Table of Contents
Family-Focused Juvenile Reentry Services: A Quasi-Experimental Design Evaluation of Recidivism Outcomes
Kristin Winokur Early, Steven F. Chapman, and Gregory A. Hand
Justice Research Center, Tallahassee, Florida
Kristin Winokur Early, Department of Research, Justice Research Center; Steven F. Chapman, Department of Research, Justice Research Center; Gregory A. Hand, Data Management Department, Justice Research Center.
Correspondence concerning this article should be addressed to: Kristin Winokur Early, Justice Research Center, 2898 Mahan Drive, Suite 4, Tallahassee, FL 32308; E-mail: email@example.com
KEYWORDS: reentry interventions, juvenile offenders, family therapy, recidivism, Parenting with Love and LimitsTM
Previous studies have evaluated the effects of community-based, family-focused juvenile probation services on recidivism. Many states are beginning to use such services as part of reentry programming for youth released from residential custody. Little is known, however, about whether these models effectively reduce rates of reoffending among youth transitioning from confinement. The current study used a quasi-experimental design to compare the family-focused Parenting with Love and LimitsTM (PLL) reentry services with standard aftercare offered through the St. Joseph County Probate Court in Indiana. We used intent-to-treat and protocol adherence analyses to evaluate recidivism outcomes. Youth released from PLL had lower rates of reoffending than those receiving standard aftercare, with statistically significant differences found for subsequent rates of juvenile readjudication. Effect sizes for the intervention ranged from -0.112 for rearrest to -0.221 for readjudication. Lengths of service were significantly shorter for the treatment sample than for the matched comparison group by an average of 2 months, suggesting that the intervention can serve more clients per year than standard aftercare while reducing costs associated with residential commitment. Findings have important implications for research and the implementation of juvenile reentry programs and strategies.
Nationally, juvenile arrest rates have declined to their lowest levels since 1980 (Puzzanchera & Adams, 2011). However, recidivism rates for youth released from juvenile correctional facilities have failed to keep pace. The number of arrests involving juveniles in the United States declined by 17% between 2000 and 2009 (Puzzanchera & Adams, 2011), yet recidivism trends reported by various states either have remained relatively stable or revealed only incremental decreases over time (Feyerherm, 2011; Florida Department of Juvenile Justice, 2011; Noreus & Foley, 2012; Pate, 2008; Rogan, 2008; Virginia Department of Corrections, 2011). Overall rates of recidivism for juveniles released from residential commitment are high. The Casey Foundation reports that 68% to 82% of these youth are rearrested within two years of release, and 38% to 58% are subsequently adjudicated or convicted for a new offense (Mendel, 2011).
Aftercare services in the juvenile justice system have historically been underfunded and have emphasized surveillance and community restraint, with little in the way of treatment interventions designed to address offender risks and needs, or the family and community dynamics to which youth return following residential commitment (Bouffard & Bergseth, 2008). Many states are looking to community-based juvenile reentry services that engage parents and caregivers in the treatment process as a way to reduce high rates of recidivism among youth released from correctional custody. These family-focused interventions are based on the theory that the family plays a pivotal role in reducing risk—directly, through social support and the exercise of supervision and guidance, and indirectly, by mitigating the influence of antisocial peers, antisocial thought patterns, and other potential risk factors.
Community-based programming has been found in several systematic reviews to provide larger effect sizes in reducing recidivism than traditional institutional interventions. For example, Andrews and colleagues (1990) found that the positive effects of appropriate correctional treatments in residential facilities were smaller than those in community-based facilities (0.20 for residential versus 0.35 in the community). In addition, the negative effects of inappropriate programming were more pronounced in residential settings (-0.15) than in the community (-0.06). Lipsey (2009) reported that recidivism effect sizes were largely similar whether juveniles at a given risk level received treatment services within the community or in a residential setting. Expanding upon their earlier research, Andrews and Bonta (2006) reached similar conclusions about the effectiveness of community-based treatment, finding the mean effect size of appropriate institutional programming was less than that of appropriate community-based programming (0.17 versus 0.35, respectively).
One of the advantages of community-based treatment for delinquent youth is that it offers the opportunity to intervene not only with the youth, but also to target risk factors associated with parents and the family. Juvenile offenders released from confinement often return to disorganized, chaotic family environments. The youth may have attained skills while in residential commitment, but the family may have remained largely unchanged in the interim. Addressing this issue becomes critical to reducing juvenile recidivism. Greenwood (2008) notes that "the most successful community-based programs are those that emphasize family interactions, probably because they focus on providing skills to the adults who are in the best position to supervise and train the child" (p. 198).
Family factors have a well-established link to antisocial behavior among youth, from classic research conducted by the Gluecks during the 1950s to today (Henggeler & Borduin, 1990; Quinn, 2004). Utilizing an ecological systems framework, Patterson and colleagues (1992) developed a social interactional, coercive family process model that mapped the developmental progression of antisocial boys to future delinquency and crime, with a focus on the influence of poor parental family management skills (Patterson, Reid, & Dishion, 1992). Longitudinal research from Patterson, Forgatch, Yoerger, & Stoolmiller (1998) found that family relationships and functions were related to the development of antisocial behavior in boys. They found that poor parental family management skills—such as disrupted parental discipline and inadequate monitoring—were strongly related to early antisocial behavior, arrest before age 14, and chronic offending by age 18. While intrafamilial dynamics have a strong influence, other researchers have pointed out that the involvement of family members themselves in anti-social behavior and crime is a risk factor for anti-social behavior among youth (Eddy & Reid, 2002). Further, meta-analytic reviews of the literature list negative parent-child relationships and poor parenting practices as being among the stronger predictors of delinquency (Andrews & Bonta, 2006; Lipsey & Derzon, 1998).
A number of community-based, family-centered treatment models have been used as frontend, diversionary or probation interventions with positive results (Butler, Baruch, Hickey, & Fonagy, 2011; Gordon, Graves, & Arbuthnot, 1995; Henggeler, Melton, & Smith, 1992; Sexton, & Turner, 2010; Winokur Early, Hand, Blankenship, & Chapman 2012). Among these are Functional Family Therapy (FFT), Multi-Systemic Therapy® (MST), Multidimensional Treatment Foster Care® (MTFC), Parenting with Love and LimitsTM (PLL), and other programs aimed not only at treating the offender, but at strengthening the family as the enduring social environment and source of social control for youth. Strong empirical evidence of the effectiveness of many of these programs has resulted in their being designated as evidence-based or model programs by groups such as the University of Colorado Blueprints for Violence Prevention project (Mihalic, Irwin, Elliott, Fagan, & Hansen, 2001), the Substance Abuse and Mental Health Administration’s National Registry of Evidence-Based Programs and Practices (http://www.nrepp.samhsa.gov), and the Model Programs Guide from the Office of Juvenile Justice and Delinquency Prevention (OJJDP) (http://www.ojjdp.gov31/mpg/).
There is additional evidence for the effectiveness of generic family counseling services in reducing juvenile recidivism (Lipsey, Howell, Kelly, Chapman, & Carver, 2010). In a recent metaanalysis, Lipsey and colleagues (2010) found that family counseling programs showed positive effects on recidivism in general—and although model programs produced varying degrees of positive results, "some no-name programs produced effects even larger than those found for the model programs" (p. 26). Yet, not all family counseling programs have achieved positive results. Further research is needed to identify specific characteristics that distinguish those that work from those that do not, as well as those that are effective as juvenile reentry interventions.
