging client capabilities, talents, and resources to support change and solve problems from a positive perspective. Belief in a client’s strength and focusing on his or her ability to change can foster motivation rather than resistance (Clark and Lee, 2005). AVAILABLE SERVICES AND THEIR EFFECTS Physical Health Services The incarcerated population in the United States is composed mostly of poor, urban, and undereducated people who have a high prevalence of health problems. They not only have higher rates of substance abuse and violence than the general population, but they also suffer high rates of physical health problems. Their generally riskier life-styles increase the prevalence of infectious diseases, such as HIV/AIDS, tuberculosis, sexually transmitted diseases, and hepatitis (see Brewer, 2001). In addition to infectious disease, their relatively higher rates of lack of access to health care, combined with poverty, substandard nutrition, and poor housing or homelessness contribute to increased risk for such chronic conditions as hypertension, cardiovascular disease, skin conditions, gastrointestinal disease, diabetes, and asthma. The effects of physical health problems on reentry for releasees has received some attention (Travis, 2005), but the effects have not been carefully or widely studied. One preliminary study suggests that releasees with
OCR for page 57
Parole, Desistance from Crime, and Community Integration medical problems are more likely to experience difficulties in reentry than those without such problems (Visher and Mallik-Kane, 2007). In recent years, legal mandates have required correctional systems to provide physical and mental health care to prisoners. And with an increase in the accreditation of correctional health care provision (by the American Correctional Association and the National Commission on Correctional Health Care), the quality of such care has come under greater scrutiny. The most innovative approach to the provision of reentry-focused health care is a program developed at the Hampden County Correctional Center in Massachusetts. Neighborhood health centers and the institution developed a collaboration that fostered both greater in-prison provision of quality health care services and more consistent linkages to needed services after a prisoner’s release. Health care providers and social service professionals work together as part of a team from each health center: a case manager with expertise in the social service needs of inmates and releasees strongly complements the health care delivered by physicians and nurses. The comprehensive approach ensures that inmates receive high-quality care and reduces the risks to good health after release (i.e., homelessness, substance abuse relapse, and lack of health insurance). An evaluation of the Hampden County program funded by the National Institute of Justice (Hammett et al., 2004) found that participants reported significantly better overall health, more interaction with community health care providers, and less frequent use of alcohol and hard drugs in the 6 months after release in comparison with the period before incarceration. However, an analysis of criminal history records revealed no relationship between recidivism and participation in the program. Moreover, there was no comparison group in this study. Replications of this program are being funded by the Robert Wood Johnson Foundation in nine sites, and an evaluation is planned. Mental Health Services The deinstitutionalization of mentally ill people that occurred during the 1960s and early 1970s rested on several assumptions. A key one was that people with mental illness could find and easily access community mental health services. Unfortunately, this assumption has not proved true. The lack of community facilities for mentally ill people has had the unintended consequence of making the criminal justice system the primary public response to problem behaviors associated with severe mental illness. Among prisoners, the rates of mental illness are two to four times higher than among the general population (Lurigio, 2001). In a 1998 survey, approximately 16 percent of those in state prisons, local jails, or on probation said they either had a mental condition or had stayed overnight in a mental hospital, unit, or treatment program (Ditton, 1999), about twice as
OCR for page 58
Parole, Desistance from Crime, and Community Integration high as estimates of mental illness from the National Survey on Drug Use and Health among the general population (see http://www.oas.samhsa.gov/nhsda.htm [accessed June 2007]). A more recent study reports that about one-half of state and federal prison inmates have a mental health problem, based on self-reports of treatment or diagnosis during the previous 12 months or on symptoms classified by interviewers using standard criteria (James and Glaze, 2006). Only about one-third of state prison inmates and one-quarter of federal inmates with a mental health problem reported having received mental health treatment since admission to prison. Until the early 1990s, most mentally ill defendants could expect to be processed by the criminal justice system in the same manner as defendants without mental illness. In the past 10−15 years, however, more innovative approaches have been initiated. Two federal funding initiatives—the targeted capacity expansion diversion program of the Substance Abuse and Mental Health Services Administration and the mental health courts grant program of the Bureau of Justice Assistance—have provided resources for new approaches for mentally ill offenders. In 1992, a national survey of jail diversion programs estimated that only about 52 jails had some kind of diversion program for offenders with mental illness (Monahan and Steadman, 1994); by 2003, the number had increased to 294 (U.S. Department of Health and Human Services, 2004). The approaches are not uniform. They include a variety of interventions that facilitate the provision of services to offenders in order to speed their release from incarceration. The overriding purpose of diversion programs is to provide services and reduce the length of incarceration; some also attempt to reduce or dismiss the charges brought against offenders. A cross-site evaluation of nine jail diversion programs found that, in general, the programs decreased the number of days spent