|Home | About | Journals | Submit | Contact Us | Français|
Fostering Healthy Futures (FHF) is a randomized controlled trial of an innovative preventive intervention for preadolescent youth (ages 9–11) placed in out-of-home care. The program is designed to promote child well-being by identifying and addressing mental health issues, preventing adolescent risk behaviors, and promoting competence. This paper describes the design, implementation, and uptake of the FHF program as well as our approaches to the challenges of conducting research-based prevention work within a child welfare setting.
Fostering Healthy Futures (FHF) is a randomized controlled trial of an innovative prevention program for preadolescent youth (ages 9–11) placed in out-of-home care. The program is designed to promote child well-being by identifying and addressing mental health issues, preventing adolescent risk behaviors, and promoting competence. The prevention program was designed within a solid theoretical framework that emanated from our pre-intervention research. FHF provides cognitive, academic, and mental health assessments for all eligible youth. Half of the youth are then randomly selected to participate in a nine-month prevention program, which is manualized and consists of mentoring and therapeutic skills groups.
Based on their review of the epidemiological literature, Landsverk and Garland (1998, p.194) report that “between one-half and two-thirds of the children entering foster care exhibit behavior or social competency problems warranting mental health services.” Studies of Medicaid claims suggest that as many as 57% of youth in foster care meet Medicaid criteria for a mental health disorder (dosReis, Zito, Safer, & Soeken, 2001). In addition, children with a history of maltreatment and/or foster care placement are at greater risk for delayed cognitive development and poor school performance (Courtney, Roderick, Smithgall, Gladden, & Nagaoka, 2004; Veltman & Browne, 2001). Despite high rates of problems, not all youth in foster care receive needed services (Burns et al., 2004). For instance, one study found that 80% of 6–12-year-old children in foster care had received a psychiatric diagnosis, but only 50% had received services (Zima, Bussing, Yang, & Belin, 2000). Our assessment protocol is designed to identify behavioral, mental health, cognitive and academic problems that have not yet been identified.
The theoretical model underlying FHF is informed by the resilience literature, which identifies those factors that are associated with adaptive functioning among children at high risk for adverse outcomes. The FHF theoretical model posits that maltreatment and placement in foster care interact with attachment, temperament, physiological, and neurological variables to result in psychological, social and behavioral functioning that often deviates from normative development. Impairment in these areas is hypothesized to lead to mental health problems, risk behaviors, low levels of competence, and poor quality of life (Milan & Pinderhughes, 2000; Pynoos, Steinberg, & Wraith, 1995; Schofield & Beek, 2005).
As depicted in the FHF intervention model (Figure 1), our prevention program is designed to promote adaptive functioning in order to foster resilience. The FHF intervention targets the psychological, social, and behavioral domains because they have been identified by the resilience literature as particularly relevant for high-risk youth and because they are potentially malleable. Furthermore, child development researchers have concluded that the most salient developmental tasks for preadolescent children are to “regulate their emotions; to modulate their behavior and adapt to the school environment; to form positive relationships with peers; and to achieve and maintain a positive sense of self” (Bolger & Patterson, 2003, p. 163). Improvement in these domains is hypothesized to produce better distal outcomes and fewer adverse life-course outcomes. The intervention model also recognizes that intervention effects may differ depending upon a number of hypothesized moderating variables. Although our model acknowledges that children and families are embedded in multiple systems (Brofenbrenner, 1979, 1995), it focuses on building individual strengths so that children will have the inner resources to cope effectively with what can be highly chaotic and uncontrollable life circumstances.
FHF was developed based on a number of pre-intervention efforts (NIH, 1998). First, we reviewed the literature to identify: (a) risk and protective factors for children placed in foster care, and (b) evidence-based interventions that could accommodate the unique challenges presented by this population. Our own clinical and research work with this population also informed the intervention design (cf. Taussig, 1992; Taussig, 2002; Taussig & Talmi, 2001).
After we identified mentoring and skills groups as possible intervention components, we conducted focus groups with: (a) biological parents whose children were placed in out-of-home care, (b) foster and kinship caregivers, (c) children in out-of-home placement who received mentoring, (d) mentors of youth in out-of-home care, and (e) social work students. Each participant group felt that mentoring and skills groups would be beneficial for youth in out-ofhome care. The groups also provided invaluable feedback on programmatic issues, such as skills group session content, timing and location of groups, and methods for engaging children and families.