Evidence on the effectiveness of juvenile reentry services in general is relatively scant. Research has tended to support the finding that intensive aftercare supervision alone is ineffective in reducing juvenile recidivism (Bouffard & Bergseth, 2008; Petersilia & Turner, 1993). Largescale, federally funded initiatives have been undertaken over the last two decades to reform juvenile aftercare through the implementation of intensive supervision models that incorporate case management and treatment services focused on offender risks, needs, and strengths, including the Intensive Aftercare Program (IAP) (Altschuler & Armstrong, 1994) and the Serious and Violent Offender Reentry Initiative (SVORI) (Winterfield & Brumbaugh, 2005). Demonstration IAP programs have been introduced in Colorado, Nevada, and Virginia and included a randomized clinical trial. While some intermediate outcomes such as shorter lengths of commitment were reported for treatment participants, researchers found little difference in the prevalence of reoffending between the treatment and comparison groups over a 12-month follow-up period (Wiebush, Wagner, McNulty, Wang, & Le, 2005). Similarly, results of the impact of SVORI programming showed few differences in official post-release arrest or incarceration between those receiving SVORI and those not in the treatment group (Lattimore & Steffey, 2009). While SVORI could be characterized as an "overlay" intervention used to enhance or expand existing programs (Lattimore & Visher, 2009), the null findings generally reported with the official measures of recidivism underscore the need for additional research to identify effective aftercare models and strategies for delinquent youth.
The National Reentry Resource Center Advisory Committee on Juvenile Justice recently highlighted promising or emerging practices for youth reentry, including: 1) cognitive-behavioral approaches reflective of adolescent brain development, 2) strengths-based strategies emphasizing positive youth development, 3) meaningful family and community engagement in the process, 4) emphasis on education and employment, and 5) development of lifelong connections to facilitate successful transition to adulthood (Bilchik, 2011). What emerges from integrating these practices is a comprehensive system of care. In a growing number of communities, agencies are formally collaborating to provide a wide array of individualized services and support networks for youth reentry using a "wraparound" case management strategy (Burns & Goldman, 1999; VanDenBerg, Bruns, & Burchard, 2008). Unifying efforts among community social service agencies, initiatives such as Wraparound Milwaukee (Kamradt, 2000), Connections in Clark County, Washington (Koroloff, Pullman, Savage, Kerbs, & Mazzone, 2004), and the Repeat Offender Prevention Program (ROPP) in California (State of California Board of Corrections, 2002) have developed wraparound models that have been evaluated and found to reduce juvenile recidivism and improve youth behaviors, socialization, and academic performance. These findings support the need for reentry models that effectively mobilize community resources in the delivery of comprehensive, individualized systems of care for youth.
Family-Focused Reentry Services
A number of jurisdictions have experimented with the use of family-focused treatments as part of reentry programming for youth released from residential commitment. Examples include:
- New York State Office of Children and Family Services (OCFS) initiated MST interventions in March 2000 for youth released from OCFS residential facilities (Mitchell-Herzfeld et al., 2008).
- Maryland Department of Juvenile Services (DJS) recently expanded evidence-based programming through the implementation of FFT services with youth released from residential commitment in Baltimore (Rogan, 2008; VisionQuest, 2012).
- Washington State Department of Social and Health Services piloted the Family Integrated TransitionsTM (FITTM) Program in 2001. The program provides a combination of evidence-based approaches including MST, Dialectical Behavior Therapy (DBT), Motivational Enhancement Therapy, and Relapse Prevention Services (Aos, 2004).
Similar programs have been established on a smaller scale in communities around the country. Most, however, have not been fully evaluated. In their meta-analysis of systematic reviews on correctional rehabilitation, Lipsey and Cullen (2007) found that although there have been studies of such evidence-based programs, few separated out effects of community-based versus residential treatment. In addition, few outcome evaluations specifically examined the impact of family-focused juvenile reentry interventions (Lipsey & Cullen, 2007). Two exceptions are the interventions implemented through the New York Office of Children and Family Services and the Washington State Department of Social and Health Services.
Mitchell-Herzfeld and colleagues (2008) evaluated the impact of the pilot implementation of MST services with youth released from New York’s OCFS facilities. Contrary to positive outcomes achieved when MST has been used as a diversionary intervention, MST was not effective in decreasing recidivism among the aftercare population in New York. Rearrest rates were generally high for youth released from both MST and the control groups, ranging from 85% to 90%. Furthermore, boys in the MST group were significantly more likely than those in the control group to be rearrested for a violent felony offense following program completion. The receipt of MST services increased the odds of reconviction among girls in the pilot study and increased the likelihood of boys being reincarcerated. The authors concluded that two primary factors contributed to the results in New York: 1) the severity of problems facing OCFS youth and their families, which included high rates of mental health and substance abuse problems; and 2) the decision to use MST as a reentry intervention rather than as a front-end diversionary treatment. The latter was deemed problematic because MST therapists had greater difficulty engaging youth and their families and in reducing negative peer associations among youth, both inhibited by their residential incarceration preceding MST services (Mitchell- Herzfeld et al., 2008).
In contrast to the New York experience with MST, the evaluation of the Washington State FIT aftercare program yielded positive outcomes for youth who received FIT treatment (Aos, 2004; Trupin, Kerns, Walker, DeRobertis, & Stewart, 2011). The FIT model was pilot tested with youth identified in Washington to be at high risk for reoffending following release from a residential facility—specifically, those with co-occurring substance abuse and mental disorders (Trupin, Turner, Stewart, & Wood, 2004). Researchers found the pilot implementation of the model, which combined multiple evidence-based approaches, significantly lowered rates of felony recidivism for youth who received the treatment compared with offenders who were eligible for FIT but did not reside in one of the four counties in Washington in which FIT was being tested. Mean adjusted reconviction, felony reconviction, and violent felony reconviction outcomes had effect sizes of -0.126, -0.289, and -0.093, respectively. In a subsequent evaluation of FIT, Trupin and colleagues (2011) found that although participation in the program was associated with a 30% reduction in felony recidivism, this reduction did not appear related to overall, violent felony, or misdemeanor recidivism. Although some results have been promising, more research on the application of familyfocused models with juvenile reentry populations is needed.
In summary, although evidence has accumulated on the effectiveness of front-end, communitybased services in reducing juvenile recidivism (Aos, Barnoski, & Lieb, 1998; Barton, Alexander, Waldron, Turner, & Warburton, 1985; Henggeler, Cunningham, Pickrel, Schoenwald, & Brondino, 1996; Szapocznik & Williams, 2000), less is known about their impact on and application with juvenile offenders transitioning from residential confinement back to the community. Historically overlooked in the juvenile justice system, aftercare programming has had scant success in reducing the prevalence, frequency, or seriousness of reoffending (Lattimore & Visher, 2009); MacKenzie, 1999; Wiebush et al., 2005).
Studies have documented the effectiveness of family therapy interventions in preventing delinquency (Lipsey et al., 2010). Yet, when applied with juvenile reentry populations, the results have not always been positive (Mitchell-Herzfeld et al., 2008). Further research is needed to determine whether family-focused juvenile reentry programs can effectively reduce recidivism, and whether specific treatment and implementation strategies are more effective than others with youth transitioning from residential confinement back to their families and communities. The current research seeks to address these issues and fill this gap in the empirical evidence on juvenile aftercare.
