Gang-involved youth are responsible for a disproportionate share of crime in communities. Successfully tackling gang involvement and associated criminal activities should result in a substantial reduction of crime — including violent crime — in affected neighborhoods. To that end, many programs and strategies have been developed to prevent gang involvement, reduce the criminal activities of gang members, and help individuals disengage from gangs.
Yet gang crime and violence remain pressing problems. It is clear that meaningful inroads on gang-related crime will require, among other initiatives, investments in rigorous research and evaluation studies to identify effective anti-gang programs and strategies that can be readily implemented.
In 2009, researchers began exploring whether Functional Family Therapy (FFT), an evidence-based program demonstrated to reduce delinquency and substance abuse in a general delinquency population, could be modified for a population at risk of gang involvement or currently involved in gangs. They developed and implemented the new program, Functional Family Therapy–Gangs (FFT‑G), and NIJ-funded researchers conducted a randomized controlled trial to determine whether the new FFT‑G program reduced gang involvement and whether positive outcomes could be sustained over time.
In 2018, the researchers concluded the study with encouraging results: FFT‑G significantly reduced delinquency for program participants and cost less than treatment as usual. Positive results were more pronounced for those youth at highest risk of gang involvement. In general, differences in official recidivism outcomes were large and clinically meaningful.
The findings came with certain caveats, however. The study was limited to a single location, and the data were insufficient to demonstrate the program’s impact on gang involvement as distinct from delinquent activity. In addition, there were issues associated with the contamination of the control group and remaining questions about the long-term effects of the gang-specific therapy program and why the program brought about observed outcomes. A new, second-phase study funded by NIJ aims to address some of these unanswered questions.
Researchers Terence Thornberry and Denise Gottfredson from the University of Maryland and their colleagues took note of the absence of anti-gang programs that had been rigorously evaluated and found to be effective. In 2009, with funding from the Office of Juvenile Justice and Delinquency Prevention and the Centers for Disease Control and Prevention, they began developing an evidence-based program to demonstrably prevent and reduce gang membership and the negative impacts of gangs.
The team chose an approach that would both address the weaknesses of and build on the knowledge gained from previous research and program implementation efforts. Their work involved a number of steps: identifying an existing evidence-based program not directly focused on gang involvement; seeking the insight of a group of experts; modifying and implementing the program to focus on gangs; and rigorously evaluating the program using scientific principles to compare the participant group with a control group of individuals who did not receive the program.
The team first searched for an appropriate existing program model that a rigorous evaluation had shown to be effective. They sought a program that satisfied the exacting evaluation standards of the Blueprints for Healthy Youth Development project. An advisory board of researchers from various backgrounds systematically reviewed Blueprints programs to narrow the candidate list. The team and advisory board members consulted with program developers about the programs on the candidate list.
Functional Family Therapy, first developed by James F. Alexander at the University of Utah, was considered the most likely of all Blueprints programs to effectively address two key populations: those at risk of gang involvement and those already involved in gangs. FFT was also desirable because it had undergone several rigorous evaluations, including randomized controlled trials, which provided strong support of its effectiveness. A meta-analysis of studies of FFT further substantiated its effectiveness.
Adapting FFT for use as an anti-gang program required many steps; ultimately, modifications were limited though still significant. From 2010 to 2011, the team and advisory board members collaborated with Alexander and other program developers to adapt the program for youth who are gang-involved or at risk of gang involvement. Given the previous success of the FFT program, the team kept the FFT clinical model and time in treatment (12 to 15 sessions over about three months) for the new FFT‑G program. However, they modified the training and guidance provided to therapists. The training manual was supplemented to cover gang-related risk factors and group processes associated with gangs.
In addition, they added gang-related examples to the program design to demonstrate how to apply the clinical approaches of FFT to FFT‑G. Finally, FFT-certified therapists received 12 additional hours of training on the new FFT‑G model and greater supervision and feedback from the national FFT office on their FFT‑G cases. Discussions during therapeutic sessions with youth and adults emphasized the impact of gang membership on youth’s behavior and development.
Together, carefully devised changes to the program model were expected to better address the risk factors associated with gang affiliation and challenges associated with engaging the families of gang-involved youth. The team designed FFT‑G to reduce participants’ involvement in delinquency and substance abuse by addressing negative peer relationships, normative beliefs about rules and laws, constructive use of time, unhealthy family functioning, and negative parent behavior and substance abuse.
Implementing and Evaluating FFT-G
The team chose to implement and evaluate the new program in the Philadelphia Juvenile and Family Court. This gave them access to the city’s chronically high levels of gang activity and to a sufficient number of youth who were eligible for FFT‑G services.