Finally, we met extensively with community partners to discuss program implementation. Community partners included departments of human services in participating counties and schools of social work (whose students serve as mentors in the intervention). Our discussions with these partners helped us develop: systems for identifying eligible youth through child welfare records, methods for maintaining families’ confidentiality, means for effective communication between caseworkers and FHF clinical staff, and procedures and policies that would permit social work students to drive participating youth.
All of these pre-intervention efforts were designed to maximize the likelihood that the FHF intervention would be developmentally-sensitive, contextually-relevant, generalizable, and replicable in a range of settings (Hoagwood, Burns, Kiser, Ringeisen, & Schoenwald, 2001).
The FHF program recruits all 9–11-year-old children from participating counties who are court ordered into out-of-home care during the preceding 12 months as a result of maltreatment. Out-of-home care includes foster and kinship care, group homes, residential treatment centers, and psychiatric hospitals. It is important to note that FHF is a voluntary program and that children and families are not court-ordered to participate. The program is “above and beyond treatment as usual,” and is never meant to replace other services children and families may receive.
Once enrolled, youth receive a free cognitive, academic, and mental health screening assessment; half are then randomly selected to participate in the 9-month mentoring and skills training intervention. If youth change placements or reunify with their parents during the intervention, their participation continues with appropriate consent. Youth in both the “assessment only” and the “assessment plus intervention” groups are assessed pre-randomization, immediately post-intervention, and at a 6-month follow-up. Because solid prevention research requires ecologically valid measurement across relevant settings (Capaldi, Chamberlain, Fetrow, & Wilson, 1997), our assessment protocol measures variables of interest using a combination of youth-report, caregiver-report, teacher-report, and staff-report measures and also provides for data abstraction from child welfare, juvenile justice, and academic records.
The FHF program screens all children enrolled in the randomized controlled trial (RCT) for cognitive, educational, and mental health problems, using standardized tests of intellectual ability (Kaufman Brief Intelligence Test) and academic achievement (Wechsler Individual Achievement Test), as well as normed caregiver- and youth-report measures of psychological functioning (Children’s Depression Inventory, Revised Children’s Manifest Anxiety Scale, Trauma Symptom Checklist, Child Behavior Checklist). Scores generated from these assessment tools, as well as our behavioral observations, enable us to identify potential problems in cognitive, academic, and mental health domains. These findings and accompanying recommendations are summarized in a 3-5-page report provided to children’s caseworkers. Caseworkers are encouraged to use the reports to advocate for educational and mental health services, as appropriate.
A standardized screening assessment can be a valuable tool for identifying potential problems in a relatively efficient manner. During the first year of our program, 54% of the children scored substantially lower on tests of academic achievement than they did on standardized tests of intellectual functioning. According to their caregivers, however, only 8% of the underachieving children were receiving special education services. In addition, during our most recent recruitment period, 10% of the children assessed reported current suicidal ideation or intent. These findings were reported to children’s current caregivers and to their caseworkers and in many cases, neither was aware of the child’s current suicidal ideation or plans.
At the end of the assessment phase, approximately half of the children enrolled in the clinical trial are randomly selected to participate in the intensive intervention, which includes skills groups and mentoring.
FHF skills groups meet weekly for 30 weeks during the academic year and include eight youth and two group facilitators. Mentors transport youth to and from group, and attend separate group supervision while children participate in skills training. Following the group activities, mentors and youth have dinner together with program staff.
FHF groups are highly structured. Each weekly session follows the same predictable routine, during which children: (a) “check in,” and are given an opportunity to process current feelings and events; (b) review the previous week’s lesson; (c) are introduced to a new lesson with an “icebreaking” story or relaxation exercise; (d) are taught a new lesson, mostly through Socratic methods; and (e) participate in an activity, which is designed to help them practice new skills, and usually involves a role play, game, drawing, or small group activity.