This study sought to broaden the evidence base by evaluating the impact of a manualized reentry intervention with youth and their families using group and family therapy, begun while the youth were incarcerated and continuing through transition and into aftercare. Specifically, we evaluated the effectiveness of a new program, PLL Reentry, by comparing process (program completion, length of service) and recidivism (rearrest, readjudication, and recommitment) outcomes for youth receiving services compared with a matched sample of youth receiving standard aftercare services in the study site. The program engaged families early on in the service delivery process using motivational interviewing techniques; individual, group, and family therapy; trauma and wound work; and a wraparound case management approach to transitioning youth back to the community. The study is therefore a critical step in expanding the knowledge base about the overall effectiveness of family-focused juvenile reentry, as well as the impact of specific programming and implementation strategies.
PLL began in 2000 as a family-focused system of care for at-risk and delinquent youth and their families (Sells, 1998). The intervention was initially implemented as a diversion and probation overlay service for court-involved youth, with demonstrated success in reducing behavioral problems and recurring substance abuse (Smith, Sells, Rodman, & Reynolds, 2006), as well as subsequent juvenile justice system involvement among youth served (Hand, Winokur Early, & Blankenship, 2011). PLL has been replicated in 13 states and in Holland, and has received an Exemplary rating in the OJJDP Model Programs Guide (http://www.ojjdp.gov/mpg/). In 2007, PLL introduced its reentry model as part of a pilot demonstration project implemented through the St. Joseph County Probate Court in Indiana.
Based on a family systems framework, PLL Reentry targets juvenile offenders ages 14 to 17 who have serious emotional and behavioral problems, including aggression, criminality, drug or alcohol abuse, sexual offending, conduct disorder, running away, and/or chronic truancy. PLL Reentry is a program (Sells, 1998; Smith et al., 2006; Sells, Smith & Sprenkle, 1995) that integrates principles of structural family therapy with comprehensive fidelity protocols (Sells, 2002). The approach is grounded in Family Systems Theory, which has support in the literature to be an effective method for reducing adolescent conduct disorders (Lambie & Rokutani, 2002; Rowe, Parker-Sloat, Schwartz, & Liddle, 2003; Springer & Orsbon, 2002).
PLL Reentry is divided into three implementation stages: Stage I—Intensive, Stage II—Transition, and Stage III—Aftercare (Table 1). Juvenile aftercare services have historically begun after a youth’s release from residential commitment. PLL Reentry begins with the youth and family during the period in which the youth is confined. Continuity in services is established by having the same PLL therapist work with the youth and family from the initial commitment stage through aftercare and post-release treatment. The model is designed to facilitate a youth’s readiness for change and increasing commitment to the program, thereby reducing overall lengths of confinement and earlier release to the community. On average, Stage I lasts approximately 3 months, Stage II lasts approximately 1 to 2 months, and Stage III involves receiving 3 months of services in the community, with additional interventions and relapse prevention sessions as needed (Sells & Souder, 2007).
Using a quasi-experimental design, the current study examined the effectiveness of PLL Reentry in reducing lengths of residential confinement and reducing juvenile recidivism. We compared the outcomes of youth receiving PLL (treatment group) with those of a matched sample of youth who received standard aftercare programming (comparison group) in the study site.
The questions that guided the research were, first, does the family-focused treatment intervention reduce residential lengths of service compared with standard programming? This is of particular importance given recent research (Bouffard & Bergseth, 2008) suggesting the failure of MST to reduce recidivism among reentry youth, which may have been due in part to services not formally beginning until a youth’s release from confinement. Greater involvement with negative peers exacerbated by longer lengths of confinement may also have contributed to these results (Bouffard & Bergseth, 2008).
Table 1. PLL Reentry Model
|Stage||Stage I Intensive (3 months)||Stage II Transition (1–2 months)||Stage III Aftercare (3 months)|
Youth Commitment Status
|Residential Commitment||Residential Commitment||Post-Commitment|
Motivational Interviewing: 1 to 2 sessions lasting on average 1 hour in duration
Parent-Only Group Modules: conducted in the community lasting 2 hours on average per group:
Family Therapy I—Setting the Terms for Aftercare: 3 to 4 family therapy sessions lasting 1½ to 2 hours in duration
Family Therapy II—Customizing the Aftercare Plan: 3 to 4 additional family therapy sessions lasting 1½ to 2 hours in duration
Benchmark Meeting: The family, PLL Reentry therapist, probation officer, and residential facility staff meet to review youth’s performance in residential program, change in assessed risks and needs, aftercare plan and family participation
No Parent-Only Group Modules
Family Therapy III—Role Playing & Troubleshooting: Four sessions lasting 2 hours in duration focused on family role play practice, implementation of the aftercare plan, and preplan troubleshooting to address transition obstacles and techniques for handling resistance
Transition Services: Wraparound services in the community are identified and arranged including job and/or vocational placement, school reintegration, medication management, and mentoring
Benchmark Meeting: The family, PLL Reentry therapist, probation officer, and residential facility staff meet to review youth’s performance in residential program, change in assessed risks and needs, aftercare plan and prepare for transition
Family Therapy IV—Aftercare Maintenance: Therapy sessions occurring 2 to 4 times per week over the course of a minimum of 3 months, depending on need, and including participation of community wraparound stakeholders as applicable
Relapse Prevention: Calls backs to family every 30 days for 3 months postgraduation from PLL Reentry to monitor aftercare plan progress and address any obstacles
Refresher Sessions: Additional family therapy sessions as needed if relapse is imminent or has occurred
Second, which group achieves lower rates of reoffending: those receiving family-focused reentry services that begin while the juvenile is incarcerated and employ specific treatment strategies (such as motivational interviewing; individual, group and family therapy; trauma and wound work; and wraparound case management) or those receiving standard aftercare services? We examined multiple measures of the prevalence and severity of reoffending, including rates of rearrest, felony arrest, readjudication, felony adjudication, and recommitment. We hypothesized that committed youth receiving the familyfocused reentry would achieve lower rates on each of these outcomes than youth who received standard aftercare services.
During the 18-month period from August 2007 to February 2009, all eligible juvenile offenders transitioning from residential commitment back to the community in St. Joseph County, Indiana, were assigned to receive PLL Reentry services. Prior to that time, the county had used traditional community restraint and supervision reentry services with youth. Seeking a more effective mechanism for curbing juvenile recidivism, the county pilot tested the PLL Reentry model with all youth released from residential commitment. A youth was deemed ineligible for program referral only if a parent or caregiver was unavailable, which did not occur in the study site. Because the intervention was pilot tested with all eligible youth, an experimental design was not possible. Instead, we used a quasi-experimental design featuring a comparison group identified through official records of committed youth released to standard probation services during the 18-month period preceding the implementation of PLL Reentry.1
The current analysis examined a total of 354 cases, which consisted of all 201 cases of youth released from standard reentry services in the study site during the 18 months preceding PLL implementation (February 2006 to August 2007) and all 153 PLL cases processed during the following 18-month period (August 2007 to February 2009). PLL cases were matched to standard reentry cases using propensity score matching (PSM), yielding 153 pairs of treatment and comparison reentry cases.