Formulating an approach that would enable a rigorous study in a real-world setting was a challenging task that required extensive discussion and negotiation with court staff, judges, and treatment staff. The team faced concerns about the impact of labeling youth “gang members,” resistance to the notion that a gang problem exists in Philadelphia, apprehension about the public safety implications of increasing reliance on community-based services such as FFT‑G, and challenges with recruiting a sufficient number of eligible youth.
Ultimately, the team addressed these concerns and undertook a rigorous study to determine whether FFT‑G can reduce delinquency and substance abuse among a population at risk of involvement or currently involved in gangs. Among other arrangements, the team secured a commitment from a judge that permitted them to undertake a randomized controlled trial of FFT‑G.
Eligible youth included 11- to 17-year-old males from across the city whose court cases were assigned to the judge participating in the study, who had been adjudicated delinquent and ordered by the judge to receive family services, and who had not been referred to FFT in the previous year. Due to challenges with recruiting youth for the program, eligibility criteria did not include a measure of gang involvement — although the neighborhoods from which most of the participants were drawn were known to have elevated gang activity. Youth and families amenable to study participation were randomly assigned to receive probation and FFT‑G or treatment as usual. Treatment as usual included probation and the Family Therapy Treatment Program (FTTP). FTTP — developed for a general, non-gang-involved population — has goals and a dosage that are similar to those in FFT and FFT‑G, but it has not been rigorously evaluated. While participating in their assigned program, study participants continued to be eligible for other social services from the city.
A total of 129 families participated in the study. Of those assigned to FFT-G, 80% received at least one FFT‑G session and just over half completed the program. Unfortunately, few of the families assigned to FTTP actually received any services from FTTP (only 11 of 63). It appears that FTTP staff had problems engaging with the assigned families, and as a result few participated in the program. Consequently, the judge reassigned some of the families from FTTP (about 20%) to FFT to improve the likelihood that they would receive family services.
The study used data from multiple sources to assess the extent to which FFT-G accomplished its goals. The study measured youth gang involvement and offending, beginning when the participants were enrolled in the study and concluding 18 months later. In addition, the researchers collected information on program fidelity and participant involvement in the assigned programs (either FFT‑G or FTTP) along with other community services that participants received following random assignment.
To measure gang involvement, the researchers interviewed youth at study enrollment and again six months later — the time at which involvement in FFT‑G or FTTP was scheduled to end and when the researchers expected to begin observing desirable changes. To measure offending, they used youth self-reports of delinquency, violence, and alcohol and substance abuse obtained during the two interviews, as well as Philadelphia juvenile and adult court records from initial court involvement until 18 months following study enrollment. Researchers also used parent and caregiver interviews to measure adolescent substance abuse.
Using data from the city’s Community Behavioral Health office, the team tracked the type and costs of additional community services that these youth received for the six months following study enrollment. Program administrative data were the source of information on fidelity and involvement in FFT‑G and FTTP. Together, these data provided a picture of which services the study participants actually received and how these services affected their behavior over time.
The study results indicated that FFT‑G reduced delinquency for program participants and cost less than the usual treatment, FTTP. Importantly, those at the highest risk of gang involvement had the best results.
Though there was little indication that FFT‑G produced desired results within the first six months that participants were in the program, favorable results emerged by 18 months. By that point, those assigned to FFT‑G were significantly less likely to have drug charges (11% vs. 22%) or to be adjudicated delinquent (23% vs. 38%) than those assigned to FTTP.
Results were even more robust for FFT-G when:
- The cases of those reassigned from FTTP to FFT were removed from analysis.
- Analysis was limited to those study participants at highest risk of gang involvement.
At 18 months following assignment to FFT-G (i.e., 12 months after program enrollment concluded), FFT‑G participants at high risk of gang involvement had better outcomes regarding arrests (prevalence and number) and charges for felonies, crimes against persons, and property crimes. They were also less likely to be adjudicated delinquent. In general, differences between the treatment and control groups in official recidivism measures were large and clinically meaningful.
The team found that receipt of other community services was common regardless of whether participants were assigned to FFT‑G or FTTP. However, there was variation in the types of concurrent services received: Those assigned to FTTP (whether or not they participated) were more likely to receive residential services. Those assigned to FFT‑G spent fewer days in residential placements (135 vs. 191 days) and were more likely to receive other community-based services. Because residential services were more expensive than FFT-G or FTTP, the cost per youth served within six months of assignment to FFT-G or FTTP, for all services received, was lower for FFT-G youth than for those assigned to the usual FTTP treatment ($10,197 vs. $12,368). Thus, the team concluded that FFT-G did a better job of replacing more expensive services than FTTP.