The FHF skills group uses a manualized curriculum that combines traditional cognitivebehavioral skills group activities with process-oriented material. Units address the following topics:
|•||Emotion Recognition||•||Dealing with Worry||•||Drugs/Alcohol|
|•||Perspective Taking||•||Cultural Competence||•||Anatomy and Puberty|
|•||Problem Solving||•||Change and Loss||•||Dating Pressure|
|•||Communication||•||Healthy Relationships||•||Abuse Prevention|
|•||Anger Management||•||Peer Pressure||•||Future Orientation|
The FHF curriculum includes materials from evidence-based skills group programs, including PATHS (Promoting Alternative Thinking Strategies, Kusché & Greenberg, 1994) and Second Step (Committee for Children, 2001), which are incorporated in the following units: emotion recognition, problem solving, perspective-taking, communication, anger management, and dealing with peer pressure. Materials from multiple other sources, as well as project designed curricula, are seamlessly integrated into each lesson, as detailed in our skills group manual.
The FHF skills group curriculum is tailored to meet the needs of children placed in out-of-home care by incorporating units on coping with change and loss, dealing with worry, and establishing healthy relationships. In addition, all units include activities relevant to the foster care experience, and role plays are designed to mirror problems faced by children in out-of-home care. In the middle of the year, the children organize, plan, and host a “panel night,” during which adults who grew up in out-of-home care share their own experiences. Finally, the FHF curriculum supports children in creating lifebooks, which use verbal and nonverbal media to chronicle children’s pasts and document their hopes and plans for the future. The skills group curriculum includes lifebook activities that encourage youth to practice newly learned skills with their mentors and others in home, school, and community environments.
A large body of evidence suggests that skills training curricula are effective in reducing risk and promoting resilience with diverse populations, including maltreated youth (Berliner & Kolko, 2000; Deblinger, Stauffer, & Steer, 2001; Greenberg, Domitrovich, & Bumbarger, 2000). Skills group programs have been criticized, however, because they often fail to provide for reinforcement of skills in natural environments, and because they target youth with identified behavior problems who may learn negative behaviors from one another in the group setting (i.e. “deviance training”) (Capaldi, Dishion, Stoolmiller, & Yoerger, 2001; Dishion, McCord, & Poulin, 1999).
The FHF skills group program attempts to addresses these criticisms. It provides ample opportunities for children to practice the skills they learn from group in naturalistic settings (e.g. with their mentors), or in unstructured, yet highly supervised, group activities (e.g. in dinner after group). FHF skills groups are also designed to minimize deviance training. Groups are comprised of heterogeneous children, including highly prosocial youth and those with significant behavioral and emotional problems. All group activities are closely monitored by group leaders who have been trained to recognize deviant behavior and respond appropriately. Furthermore, children are supervised at all times during the program and do not associate with one another outside of program activities.
The mentoring component of the FHF program is a manualized intervention that provides one-on-one mentoring for each child. The FHF program also serves as a field placement site for social work students who mentor and advocate on behalf of two youth placed in foster care. Spending 16–24 hours per week on field placement with the FHF project, social work students attempt to reduce those risk factors and promote those protective factors identified by the resilience literature as malleable and particularly salient for high-risk youth.
The structured mentoring component of FHF supports mentors as they:
All of the intervention strategies employed by mentors are individually tailored for each child, based on the children’s presenting problems, strengths, and interests, as well as their family and placement issues. For example, mentors facilitate a career shadow activity for each of their children. In this activity, adults (who match the children on gender and race/ethnicity) in the children’s professions of interest, show the children their places of employment, talk to them about their careers, and encourage them to stay in school and make positive life choices.
Evaluations of mentoring programs suggest that mentoring can have a positive impact on risk and protective factors identified by the resilience research (Jekielek, Moore, Hair, & Scarupa, 2002; Keating, Tomishima, Foster, & Alessandri, 2002; Rhodes, Grossman, & Resch, 2000). A meta-analysis of 55 mentoring programs concluded that that the effects of mentoring programs were significantly enhanced when programs incorporated several “best practices” (Dubois, Holloway, Valentine, & Cooper, 2002). Programs that used mentors with prior experience in a helping role or profession, those that provided for ongoing training of mentors, and those that provided structured activities for mentors and participating youth had the most beneficial effect on youth identified as high risk.
The FHF mentoring program incorporates all of these best practices. It uses students enrolled in social work programs because they are training to work in a helping profession. The program provides four hours per week of supervision for mentors, including individual supervision, group supervision, and didactic training. Finally, the FHF program provides structure for mentors’ work with participating youth. Mentors use group and individual supervision to set goals for their weekly mentoring activities and to plan activities that will help accomplish those goals.