We examined a number of output and outcome measures in addressing the study research questions, including rates of program completion, lengths of service, and five measures of subsequent reoffending and placement. The first measure was reported only for youth served by PLL Reentry, as they were deemed to have received a treatment intervention or dosage; this was operationally in contrast to compliance with the conditions of standard aftercare supervision tracked for youth in the comparison group. The PLL Reentry completion rate provided an indicator of program retention and was defined as any case designated by the St. Joseph County Probate Court as having successfully completed the requirements of the intervention. Program completion requirements included the following criteria:
• Family completion of five or more parent-only group modules;
• Youth and family completion of eight or more family therapy sessions;
• No reports of curfew violations or running away over the course of service delivery;
• No reports of school truancy or failing grades;
• No reports of law violations or problems in the home over the course of service delivery; and
• Stabilization of any mental health issues, as applicable.
Length of service was measured as the number of days between a youth’s admission to residential commitment and release from the treatment or comparison group services. In addition, given that PLL Reentry providers maintained records of client referral and exit dates, it was also possible to measure the specific length of service delivery for PLL services. This measure, however, could not be calculated for comparison group youth, as there was no comparable reentry treatment dosage period.
We used five measures of recidivism: rearrest, felony arrest, readjudication, felony readjudication, and recommitment. Arrest rates provide an indicator of subsequent court involvement and system impacts, but do not necessarily indicate that a youth has been found to have committed a subsequent crime. As such, the study also examined whether youth were subsequently adjudicated for a juvenile offense. We tracked these measures uniformly for all study youth for 12 months following release from either treatment or comparison services. Additional outcomes included the nature of subsequent arrests and adjudications, with classifications including felony, misdemeanor, or status offense.
The primary independent variable of interest was participation in the family-focused juvenile reentry program versus standard aftercare probation services (yes/no). Demographic and offender characteristics of youth in the treatment and comparison groups were used in the matching process, as well as in subsequent analyses examining the study outcomes. Youth characteristics included gender (male/female), race (White/ non-White), ethnicity (Hispanic/non-Hispanic), average age at release from reentry services (in years), most serious current offense (felony, misdemeanor, or non-law violation2), number of prior juvenile adjudications, and most serious prior adjudicated offense (felony, misdemeanor, or non-law violation).
The St. Joseph County Probate Court maintains jurisdiction over all juvenile matters and is responsible for tracking official offender case information in the Quest Case Management System (Courts, 2009). The PLL Reentry program simultaneously maintained a reporting system and tracked all youth receiving services in the study site. We obtained socio-demographic, legal, and service delivery data from the PLL Reentry system. We cross-referenced information on offender histories, dates of service delivery, and sociodemographic data against official court data extracted from the Quest system by St. Joseph County staff. We used the Quest extracts as the official source for determination of youth served, outcome measures, and independent variables for the treatment and comparison reentry groups.
We used PSM to achieve an equivalent comparison group from the population of reentry youth who received services during the 18 months before PLL implementation. For the evaluation, we used the probabilities produced by a logistic regression model to calculate the propensity score as the probability of a youth being assigned to PLL Reentry versus standard reentry services. We adopted an intent-to-treat approach (all matches that were admitted to services with intent to treat, without regard to completion status) to help reduce the bias that occurs when youth with more difficult problems drop out or are rejected due to noncompliance. The intentto- treat approach aims to determine the outputs and outcomes of the treatment as implemented in the study site, which includes implications for placement and retention policies, as well as model fidelity and practitioner competence. We also used a second protocol adherence approach, selecting youth who complied with program requirements and completed services, to focus on the efficacy of the treatment.
The majority of the total study sample involved male (88%), non-White (54%) offenders ages 16 to 18 at the time of release from reentry services (Table 2). Slightly less than one-half the sample were White youth (47%), 44% were African American, slightly less than 9% were classified as multiracial, and 7% were Hispanic. Felonies were the most serious offenses for which youth in the sample were disposed to residential confinement, applying to slightly more than half (51%). Almost as many youth (45%) were disposed to residential confinement for misdemeanors, and less than 5% were confined for non-law infractions. The majority of the full sample had a prior adjudication, most involving a felony. Before their current offense, which resulted in residential placement and subsequent aftercare programming, youth in the sample had an average of one prior adjudicated offense. We tracked all 354 youth in the sample for recidivism through official records for a period of 12 months postrelease from reentry or standard probation aftercare services.
Table 2. Sample Characteristics
|Total Sample (N=354)||PLL Reentry (N=153)||Standard Reentry (N=201)|
|Percent African American||44.4||42.5||45.7|
|Percent Native American||0.3||0.7||0.0|
|Age at release (years)—Mean (SD)||17.2(.18)||17.4(1.13)||17.1(1.21)|
|Most serious current offense|
|Percent non-law violation||4.6||2.0||6.5|
|Total prior adjudications—Mean (SD)||1.07(1.24)||1.01(1.17)||1.10(1.29)|
|Most serious prior adjudication|
|Percent non-law violation||11.6||10.5||12.5|
|No prior adjudication||42.6||42.4||42.8|
|Percent followed 12 months||100.0||100.0||100.0|
We initially examined differences in the treatment and comparison samples using bivariate analyses (Table 3). PLL served a higher percentage of males, Whites, and Hispanics than were served through standard aftercare programming. On average, youth in the standard reentry sample had less serious offense histories than those in the treatment sample. Slightly less than 60% of the PLL Reentry sample had a felony as the most serious current offense. In contrast, youth receiving standard aftercare services were more often committed for a misdemeanor or non-law violation. The treatment sample was somewhat older than the comparison group at the time of release. To control for sampling bias associated with the propensity to have been served by standard reentry services as opposed to PLL Reentry, a logistic regression model, including the variables examined at the bivariate level, was calculated. It is important to note that these were measures unaffected by the treatment of interest and thus appropriate for use in estimating the propensity for placement to the intervention. Furthermore, although some of the variables may have been significantly related to one another and therefore collinear, propensity score estimation (PSE) is less focused on parameter estimation of the model and more focused on achieving balance in the covariates (Augurzky & Schmidt, 2001). In addition, Stuart (2010) notes that the inclusion of variables that are unrelated to treatment assignment are of little influence in the propensity score model. Rather, the potential for an increase in bias is more likely to result from the exclusion of important confounders (Stuart, 2010).
Table 3. Baseline Covariates Before and After Propensity Score Matching (PSM)
|Standard Reentry (SR)||PLL Reentry to Post-PSM SR|
|PLL Reentry (N=153)||Post-PSM (N=153)||Pre-PSM (N=201)||Test Statistic (p Value)|
|Age at release (years)—Mean (SD)||17.4
|Most serious current offense|
|Percent non-law violation||2.0||5.2||6.5*||X2=2.36
|Total prior adjudications—Mean (SD)||1.01
|Most serious prior offense|
|Percent non-law violation||10.5||11..1||12.5||X2=0.03
We used the probabilities generated from the model as the estimate of the propensity score (Table 4). Based on the multivariate analysis, the significant variables that distinguished juveniles in PLL versus standard reentry were age and the nature of the most serious current offense (p<.10). Using the propensity scores, we matched each PLL Reentry case to the nearest neighboring standard reentry case with the exact same or closest score. Nearest neighbors were selected at random to avoid bias, with unselected cases replaced in the pool for the next PLL match.