incarcerated, thereby reducing criminal justice costs, and also reduced rearrests among participants (U.S. Department of Health and Human Services, 2004). A different approach for mentally ill offenders is mental health courts, a judicially managed program that relies on therapeutic jurisprudence (modeled after drug courts), in which the primary focus is on reduction or dismissal of the charges when a specified treatment regimen is successfully completed. In the late 1990s, only a few such courts accepted cases; since then, some 70 others have been established or are in planning stages. In 2000, the Mental Health Courts Grant Program was created by the America’s Law Enforcement and Mental Health Project Act. The program is managed by the Bureau of Justice Assistance, which provided grants to 37 courts in 2002 and 2003. The agency also funded technical assistance for all existing courts through 2006. As of July 2006, there were 113 mental health courts that responded to a survey conducted by the National Alliance for the Mentally Ill, the National GAINS Center for People with Co-Occurring
OCR for page 59
Parole, Desistance from Crime, and Community Integration Disorders in the Justice System, the TAPA Center for Jail Diversion, and the Council of State Governments.8 Under a court’s authority, defendants undergo regular therapy sessions and their medication is often monitored, sometimes allowing them to avoid prison time. Because of the newness of these courts, little is known about their effectiveness or whether the effects for parolees may differ from those for other criminal justice populations. Current research examining mental health courts is under way. Mentoring Programs Mentoring programs have a long history in delinquency prevention, and evaluation studies show positive effects for at-risk and delinquent youth (see Branch and Tierney, 2000; Herrera et al., 2000). The goal of mentoring is to support the development of prosocial life-styles, thereby reducing an individual’s exposure to or the temptations of risky and problem behaviors. Certain elements are essential to effective mentoring programs, including a high level of contact between mentor and mentee and a relationship that defines the mentor as a trusted adviser and supporter rather than an authority figure. There are several factors that serve as prerequisites for successful mentoring programs, including: (1) volunteer screening to eliminate inappropriate mentors, (2) communication and limit-setting training for mentors, (3) procedures that take account of mentors’ and mentees’ preferences, and (4) intensive supervision and support of each match (Center for Substance Abuse Prevention, 2000). An analysis (Jolliffe and Farrington, 2007) of 18 studies in the United States and the United Kingdom on the impact of mentoring on ex-offenders and recidivism found the following: Of 18 studies, 7 showed that mentoring of ex-offenders had a statistically positive effect. Subsequent offending was reduced by 4 to 11 percent. The methodologically superior studies did not show a significant reduction in reoffending. Programs that reduced recidivism used more frequent and longer mentoring sessions: once or more per week and 5 hours or more per session. 8 For the purpose of the survey, mental health courts were defined as adult criminal courts that (1) had a separate docket dedicated to persons with mental illnesses; (2) diverted criminal defendants from jail into treatment programs; and (3) monitored the defendants during treatment and had the ability to impose criminal sanctions for failures to comply with program requirements.
OCR for page 60
Parole, Desistance from Crime, and Community Integration Mentoring was only successful in reducing recidivism when included in a menu of interventions. The benefits of mentoring did not appear to persist after mentoring ended. Maruna (2001) has argued that the benefits of the mentoring process are often greater for the mentor than for the mentee (see also Cressy’s (1955) notion of “reflexive reformation”). Adult mentoring programs for former prisoners are being studied in the context of the DOL’s program, Ready4Work (R4W) Program, which also has a strong faith-based component. The mentors and coaches are drawn from the community, especially faith-based organizations. Focus groups indicate that the clients (potential mentees) want mentoring, and they express a particular preference for mentors who have had a prisoner/reentry experience. Both individual and group mentoring models are being implemented. 9 Once they are involved in the program, mentees continue to want mentoring, and they also want to become mentors themselves, participating as both a mentor and a mentee. Several Ready4Work participants have “graduated” and become mentors. A preliminary report provides examples of mentoring in the R4W sites that appear promising (Jucovy, 2006). An evaluation of the program now under way will examine the frequency and duration of participation in the mentoring aspects of R4W as it relates to a releasee’s successful reentry. The DOL has included mentoring as a key component in its new Prisoner Reentry Initiative, again with an emphasis on faith-based organizations as the source of the mentors. Despite the variety and promising prospects of the approaches just described, community services currently available for new releasees may not be meeting their needs. In a survey of men released from state prison to Chicago, 48 percent said that they had used some services in the 2 months since release (MacKenzie, 2006; La Vigne and Cowan, 2005; Giordano et al., 2002), but when asked what services were most useful, 17 percent said that none of them was useful. When asked “what would be most helpful right now” (6 months postrelease), 64 percent mentioned a job or job training, 53 percent mentioned financial support, 24 percent mentioned education, 38 percent mentioned housing, and 41 percent mentioned health insurance (La Vigne et al., 2006). Nonetheless, in community corrections, there is a research base suggesting what works at the individual offender level. The effects of a variety of programs for those returning to communities have been examined in com- 9 Preliminary evidence suggests that group mentoring may not be as powerful as traditional mentoring (Herrera et al., 2002), although it may be a less costly alternative.