It is challenging to a conduct a scientifically-rigorous evaluation of a prevention program within a child welfare setting, but necessary if we are to identify the most efficacious interventions for one of our most vulnerable populations. Below, we delineate some of the research and clinical challenges we have confronted in designing and implementing FHF.
There are many challenges in recruiting and retaining participants for a RCT in a foster care population because youth can change placements and/or legal custody multiple times during the course of the study. In fact, we have over 20 consent, assent, and HIPAA forms to accommodate the fact that youth may change placements and custody at different stages in the research study. Despite these difficulties, we have been highly successful in recruiting and retaining participants. To date, we have recruited 93% of the eligible youth and have retained over 90% of those recruited at subsequent assessment timepoints. We attribute our recruitment success to the fact that we provide the following: (a) no-cost screenings assessments, (b) appropriate financial compensation for participation in research interviews, (c) flexibility in scheduling locations, days, and times for interviews, and (d) a chance for youth to participate in an intervention that was specifically designed to meet the needs of children and families.
We have also been successful in engaging and retaining participants for the preventive intervention. To date, almost all of the youth randomly selected to participate in the intervention (95%) have chosen to enroll and only 7% have withdrawn. This is striking, given that more than half of the children changed placements, and 25% reunified, during the 9-month intervention. The rate of participation in the intervention is also very high. During the first three years of the program, youth attended an average of 25 (of 30) skills group sessions and mentors met with youth an average of 26 (of 30) times. Of note is the fact that none of the youth who reunified during the intervention dropped out of the program. This is a tribute to the excellent alliances and trust our mentors establish with biological parents before children reunify.
Several aspects of the intervention design may be particularly compelling for youth and their caregivers. FHF mentors provide all transportation for the program, help children and families access resources, and serve as an ally during an often tumultuous time. In addition, although the intervention is manualized, it allows for flexible and diverse approaches that reorient toward a child’s ever-changing needs (e.g. as they move placements or reunify with biological families). Although the FHF program stresses advocacy on behalf of youth and families, it is important to note that the program does not weigh in on placement decisions or assume other responsibilities that are more traditionally considered the role of caseworkers. Instead, FHF works to empower youth and families to advocate on their own behalf and use available community resources.
It is a challenge to design an intervention that is amenable to scientific evaluation but is also generalizable and replicable in diverse community settings. To this end, we have taken steps to maximize the generalizability of our findings by limiting our exclusion criteria. For example, we do not exclude youth with significant mental health and behavior problems, youth with mild developmental delays, or youth in restrictive placements (including psychiatric hospitals and RTCs). The fact that there is not differential attrition by treatment group also suggests that our findings may be generalizable. We believe our program and research methods, coupled with our high recruitment and retention rates, will maximize the generalizability of our findings and enhance our ability to successfully disseminate the program, should the quantitative results demonstrate efficacy.
In order to disseminate the program, however, we must also demonstrate that we can deliver the content and process of the intervention in the same manner to all participants (Dumas, Lynch, Laughlin, Smith, & Prinz, 2001). To this end, we have developed and revised program manuals to guide the intervention. We provide extensive orientation and ongoing supervision for the students who serve as mentors and research interviewers on the project. We document the duration and content of each mentoring and skills group activity and rate participants’ engagement. For example, the skills group program is comprised of 108 discrete activities delivered over 30 sessions. Group leaders track how many activities are completed in each group each week. During the first 3 years, an average of 96% of the activities were completed. Measuring program process will also enable us to examine which aspects of our program have the greatest impact.
Children involved in the child welfare system are extremely heterogeneous with regard to race and ethnicity. To complicate matters, children in foster care can be placed in neighborhoods strikingly different from their own, with caregivers of different racial, ethnic, and socioeconomic backgrounds; even these differences are not static because they change when children change placements. In the FHF program, these differences are often mirrored in the matches we make between mentors and participating youth. It is not possible to pair all youth with mentors who match on ethnicity, as 80% of our mentors are Caucasian. Nor is it possible to pair all boys with mentors who match on gender, because the social work students who serve as mentors are predominantly female (81%).