Table 4. Logistic Regression of Placement in Treatment Group Pre-PSM
|Age at release||.251||.097||6.632||.010||1.285|
|Worst current offense (felony)||.623||.224||7.756||.005||1.864|
|Total prior adjudications||-.131||.109||1.452||.228||.877|
|Worst prior (non-law violation)||-.067||.356||.035||.852||.936|
The matching procedure yielded 153 pairs of PLL and standard reentry cases. Following the PSM adjustments, the final sample compositions were more balanced than the non-matched groups. Examination of the covariates following the PSM protocol revealed no statistically significant differences between the treatment and matched comparison groups (refer back to Table 3).
One of the primary goals of family-focused reentry services is the effective engagement of parents and caregivers in juvenile rehabilitation. We hypothesized that this engagement would foster change not only within the youth, but also within the family to which the youth returns following release from residential confinement. Prior research cited difficulties engaging families after youth returned to the community as a decided obstacle to full model service delivery. Rather than beginning services after youth were released from residential confinement, PLL Reentry began group and family therapy while the youth was still committed. This model characteristic may have contributed to the program completion rates. Of the 153 cases admitted, 124 (81%) successfully completed the program. All of the female clients and their families completed services, while 79% of the males and their families completed services. A smaller proportion of African American youth (74%) completed services compared with White (87%) and Hispanic youth (77%). Younger youth exhibited higher rates of completion than youth who were older at the time of release, with 83% of youth under age 17 completing PLL Reentry services compared with 79% of youth who were age 19 or older.
Much has been written of the deleterious impact of justice system involvement on future outcomes for youth. The current study hypothesized that successful family engagement early in the rehabilitative process would decrease overall lengths of service.3 The findings support the hypothesis, as the average length of service for the non-matched standard reentry sample was just under 2 months longer than the PLL sample. When we examined length of service for the 153 matched pairs, the difference between the treatment and comparison group was significantly larger (t = 2.63, df = 219, p < .01, two-tailed), with matched reentry youth averaging 442 days of service and PLL Reentry youth averaging 371 days of service. Treatment youth who successfully completed the PLL model requirements achieved a slightly lower average of 363 service days between the start of residential placement and completion of aftercare programming.
We examined recidivism outcomes for the treatment and matched comparison reentry samples from both an intent-to-treat approach and from a protocol adherence approach. The former examined the effect of the intervention for all youth admitted to the program, regardless of whether services were successfully completed. The latter approach examined outcomes for participants who completed the full course of treatment. The results of the intent-to-treat protocol presented in Table 5 reveal that recidivism prevalence rates were lower for youth admitted to PLL Reentry services than recidivism rates for their counterparts admitted to standard aftercare programming. This held true for rates of rearrest (29.4% versus 34.6%), as well as rates of readjudication (17.7% versus 26.8%), a difference that was statistically significant at the 0.05 level. The rate of recommitment for the treatment group was 33% lower than that for the comparison group. When we considered the severity of reoffending as measured by subsequent felony arrests and adjudications, the rates were again lower for the treatment sample. Of particular note, the rate of felony adjudications for youth released from standard reentry services was nearly double that of the PLL Reentry group (13.7% and 7.2%, respectively), a difference that was statistically significant at the 0.10 level (t = 1.87, df = 282, p = .06, two-tailed). The magnitude of the treatment effect for readjudication was measured using Cohen’s d, which produced an effect size of -0.221. For rearrest, the effect size was -0.112, for felony arrest, -0.144, and for felony adjudication, -0.214. The outcome for recommitment 12-months post-release had an effect size of -0.133.
Table 5. Treatment and Comparison Group Outcomes, Intent-to-Treat Approach (n = 306)
|PLL Reentry||Matched Standard Reentry||t-Test Statistica||dfb||Significance (2-tailed)|
|Felony arrest rate||17.7%||23.5%||1.27||301||0.20|
|Felony adjudication rate||7.2%||13.7%||1.87||282||0.06|
|Average length of service (days)c||370.9||441.9||2.63||219||0.01|
We further evaluated the impact of PLL Reentry on subsequent juvenile justice system involvement among participants using a protocol adherence approach (Table 6). We examined outcomes for the cases in which the youth and family completed the full course of PLL Reentry services, and outcomes for their matched pairs within the comparison group (n=248). Rates of reoffending among youth completing PLL Reentry were lower than for those of the sample of all cases admitted to the program, suggesting an association between completion of the full course of treatment and improved outcomes. The treatment group outperformed the comparison sample on each measure. The results revealed more pronounced differences between the samples on rearrest, felony arrest, and recommitment than found with the intent-to-treat analysis. Slightly less than 17% of the PLL youth were readjudicated compared with 25.8% of the comparison group, and 13.7% of PLL youth were recommitted compared with 20.2% of the youth completing standard reentry. The severity of reoffending was greater among comparison cases, with a felony adjudication rate nearly double that of the treatment group. Effect sizes for these results ranged from -0.139 to -0.219. For rearrest, the effect size was -0.139; for felony arrest, -0.204; for readjudication, -0.217; for felony adjudication, -0.219; and an effect size of -0.172 corresponded to the mean differences between the groups on subsequent residential confinement.
Table 6. Treatment and Comparison Group Outcomes, Protocol Adherence Approach (n=248)
|PLL Reentry||Matched Standard Reentry||t-Test Statistica||dfb||Significance (2-tailed)|
|Felony arrest rate||15.3%||23.4%||1.61||240||0.11|
|Felony adjudication rate||6.5%||12.9%||1.72||226||0.09|
|Average length of service (days)c||363.7||434.9||2.47||173||0.02|
Although mean differences between the treatment and comparison samples were generally larger with the protocol adherence analysis than the intent-to-treat approach, they were not large enough to reach statistical significance at the 0.05 level. This was, in part, due to the smaller sample of completion cases. Those completing PLL Reentry did, however, exhibit lower rates of reoffending on all five outcome measures than youth who dropped out of treatment (Figure 1).
Figure 1. Outcomes for PLL Reentry Completers vs. Dropouts
Summary and Conclusions
Study results from multi-year, multi-site, federally funded aftercare initiatives have found scant evidence that intensive aftercare programming reduces the prevalence or seriousness of subsequent juvenile court involvement. The empirical literature is replete with evidence of the effectiveness of community-based and family- focused juvenile justice programs in reducing recidivism among youth disposed to diversion and probation. In contrast, little is known about the effectiveness of such programs used as reentry interventions with youth transitioning from residential confinement back to their communities and families. Even less is known about which reentry program components work and which do not. The results of this study suggest that PLL family-focused reentry services, when implemented with both youth and their families early in the juvenile’s residential confinement period, can reduce the prevalence and seriousness of subsequent offending among youth served.
The current study sought to expand the research by evaluating a widely used family-focused intervention, not as a front-end diversionary service, but rather as an aftercare program for youth transitioning from residential confinement. Using a quasi-experimental design, the study examined the impact of PLL Reentry on juvenile recidivism compared with a matched sample of youth who received standard aftercare programming through the St. Joseph County Probate Court in Indiana. The evaluation was situated in a "real world" setting, and although an experimental randomized trial was not possible, biases were minimized through the use of propensity score matching. Data were analyzed using both an intent-to-treat approach and a protocol adherence approach.