OCR for page 61
Parole, Desistance from Crime, and Community Integration prehensive reviews of evaluations. One showed that parolees who stayed in halfway houses after release committed less severe and less frequent crimes (Seiter and Kadela, 2003). A second review showed that those intensive supervision programs with a strong treatment component had a sizable effect on recidivism (Aos et al., 2006). A third review concluded that human service-oriented programs are much more effective than those based on a control or deterrent philosophy. All of the strategies identified as effective by MacKenzie (2006) target dynamic criminogenic factors, are skill oriented, are based on cognitive-behavioral models, and treat multiple offender deficits simultaneously in particular, there is growing consensus that practices focusing on individual-level change, including cognitive change, education, and drug treatment, are likely to be more effective than other strategies, such as programs that only increase opportunities for work, reunite families, or provide housing (MacKenzie, 2006; Andrews and Bonta, 2003). These conclusions are consistent with several large meta-analyses of the evaluation literature (Aos et al., 2006; Andrews et al., 1990; Lipsey, 1995; Lipsey and Cullen, 2007). These findings seem to be somewhat at odds with the longitudinal research on desistance, which highlights the conditions that lead to law-abiding behavior, such as having a stable marriage and having strong ties to work (see Sampson and Laub, 1993; Laub and Sampson, 2003). However, it may be that programs associated with the desistance findings have had weak designs or implementation problems or that the evaluations have been flawed. Or it may be that individual-level change is a prerequisite for the conditions under which desistance takes hold. Or it may be that programs that target these conditions would be more successful if operated in close connection to individual change modalities. Best Practices Design and Implementation Problems The only evaluation of a contemporary prisoner reentry program to use a random assignment design, Project Greenlight in New York, demonstrates the implementation difficulties facing these programs (Wilson and Davis, 2006). Project Greenlight was developed by the Vera Institute of Justice on the basis of research and best practice models, and the institute believed it was creating an evidence-based reentry program. However, a thorough examination of the proposed model and its implementation revealed that the program modified best practices to fit institutional requirements, was delivered ineffectively, did not match individual needs to services, and failed to implement any postrelease continuation of services and support (Wilson and Davis, 2006; see also Visher, 2006; Marlowe, 2006). The evaluation found that the program participants performed significantly worse than a control group on multiple measures of recidivism after
OCR for page 62
Parole, Desistance from Crime, and Community Integration 1 year, a finding that the evaluators attribute to a combination of implementation difficulties, program design, and a mismatch between participant needs and program content. In response to the evaluation report, Marlowe (2006) argues that the evidence base for the program was flawed from the beginning, with weak designs and unproven, unstandardized interventions. Edward Rhine of the Ohio Department of Rehabilitation and Correction-and colleagues (2006) are more optimistic about reentry programming in general, and point out that Project Greenlight was not notably different from other failed reentry programs and that the treatment was not delivered appropriately (see also Wilson and Davis, 2006; Visher, 2006; Marlowe, 2006). CONCLUSIONS This review of programs and services for former prisoners suggests three main conclusions regarding their effectiveness in reducing recidivism and problem behaviors. First, there is scientific evidence that several programs and approaches reduce violations of community supervision requirements, arrests for new crimes, and drug use. They include cognitive-behavioral therapeutic approaches and frequent testing for drug use, coupled with treatment. Mentoring programs and comprehensive multiservice employment initiatives show promise but require further, more rigorous research. Second, inadequate implementation of program principles and procedures appears to be a significant obstacle in the way of being able to determine program effectiveness or finding out whether a program might have benefits for participants. Third, a major limitation of current program evaluation results is the failure to account fully for self-selection bias. Random assignment of persons to treatment and control conditions remains rare in research on the reentry process. Greater use of experimental designs, when such designs are feasible, is essential for drawing valid conclusions about reentry program effectiveness. When such designs are not feasible, greater attention should be paid to the selection of comparison groups and statistical adjustments for existing differences between program participants and nonparticipants. Although the field has moved beyond “nothing works” in assessments of program effects on reentry outcomes, it can identify with high confidence only a very few best practices for reducing recidivism and enhancing desistance among people leaving prison to return to local communities. More research, especially more experimental research, is needed to identify interventions that could significantly improve outcomes of community supervision for parolees (see Chapter 6).