Toward Understanding Homelessness: The 2007 National Symposium on Homelessness Research. Homeless Youth in the United States: Recent Research Findings and Intervention Approaches. Intervention Research


Much progress has been made in providing services to homeless youth and families since 1987 when the Stuart B. McKinney Homeless Assistance Act (Public Law 100-77) was signed into law. There now exist a vast array of shelters and other emergency services to address the diverse needs of homeless individuals and families, including homeless youth (Toro & Warren, 1999). Increased funding from the federal government as well as other sources has also led to the development of new interventions. Although many of these interventions are designed to help homeless youth become and remain housed, some include other components such as mental health services, alcohol and other drug treatment, or HIV/AIDS risk reduction.

Unfortunately, few of these new interventions have been formally evaluated, and when evaluations have been done, rigorous experimental or quasi-experimental designs have generally not been used. In fact, we are aware of only one rigorous evaluation of a program for homeless youth (Cauce et al., 1998). Below, we describe some of the interventions that have been evaluated, discuss the results of those evaluations, and suggest directions for future research on promising interventions, even if those interventions have yet to be tested among homeless youth.

Case management. Recognizing the multiple and diverse needs of homeless youth, Robertson and Toro (1999) advocated for a comprehensive and intensive case management approach that would address the unique needs of each homeless youth. Such an approach could be implemented in existing shelters and drop-in centers, and the relationship that developed between homeless youth and their case managers could become an important resource for the homeless youth and their families.

Intensive case management has been used successfully with homeless families and adults (Homan et al., 1993; James, Smith, & Mann, 1991; Toro et al., 1997). At least some research suggests that it might also be effective with homeless youth (Paradise et al., 2001). For example, Cauce et al. (1994) evaluated an intensive case management program for homeless youth in King County (Seattle), Washington. Youth were randomly assigned to either intensive or regular case management. Both groups experienced improved psychological well-being and a reduction in problem behaviors after the first three months of the intervention. However, youth who received intensive case management exhibited less aggression, fewer externalizing behaviors, and more satisfaction with their quality of life than youth who received “treatment as usual.”

Another promising service model is Urban Peak Denver, which provides overnight shelter as well as a variety of other services to homeless youth between the ages of 15 and 21 years. A case manager conducts a needs assessment and develops a case plan that includes educational and employment goals. Youthcan receive shelter for as long as they are moving forward on their case plans, and those who have been discharged are followed for six months. According to Urban Peak’s Client Database, which tracks the housing outcomes of youth who receive services, the percentage who experienced a positive housing outcome (e.g., moving into their own apartment, obtaining permanent supportive housing, or returning to their family of origin) ranged from a low of 48 percent in 2000 to a high of 65 percent in 2003 (Burt, Pearson, & Montgomery, 2005).

Family-focused interventions. Although many programs work primarily, if not exclusively, with the youth who are homeless, others have targeted the family. This makes sense given that youth often cite family conflict as the cause of their homelessness (Whitbeck et al., 2002; Robertson & Toro, 1999) and they often end up returning to their families of origin anyway (Toro & Janisse, 2004). Moreover, at least one study found that youth who return home to live with their parents experience more positive outcomes than other youth (Thompson, Pollio, & Bitner, 2000). Of course, this could simply reflect the fact that the youth who are able to reunify are the youth with the fewest problems.

In any event, there is relatively little evidence as to the effectiveness of family-focused interventions. One example that is sometimes cited is a 1998 study by Coco and Courtney. They described a family systems approach for preventing recidivism among runaway females. Unfortunately, their evaluation of the intervention was weak, being based on a single-case design with a simple pre and post assessment of family satisfaction to assess the impact of the intervention.

It should be recognized that there are cases in which a family-focused intervention would not be in a homeless youth’s best interest. The most obvious example is a youth who has been severely neglected or abused. Other examples would include youth who have irreconcilable differences with their families, youth who have lost contact with their families, and youth whose families are homeless or precariously housed. In these cases, efforts must be made to find alternatives such as placement in foster care or independent living. Unfortunately, placement options may be limited, and may not represent an improvement in living situation (e.g., Benedict et al., 1994; Rosenthal, et al., 1991).

Social skills training. In addition to their various service needs, many homeless youth lack what might be considered basic life skills, including meal preparation, household cleaning, time management, and budgeting (Aviles & Helfrich, 2004). Such skills are essential if they are to successfully transition out of homelessness and into successful adult functioning. Teare, Authier, and Peterson (1994) evaluated an intervention that used role-playing and a token economy to teach social skills to homeless youth receiving emergency shelter services. The assumption was that youth with social skills would experience fewer conflicts both during and after their shelter stay. The researchers reported that youth satisfaction with the program was generally high, that only 13 percent of the youth engaged in or expressed an intention to engage in self-destructive behavior, and that 69 percent did not exhibit any behaviors that were considered “out of control” (e.g., verbal or physical aggression). However, the researchers did not examine whether these outcomes reflected a change in behavior and their design did not include a comparison group that would have allowed them to assess the relationship between outcomes and participation.

School-based interventions. School contexts provide an opportunity to assess and address the needs of homeless youth. Although there is some evidence that school-based interventions can benefit school age children who are homeless with their families, we are not aware of any school-based interventions that target unaccompanied homeless youth. However, we believe that such interventions could easily be adapted for homeless youth, most of whom do attend school (even if not consistently). For example, the Empowerment Zone provided a mental health treatment package for low-income and homeless elementary school age children during summer school (Nabors, Proescher, & DeSilva, 2001). Trained teachers and mental health providers administered classroom and small group interventions and individual counseling, and parents were offered parenting classes. Results showed that parents reported a significant decrease in child behavioral problems following the intervention. Another study found favorable results for a classroom behavioral management system in which trained teacher assistants used bracelets to reinforce positive behavior among homeless elementary school age children (Nabors, Hines, & Monnier, 2002). Although these initial findings are promising, the programs need to be expanded and more thoroughly evaluated.

Other intervention research. A number of other studies have also examined the outcomes of homeless youth who received shelter services. Several of these studies have analyzed data from the Runaway and Homeless Youth Management Information System (RHYMIS). RHYMIS includes information about all of the runaway and homeless youth served by the Family and Youth Services Bureau’s (FYSB) Basic Center and Transitional Living and Street Outreach programs. RHYMIS includes demographic characteristics, services provided, and status at program exit (Family and Youth Services Bureau, 2006).

For example, Thompson et al. (2002) examined the outcomes of 261 runaway and homeless youth in four Midwestern states who received emergency shelter and crisis services, and compared their outcomes to the outcomes of 47 at-risk youth who received services from longer-term day treatment programs. Demographic information about the runaway and homeless youth was obtained from RHYMIS. Baseline data were collected from both groups at program intake. Follow-up data were collected six weeks post-discharge from the runaway or homeless youth and six weeks after intake from the comparison youth. Both the shelter youth and the day treatment youth experienced positive changes across six domains (runaway behavior, family relationships, school behavior, employment, sexual behavior, and self-esteem) and there were no significant group differences in the amount of change they experienced. Whether these improved outcomes persisted beyond the six-week observation period was not addressed.

View full report


"report.pdf" (pdf, 271.9Kb)

Note: Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®