PLL Reentry was piloted for the first time in St. Joseph County, with a total of 153 cases admitted between August 2007 and February 2009. During that time, all youth transitioning from residential commitment were referred to the aftercare intervention. Eighty-one percent of the youth and families admitted to the program completed PLL Reentry services. This is in contrast to lower rates (73% overall, 70% for boys, and 82% for girls) reported in New York when MST was used as an aftercare intervention (Mitchell-Herzfeld et al., 2008). Family involvement in the PLL treatment process began early, while the youth was still in residential placement. Rather than waiting to engage the family until after the youth was released from confinement, starting earlier may have improved the likelihood that youth would complete the full dosage of services.
All of the female clients admitted to the treatment program completed services, while 79% of the males met the completion requirements. A greater proportion of White clients completed the program than did African American or Hispanic participants. Sample sizes were small however, particularly since only 9% of the treatment sample consisted of female juvenile offenders. Nonetheless, the findings warrant further evaluation of the PLL Reentry family engagement and implementation process to better understand for whom services appear to work, as well as mechanisms for addressing any service delivery issues with youth for whom the intervention is less effective.
Program treatment components were designed to foster reductions in lengths of confinement, thereby moderating the adverse effects associated with incarceration, including those resulting from commingling with negative peers. The combined length of services for residential commitment and aftercare required just over 14½ months for the matched standard reentry cases, compared with 12.2 months for those receiving PLL Reentry, a difference of 71 days. This suggests that PLL Reentry can serve more clients in a year than the prior reentry services model while reducing costs associated with residential commitment. This potential benefit ultimately depends on the extent to which the intervention significantly reduces recidivism among youth served.
Results from the current research indicate that in addition to potential cost savings, the familyfocused reentry program also reduced recidivism compared with standard aftercare programming in the study site. Across both the intent-to-treat and protocol adherence approaches, compared with the matched standard reentry sample, youth receiving PLL Reentry had lower rates of subsequent justice system involvement on all five indicators of recidivism prevalence and seriousness measured. We found significant treatment and comparison group differences in rates of readjudication in the intent-to-treat analyses, with the observed rate for the matched reentry sample more than 51% higher than that of the treatment sample. The prevalence of rearrest was lower for PLL cases than it was for the matched comparison group, as was the rate of felony arrests. The direction of these results are opposite those found in New York with MST reentry services (Mitchell-Herzfeld et al., 2008), and similar to those found with Washington’s use of the FIT program (Aos, 2004). The magnitude of the difference between the treatment and comparison groups is contrasted with Lipsey’s (2009) metaanalysis of 548 studies in which 1-year rearrest rates were approximately 6 percentage points lower for youth receiving the target intervention compared with those who did not receive the treatment.
Overall rates of readjudication were lower for PLL Reentry compared with treatment as usual in the study site. Eighteen percent of the youth admitted to the treatment program were subsequently adjudicated for a juvenile offense within 1 year of program completion, compared with 27% of those in the comparison group. Outcomes from the evaluation of the FIT Program revealed 18-month rates of conviction for juvenile or adult offenses of 42% for the treatment and 48% for the comparison groups (Aos, 2004). The magnitude of the mean treatment effects were measured using Cohen’s d, with intent-totreat effect sizes of -0.112 for rearrest, -0.133 for recommitment, -0.214 for felony adjudication, and -0.221 for readjudication. These findings are similar to those reported by Lipsey (2009) when examining the effects of family counseling programs on recidivism, and to those of Aos (2004) in reporting effect sizes of -0.126 and -0.289 for any adjudication/conviction and felony adjudication/ conviction.4 Finally, completion of the full treatment intervention, as opposed to a reduced dosage, appears related to improved outcomes. Completers had lower rates of rearrest, felony rearrest, adjudication, felony adjudication, and recommitment than youth who dropped out of the PLL Reentry program.
The study was not without its limitations. A relatively small sample size was used in examining the effectiveness of the reentry model with juvenile offenders released from residential confinement in a northern Indiana county with a population just under 267,000, of which 79% is White and 25% is under age 18. The analysis used data from the court database, which was limited in terms of information related to the details of treatment both for the PLL Reentry and the standard aftercare groups. Research should be expanded to increase sample sizes and should make an attempt to capture more data relative to risk factors associated with recidivism, especially in the areas targeted for treatment through the PLL model.
Data related to the involvement of family and the effects on family interaction and parenting functions would also increase our understanding of the use of family-based interventions as aftercare. In addition, data related to the implementation of the PLL model at the St. Joseph site would help to analyze factors related to engagement of the youth and his or her family with the therapeutic process. While the results appear to support the hypotheses that PLL Reentry would achieve shorter lengths of confinement and lower rates of reoffending compared with traditional aftercare, the data and analyses offer no insight as to the specific strategies, services, or outside factors accounting for these findings. Possible reasons for the positive outcomes include cognitive-behavioral change in youth and/or parents, improved family functioning or communication, competent and committed staff, or perhaps increased attention stemming from program participation. The nature of the initial pilot implementation of PLL Reentry and the use of administrative data did not permit the collection or analysis of pre/post measures that might narrow the reasons for the positive results found. Additional research that incorporates assessment of youth and family change metrics, as well as indicators of staff and program characteristics for both the treatment and comparison groups, is necessary to more fully evaluate the effectiveness of this reentry model.
These limitations notwithstanding, the preliminary results from the initial implementation and evaluation of PLL Reentry offer support for the use of family-focused aftercare with youth transitioning back to their homes and communities following residential commitment. Key programmatic questions remain. What service components contributed to the reduced rates of reoffending? What implementation strategies were critical in the service delivery process? While a comprehensive process evaluation is needed to adequately address these questions, a few explanations are considered.
The current results suggest that the timing of service delivery may be a critical factor. Mitchell- Herzfeld and colleagues (2008) concluded that MST failed to reduce recidivism among New York youth committed to residential facilities because the constellation of problems facing the family were severe, and because MST was used as a post-release service. While PLL Reentry was used post-release, services began with not only the youth, but also the family, approximately 4 months before residential discharge. As such, total treatment duration ranged on average between 7 and 9 months compared with the 3 to 5 months reported by Mitchell-Herzfeld and colleagues. The earlier start of services and the effective engagement of the family before their child’s release appear to have contributed to reduced lengths of residential confinement and lower rates of recidivism. The use of individual, group, and family therapy services implemented before discharge and designed to address the severity of problems faced by families, including problems regarding communication, trauma, and supervision, for example, may have equipped parents and youth with the tools necessary to more effectively handle the youth’s transition back to the family and community. The strengthened bonds may, in turn, have aided youth in the process of abstaining from additional criminal activity.
The PLL Reentry model also used a separate manualized curriculum tailored to the aftercare population, as opposed to the front-end diversion population. Consistency in service delivery was facilitated by use of the same PLL therapist working with the youth and family, from residential commitment to release from reentry services. The model likewise incorporated a wraparound case management approach to aftercare in which teams consisting of the PLL Reentry therapist, school personnel, job placement counselors, psychiatrists and psychologists, and mentors worked collectively to address the unique needs of the youth and family. While the theoretical implications of this strengths-based approach are well established (Hawkins, Catalano, & Miller, 1992), further research is needed to explore the effects of these specific service components with reentry youth and to examine the impact of the timing of service delivery to identify strategies for achieving better outcomes. It is also important to understand what types of youth best respond to family-focused reentry services, as well as those for whom such interventions are less effective. This initial evaluation of the PLL Reentry model produced promising results, but further replication of the intervention in urban and rural areas, as well as with varying types of offenders (e.g., sex offenders, violent offenders, youth with a history of severe substance use, etc.) is needed. The research on juvenile aftercare services is still in its relative infancy. We hope that the results of the current evaluation help to expand the empirical evidence on the effectiveness of family-focused reentry services in reducing the prevalence and severity of juvenile recidivism.