Though these results are encouraging, there are reasons to be cautious and to continue exploring questions around the effectiveness of FFT-G. This study tested the program in one location, where it faced some challenges. Perhaps most significantly, there was insufficient information to report on whether FFT‑G reduced gang membership. In addition, as discussed earlier, there were concerns that the study did not test the program exclusively on a gang-involved population as well as issues with the contamination of the treatment group and weak implementation of aspects of FTTP. Finally, there are questions associated with the outcomes of FFT-G, including what its long-term effects will be and why it led to the observed outcomes.
An NIJ-supported follow-up study, which began in 2019, is exploring some of the questions left unanswered from the first FFT-G study. Among other things, the team hypothesizes that FFT-G is responsible for reducing involvement in gangs. To that end, the new study will address the following questions:
- Does participation in FFT-G result in reduced gang involvement?
- Will reductions in offending — including violent offending, drug abuse, drug sales, and negative attitudes and behavior — be sustained up to five years following assignment to FFT-G?
- Which factors are responsible for the positive outcomes identified in the initial study?
The follow-up study will take place in Philadelphia, building on the relationships already developed with the Juvenile and Family Court, program participants, and others in the city. The researchers will interview the youth from the original study, who are now at least 18 years old, to gather information on their gang involvement and membership, if any; perceptions of the services they received; physical and behavioral health; educational attainment; employment; housing; and relationships with family, partners, and peers. They will also review juvenile and adult court records on arrests, offense types, case dispositions, and instances of incarceration occurring up to five years following assignment to FFT-G or FTTP. In addition, the researchers will convene focus groups of program stakeholders, including probation officers and therapists, to explore their perceptions of FFT-G and how best to work with gang-involved youth.
As with any longitudinal study, the researchers expect to face challenges in finding and reengaging study participants after a substantial period of time. However, they have already successfully overcome challenges to advance their goal of identifying an effective anti-gang program. For example, they have navigated multiple funding streams and partners to support program development and evaluation, and they have conducted a randomized study in a juvenile court setting.
The study has strong potential to contribute to our understanding of what works and why in the effort to reduce gang involvement and associated criminality. NIJ is pleased to fund this next stage in the process to help advance this important public safety issue.
For More Information
About the Author
Mary Poulin Carlton is a social science analyst in NIJ’s Office of Research, Evaluation, and Technology.
About This Article
This article discusses the following OJP grants:
[note 1] David C. Pyrooz, Jillian J. Turanovic, Scott H. Decker, and Jun Wu, “Taking Stock of the Relationship Between Gang Membership and Offending: A Meta-Analysis,” Criminal Justice and Behavior 43 no. 3 (September 2015): 365-397, doi:10.1177/0093854815605528; and Terence P. Thornberry et al., Gangs and Delinquency in Developmental Perspective (New York: Cambridge University Press, 2003).
[note 2] See the Blueprints for Healthy Youth Development webpage for more information.
[note 3] Dan Hartnett, Alan Carr, Elena Hamilton, and Gary O’Reilly, “The Effectiveness of Functional Family Therapy for Adolescent Behavioral and Substance Misuse Problems: A Meta-Analysis,” Family Process 56 no. 3 (September 2017): 607-619, doi:10.1111/famp.12256.
[note 4] More information is available on the Functional Family Therapy website.
[note 5] The study began with support from the Smith Richardson Foundation in 2013 and continued in 2015 with NIJ funding.
[note 6] More information on implementation is available in Denise C. Gottfredson et al., “Scaling-Up Evidence-Based Programs Using a Public Funding Stream: A Randomized Trial of Functional Family Therapy for Court-Involved Youth,” Prevention Science 19 no. 7 (October 2018): 939-953, doi:10.1007/s11121-018-0936-z; and in Denise C. Gottfredson et al., “Reducing Gang Violence: A Randomized Trial of Functional Family Therapy,” Final report to the National Institute of Justice, grant number 2014-R2-CX-0001, February 2018, NCJ 251754.
[note 7] More information on the measures is available in Terence P. Thornberry et al., “Reducing Crime Among Youth At Risk for Gang Involvement: A Randomized Trial,” Criminology and Public Policy 17 no. 4 (November 2018): 953-989, doi:10.1111/1745-9133.12395.
[note 9] For information on how researchers defined risk of gang involvement, see Terence P. Thornberry et al., “Reducing Crime Among Youth.”