For these reasons, we have taken care to design an intervention and assessment protocol that is culturally sensitive. FHF incorporates components of interventions (i.e. PATHS, mentoring) that have been used successfully with heterogeneous populations. All FHF group materials were designed to be linguistically and visually sensitive for mixed cultural backgrounds, using input from experts in these areas. The mentors receive extensive training in multicultural issues through both their graduate school training and our program’s weekly seminar. Because FHF mentors often do not match the children they mentor on gender, race/ethnicity, or socioeconomic status, they are supervised closely with regard to the impact that these differences may have on children, families, and communities. Mentors are encouraged to embrace differences and to explore what these differences mean to their children. Mentors also work to facilitate the involvement of children in culturally-meaningful activities and to learn about and share their own, their mentees’, and others’ cultures. Children, with the support of their mentors, have an opportunity to give a presentation to the group about aspects of their cultural heritage. As already described, children are also paired with “career shadows” who are matched with youth on gender and race/ethnicity.
Experience in the FHF pilot trial provides some evidence that FHF is a culturally-sensitive program. For example, FHF has successfully enrolled multicultural children and families including bilingual children and monolingual Spanish-speaking caregivers. Our recruitment and retention rates do not differ as a function of race or ethnicity. Preliminary evidence from our pilot study suggests that matching youth and mentors on demographic characteristics does not influence outcomes. In the first three years of our pilot study, those ethnic minority children who were not paired with ethnic minority mentors did not differ from those with an ethnic match on rates of group attendance, number of mentor visits, or perceived support from their FHF mentors. Similarly, boys who were paired with female mentors did not differ from boys who were paired with male mentors on these indices.
In a qualitative study conducted with 100% of the first cohort of participants one year after the program ended, we asked youth and caregivers to discuss “what it was like” to have a mentor who matched or did not match on demographic characteristics. None of the children or caregivers reported any concerns about ethnic/racial similarities or differences. Similarly, all of the male youth and their caregivers reported that a mismatch in gender was not a problem, although a few caregivers of boys who had female mentors said that they had been concerned initially. These same caregivers reported, however, that they subsequently valued that their child had developed a positive relationship with a female role model.
Finally, there have been concerns about the ethics of a mentoring component that ends after 9 months. When developing the program, we found no solid empirical evidence suggesting that longer mentoring relationships were more beneficial. Furthermore, we felt that a planned and positive ending could be therapeutic for children who have often experienced unpredictable and traumatic losses. Through focus groups, we learned from youth that some short-term relationships (e.g. with teachers, camp counselors, group home leaders, therapists) had been meaningful and beneficial even though time-limited. We also learned that unplanned endings with mentors could be devastating. We were concerned that if the FHF program allowed mentors to maintain contact with youth after program completion, that they would not be able to sustain the relationship without program support and supervision. We felt that an unplanned, negative ending to the relationship might be harmful.
Our clinical experience, as well as the qualitative interviews with past program participants, suggest that the way we end the program is empowering and positive for youth involved. Children graduate from the program and receive medals and diplomas, as well as highly personalized speeches delineating all they have accomplished in the program. Our qualitative interviews with past program participants suggest that although children and families were sad when the program ended, they felt it was the right length.
Although we have not yet demonstrated empirically that the FHF program is effective in addressing mental health, behavioral and academic issues among preadolescent youth placed in foster care, we believe we have a very promising model. We have successfully negotiated many challenges of conducting a RCT within a child welfare setting, in large part because of the collaboration and support of our community partners. The ultimate goal of this research is to develop more efficacious interventions, thereby reducing adverse life-course outcomes and promoting healthy futures, not only for youth in out-of-home care, but for all youth at risk.
This project was supported by grants from the National Institute of Mental Health (1 K01 MH01972 and 1 R21 MH067618, H. Taussig, PI) and the Kempe Foundation, Pioneer Fund, Daniels Fund and Children’s Hospital Research Institute, Denver, Colorado. We thank all of the Fostering Healthy Futures staff who were instrumental in the development and implementation of the program. We also thank the Denver Department of Human Services and the Adams County Social Services Department, for their partnership in launching this program. Finally, we wish to express our appreciation to the children, families and students who made this project possible.
Heather N. Taussig, The Kempe Center for the Prevention and Treatment of Child Abuse and Neglect, University of Colorado School of Medicine, Denver, Colorado, 303-864-5216.
Sara E. Culhane, The Kempe Center for the Prevention and Treatment of Child Abuse and Neglect, University of Colorado School of Medicine, Denver, Colorado, 303-864-5847.
Daniel Hettleman, Private Practice, Denver, Colorado, 303-912-6632.