About the Authors
Kristin Winokur Early, PhD, is vice-president and director of Research, Justice Research Center, Tallahassee, Florida.
Steven F. Chapman, PhD, is senior research scientist, Justice Research Center, Tallahassee, Florida.
Gregory A. Hand, BS, is director of Data Management, Justice Research Center, Tallahassee, Florida.
Altschuler, D. M., & Armstrong, T. L. (1994). Intensive aftercare for high-risk juveniles: A Community Care model. Summary. Washington, DC: U.S. Department of Justice, Office of Justice Programs, Office of Juvenile Justice and Delinquency Prevention.
Andrews, D. A., & Bonta, J. (2006). The psychology of criminal conduct (4th ed.). Newark, NJ: LexisNexis.
Andrews, D. A., Zinger I., Hoge, R., Bonta, J., Gendreau, P., & Cullen, F. (1990). Does correctional treatment work? A clinically relevant and psychologically informed meta-analysis. Criminology, 28(3), 369–401.
Aos, S. (2004). Washington state’s family integrated transitions program for juvenile offenders: Outcome evaluation and benefit-cost analysis. Olympia, WA: Washington State Institute for Public Policy.
Aos, S., Barnoski, R., & Lieb, R. (1998). Preventive programs for young offenders: Effective and costeffective. Overcrowded Times, 9(2):1, 7–11.
Augurzky, B., & Schmidt, C. M. (2001). The propensity score: A means to an end. (IZA Discussion Paper No. 271.) Bonn, Germany: Institute for the Study of Labor (IZA). Retrieved from http://ssm.com.abstract=270919
Barton, C., Alexander, J. F., Waldron, H., Turner, C. W., & Warburton, J. (1985). Generalizing treatment effects of Functional Family Therapy: Three replications. American Journal of Family Therapy, 13(3), 16–26.
Bilchik, S. (2011). Five emerging practices in juvenile reentry. National Reentry Resource Center. New York: Council of State Governments Justice Center.
Bouffard, J., & Bergseth, K. (2008). The impact of reentry services on juvenile offenders’ recidivism. Youth Violence and Juvenile Justice, 6(3), 295–318.
Burns, B. J., & Goldman, S. K. (1999). Promising practices in wraparound for children with serious emotional disturbance and their families. In B. J. Burns & S. K. Goldman (Eds.), Systems of care: Promising practices in children’s mental health, 1998 Series, Volume IV. Washington, DC: Center for Effective Collaboration and Practice, American Institutes of Research.
Butler, S., Baruch, G., Hickey, N., & Fonagy, P. (2011). A randomized controlled trial of multisystemic therapy and a statutory therapeutic intervention for young offenders. Journal of the American Academy of Child & Adolescent Psychiatry, 20(12), 1220–1235.e2.
Courts of the 60th Judicial Circuit (2009). State of the judiciary for the courts of record of the 60th Judicial Circuit, St. Joseph County, Indiana. South Bend, IN: Courts of the 60th Judicial Circuit.
Eddy, J. M. & Reid, J. B. (2002). Interventions for antisocial behavior: Overview. In J. B. Reid, G. R. Patterson, & J. Snyder (Eds.), Antisocial behavior in children and adolescents: A developmental analysis and model for intervention (pp. 195–201). Washington, DC: American Psychological Association.
Feyerherm, W. (2011). Oregon juvenile justice system needs analysis: Juvenile crime trends and recidivism report. Salem, OR: Oregon Commission on Children and Families. Retrieved from http://www.oregon.gov/OCCF/Documents/JCP/Oregon_Juvenile_Crime_2002-2009_Trends_Analysis.pdf?ga=t
Florida Department of Juvenile Justice (2011). 2011 Comprehensive accountability report. Tallahassee, FL: Florida Department of Juvenile Justice. Retrieved from http://www.djj.state.fl.us/research/ reports/car
Glueck, S., & Glueck, E. (1950). Unraveling juvenile delinquency. New York: The Commonwealth Fund.
Gordon, D., Graves, K., & Arbuthnot, J. (1995). The effect of Functional Family Therapy for delinquents on adult criminal behavior. Criminal Justice and Behavior, 22(1), 60–73.
Greenwood, P. (2008). Prevention and intervention programs for juvenile offenders. The Future of Children, 18(2), 185–210.
Hand, G., Winokur Early, K., & Blankenship, J. (2011). Parenting with Love and Limits: Evaluation 2011. Tallahassee, FL: Justice Research Center.
Hawkins, J. D., Catalano, R. F., & Miller, J. Y. (1992). Risk and protective factors for alcohol and other drug problems in adolescence and early adulthood: Implications for substance abuse prevention. Psychological Bulletin, 112(1), 64–105.
Henggeler, S. W., & Borduin, C. M. (1990). Family therapy and beyond: A multisystemic approach to treating the behavior problems of children and adolescents. Pacific Grove, CA: Brooks/Cole.
Henggeler, S., Melton, G., & Smith, L. (1992). Family preservation using Multisystemic Therapy. Journal of Consulting and Clinical Psychology, 60, 953–961.
Henggeler, S. W., Cunningham, P. B., Pickrel, S. G., Schoenwald, S. K., & Brondino, M. J. (1996). Multisystemic Therapy: An effective violence prevention approach for serious juvenile offenders. Journal of Adolescence, 19, 47–61.
Kamradt, B. (2000). Wraparound Milwaukee: Aiding youth with mental health needs. Juvenile Justice Journal, 7(1), 14–23.
Koroloff, N., Pullman, M., Savage, P., Kerbs, J., & Mazzone, V. (2004). The Clark County, Washington Children’s System of Care evaluation summative report. Portland, OR: Portland State University, Graduate School of Social Work, Regional Research Institute for Human Services.
Lambie, G. W., & Rokutani, J. (2002). A systems approach to substance abuse identification and intervention for school counselors. Professional School Counseling, 5(5), 353–359.
Lattimore, P. K., & Steffey, D. M. (2009). The multi-site evaluation of SVORI: Methodology and analytic approach. Research Triangle Park, NC: RTI, International.
Lattimore, P. K., & Visher, C. A. (2009). The multisite evaluation of SVORI: Summary and synthesis. Research Triangle Park, NC: RTI International.
Lipsey, M. W. (2009). The primary factors that characterize effective interventions with juvenile offenders: A meta-analytic overview. Victims and Offenders, 4, 124–147.
Lipsey, M., & Cullen, F. (2007). The effectiveness of correctional rehabilitation: A review of systematic reviews. Annual Review of Law and Social Science, 3, 297–320.
Lipsey, M., & Derzon, J. (1998). Predictors of violent or serious delinquency in adolescence and early adulthood. In R. Loeber & D. P. Farrington (Eds.), Serious and violent juvenile offenders: Risk factors and successful interventions, Thousand Oaks, CA: Sage Publications.
Lipsey, M., Howell, J., Kelly, M., Chapman, G., & Carver, D. (2010). Improving the effectiveness of juvenile justice programs: A new perspective on evidence-based practice. Washington, DC: Georgetown University Public Policy Institute, Center for Juvenile Justice Reform.
MacKenzie, D. L. (1999). Commentary: The effectiveness of aftercare programs—examining the evidence. Washington, DC: U.S. Department of Justice, Office of Justice Programs, Office of Juvenile Justice and Delinquency Prevention.
Mihalic, S., Irwin, K., Elliott, D., Fagan, A., & Hansen, D. (2001). Blueprints for violence prevention. Washington, DC: U.S. Department of Justice, Office of Justice Programs, Office of Juvenile Justice and Delinquency Prevention. Retrieved from www.colorado.edu/cspv/blueprints
Mendel, R. (2011). No place for kids: The case for reducing juvenile incarceration. Baltimore, MD: The Annie E. Casey Foundation.
Mitchell-Herzfeld, S., Shady, T., Mayo, J., Kim, D., Marsh, K., Dorabawila, V., et al. (2008). Effects of Multisystemic Therapy (MST) on recidivism among juvenile delinquents in New York state. Rensselaer, NY: New York State Office of Children and Family Services.
Noreus, B., & Foley, J. (2012). Annual Maine juvenile recidivism report. Augusta, ME: Maine Department of Corrections, Division of Juvenile Services. Retrieved from http://muskie.usm.maine.edu/justiceresearch/Publications/Juvenile/Discharge_Youth_Recidivism_Report_
OJJDP Model Programs Guide. Aftercare. Washington, DC: U.S. Department of Justice, Office of Justice Programs, Office of Juvenile Justice and Delinquency Prevention. Retrieved March 6, 2012, from http://www.ojjdp.gov/mpg/progTypesAftercare.aspx
Parenting with Love and Limits (2011). PLL Reentry Delivery System. Savannah, GA: PLL.
Pate, S. (2008). Annual Maine juvenile recidivism report. Augusta, ME: Maine Department of Corrections, Division of Juvenile Services. Retrieved from http://muskie.usm.maine.edu/justiceresearch/Publications/Juvenile/Annual_Juvenile_Recidivism_
Patterson, G. R., Forgatch, M. S., Yoerger, K. L., & Stoolmiller, M. (1998). Variables that initiate and maintain an early-onset trajectory for juvenile offending. Development and Psychopathology, 10, 531–547.
Patterson, G. R., Reid, J. B., & Dishion, T. J. (1992). Antisocial boys. Eugene, OR: Castalia.
Petersilia, J., & Turner, S. (1993). Intensive probation and parole. In M. Tonry (Ed.), Crime and justice: A review of the research (pp. 281–335). Chicago: University of Chicago Press.
Puzzanchera, C., & Adams, B. (2011). Juvenile arrests 2009. Washington, DC: U.S. Department of Justice, Office of Justice Programs, Office of Juvenile Justice and Delinquency Prevention. Retrieved from http://www.ojjdp.gov/pubs/236477.pdf
Quinn, W. H. (2004). Family solutions for youth at risk: Applications to juvenile delinquency, truancy, and behavior problems. New York: Brunner-Routledge.
Rogan, A. (2008). Maryland Department of Juvenile Services: Expansion of evidence-based practices. Baltimore, MD: Maryland Department of Juvenile Services
Rowe, C. L., Parker-Sloat, E., Schwartz, S., & Liddle, H. (2003). Family therapy for early adolescent substance abuse. In S. J. Stevens & A. R. Morral (Eds.), Adolescent substance abuse treatment in the United States: Exemplary models from a national evaluation study (pp. 105–132). New York: Haworth Press.
Sells, S. P. (1998). Treating the tough adolescent: A step-by-step, family-based guide. New York: Guilford Press.
Sells, S. P. (2002). Parenting your out-of-control teenager. New York: St. Martin’s Press.
Sells, S. P., Smith, T. E., & Sprenkle, D. H. (1995). Integrating qualitative and quantitative research methods: A research model. Family Process, 34, 199–218.
Sells, S. P., & Souder, E. (2007). PLL Center of Excellence: Transportability. Savannah, GA: Savannah Family Institute.
Sexton, T., & Turner, C. (2010). The effectiveness of Functional Family Therapy for youth with behavioral problems in a community practice setting. Journal of Family Psychology, 24(3), 339–348.
Smith, T. E., Sells, S. P., Rodman, J., & Reynolds, L. R. (2006). Reducing adolescent substance abuse and delinquency: Pilot research of a family oriented psycho-education curriculum. Journal of Child and Adolescent Substance Abuse, 15(4), 105–115.
Springer, D. W., & Orsbon, S. H. (2002). Families helping families: Implementing a multifamily therapy group with substance-abusing adolescents. Health and Social Work, 27(3), 204–208.
State of California Board of Corrections (2002). Repeat offender prevention program. Sacramento, CA: Author.
Stuart, E. A. (2010). Matching methods for causal inference: A review and a look forward. Statistical Science 25(1): 1-21.
Szapocznik, J. & Williams, R. A. (2000). Brief Strategic Family Therapy: Twenty-five years of interplay among theory, research and practice in adolescent behavior problems and drug abuse. Clinical Child & Family Psychology Review, 3(2), 117–134.
Trupin, E. J., Kerns, S. E. U., Walker, S. C., DeRobertis, M. T., & Stewart, D. G. (2011). Family Integrated Transitions: A promising program for juvenile offenders with co-occurring disorders. Journal of Child & Adolescent Substance Abuse, 20, 421–436.
Trupin, E., Turner, A. P., Stewart, D., & Wood, P. (2004). Community transitions of juvenile mentally ill offenders. Behavioral Sciences and the Law, 22, 559–610.
VanDenBerg, J., Bruns, E. J., & Burchard, J. (2008). History of the wraparound process. In E. J. Bruns & J. S. Walker (Eds.), The resource guide to wraparound. Portland, OR: National Wraparound Initiative, Research and Training Center for Family Support and Children’s Mental Health. Retrieved from http://www.nwi.pdx.edu/NWI-book/Chapters/VanDenBerg-1.3-(history-of-wraparound).pdf
Virginia Department of Corrections (2011). Report to the House Appropriations Committee, January 24, 2011. Richmond, VA: Author. Retrieved from http://hac.virginia.gov/committee/files/2011/01- 24-11/Reentry.pdf
VisionQuest (2012). Baltimore FFT. VisionQuest’s Functional Family Therapy Services (FFT). Downingtown, PA: Author. Retrieved from http://www.vq.com/quest/programs/md/md_baltimoreFFT.php
Wiebush, R., Wagner, D., McNulty, B., Wang, Y. & Le, T. (2005). Implementation and outcome evaluation of the Intensive Aftercare Program: Final report. Washington, DC: U.S. Department of Justice, Office of Justice Programs, Office of Juvenile Justice and Delinquency Prevention.
Winokur Early, K., Hand, G., Blankenship, J., & Chapman, S. (2012). Redirection continues to save money and reduce recidivism. Tallahassee, FL: Justice Research Center.
Winterfield, L., & Brumbaugh, S. (2005). Characteristics of prisoner reentry programs for juveniles: The multi-site evaluation of the serious and violent offender reentry initiative. Research Triangle Park, NC: RTI International; and Washington, DC: Urban Institute.